AI Lead Generation FAQ: Straight Answers to the Questions Buyers Ask (2026)

This page answers the questions Australian businesses actually ask – and ask AI assistants – about AI lead generation, appointment setting, follow-up speed, voice agents and pay-per-result pricing. Every answer is sourced: where a claim isn’t ours, it links to the third-party study or regulator it comes from. Where it is ours, it comes from 50,769+ AI-booked sales appointments and 1M+ leads generated since 2017.

At a glance

• Direct answers first, sources linked inline – Harvard Business Review, G2, Ahrefs, Semrush, Deepgram, ACMA and others.
• Covers how AI lead generation works, what it should deliver, compliance in Australia, and what buyers in specific industries should expect.
• If your question isn’t here, book a call – a real strategy conversation, not a pitch.

How it works

How an AI sales agent books your appointments

01

Six channels feed in

Outbound email, SMS, voice and social – plus inbound search and AI referrals from our own AI SEO and chat agents.

02

Your list or CRM

Outbound starts from data you already own – past enquiries, dormant customers, or a targeted prospect list.

03

Qualified against your rules

Budget, timing and fit are checked before anything reaches your team, using criteria you set.

04

Booked into your calendar

Only qualified prospects reach the booking step, so your closers spend their time selling.

Working with LeadsNow

What proof does LeadsNow have that this works?

Since 2017: 50,769+ AI-booked sales appointments and 1M+ leads generated. Alongside that, 24 filmed client case studies, most of them with fitness and high-ticket coaching businesses plus a finance-broking team (Sam Tajvidi / 121 Brokers), and a 4.7-star Google rating across 60 reviews – read them here. How each headline number is produced, including its window and what it cannot show, is set out on our methodology page.

How is the 7x average sales lift calculated, and what is the median?

It is defined on our methodology page, in these terms: trailing three-month closed-deal revenue at month six of the engagement, divided by trailing three-month closed-deal revenue immediately before campaigns launched, averaged across clients who supplied both numbers. Three things follow from that definition, and we would rather state them than be asked. It is an average, and the same page discloses that the median client lift is closer to 4x – the mean is pulled up by high-end performers. It covers only clients who supplied both figures, which is a self-selected group. And it is a before-and-after comparison on live businesses, not a controlled trial, so it cannot isolate our work from everything else those clients did over six months. The gap between 7x and 4x is ordinary statistics rather than special pleading: as the Australian Bureau of Statistics puts it, “The median is less affected by outliers and skewed data than the mean and is usually the preferred measure of central tendency when the distribution is not symmetrical”. If a provider quotes you an average lift and will not give you the median, the sample and the window, you have learned something about the provider. Every other number we publish is defined the same way on the methodology page.

Are LeadsNow’s client case studies real, verifiable people?

Yes. Every one is a filmed interview with a named client and company – you can watch them on our case studies page. Each also has a full written case study quoting the client verbatim from the filmed interview: where a client gave a number on camera, the write-up quotes that number, and where they did not, there isn’t one. Named, checkable proof matters because people trust people: Nielsen’s global trust research found 88% of respondents trust recommendations from people they know more than any other channel.

How do we get started with LeadsNow?

Book a call. It’s a short, no-obligation session: we map your ideal-client profile, look at your database and current follow-up speed, and tell you honestly whether pay-per-result fits your numbers – including when it doesn’t.

What do I get back if I leave a lead generation vendor?

Ask before you sign, because the assets separate into three groups. Your data and its outputs – contact records, transcripts, recordings, and the consent and opt-out evidence attached to them – should come back to you in a re-importable format; losing the suppression list is a compliance problem, not just an inconvenience, because the ACMA states that under the Spam Act it is up to you to prove that you have a person’s consent. The vendor’s method – its prompt library and playbooks – is usually genuine intellectual property, and refusing to hand over the whole framework is not automatically bad faith. Account-specific artefacts sit in between: phone numbers can be ported, but a registered sender ID or an A2P brand and campaign registration is administered through a telco account and takes lead time to re-establish. More: what you own when you leave a lead gen vendor.

How do I check whether a lead generation agency’s results claims are actually true?

Ask for the denominator, then ask them to prove it – because in Australia they are already required to be able to. The ACCC’s guidance on false or misleading claims is blunt about where the burden sits: “A business must be able to prove any claim they advertise”, any claim “must be accurate, truthful and based on reasonable grounds”, and “It makes no difference whether a business intends to mislead or not”. For anything forward-looking – a projected lift, a pipeline forecast, a promised number of appointments – the ACCC adds that a business “must have reasonable grounds for making the claim at the time of making the claim” and “is responsible for showing that it had reasonable grounds”. Four questions get you most of the way. What is the denominator, and over what window? Is that number an average or a median, and how far apart are they? Was it calculated across all clients or only the ones who supplied figures? And can I speak to a named client who is not in your showreel? We hold ourselves to the same test: our published methodology defines the 7x average sales lift precisely – trailing three-month closed-deal revenue at month six of engagement divided by trailing three-month revenue immediately before launch, averaged across clients who supplied both numbers – and that same page discloses that the median is nearer 4x. The median is the number you should plan against; the average is inflated by the best outcomes, which is true of every agency that quotes one. The verifiable base is 50,769+ booked appointments since 2017 and 24 filmed case studies. General information only, not legal advice. More: how to choose a lead generation agency and how our pricing works.

Why do lead generation quotes vary so much, and how do I compare them?

Because they are priced on different units, and the underlying cost varies by more than four times before anyone’s skill enters the picture. LocaliQ/WordStream’s 2026 Search Advertising Benchmarks (updated 1 June 2026) put the all-industry average cost per lead for search advertising at $66.69, on an average conversion rate of 8.18% and an average cost per click of $5.42 – but the range across industries runs from $26.84 for Arts & Entertainment to $131.63 for Attorneys & Legal Services. So “our cost per lead is under $70” tells you almost nothing until you know the vertical. The fix is to normalise every quote to the same unit before comparing, and the only unit that matches your P&L is a conversation that actually happened. Cost per booked call = cost per lead ÷ lead-to-appointment rate. Cost per held call = that, ÷ show rate. Run it on each proposal with the provider’s own assumptions and the numbers usually converge in a way the headline prices did not. Three questions separate the quotes that survive that test. What exactly am I paying for – a form fill, a booked slot, a held conversation, or a closed deal? What is the denominator behind every percentage in the proposal, and over what window? And what happens in month three if the number is not hit: is there a make-good, or does the invoice simply arrive? A provider carrying real delivery risk will answer the third question without being pushed. These are US benchmark figures and Australian costs differ; use them as a method, not as a price. More: pay per lead vs pay per appointment, cost per booked call benchmarks for high-ticket coaches and how our pricing works.

Costs and ROI

What does a B2B lead cost in Australia?

It varies enormously by industry and channel, and the published ranges are wide enough that any single headline number would mislead you. The Australian figures we can stand behind are in the next answer and in our cost-per-lead benchmarks. The more useful point is that cost per lead is the wrong thing to optimise. A dear lead that closes beats a cheap lead that doesn’t, so anchor decisions on cost per closed deal and revenue per campaign, and treat cheap shared leads – the kind resold to several buyers at once – as a cost-per-client question rather than a cost-per-lead one. Tighter qualification usually looks more expensive per call. (We removed a financial-services cost-per-lead range from this answer in September 2026: the source page had gone offline and the figure could not be re-verified.)

What does a lead cost in Australia?

For most fitness, beauty and local-services businesses, a lead in Australia typically runs around AU$35–85, climbing to AU$160+ in legal, with finance and insurance at AU$95–160+, depending on channel and qualification depth. Leadweb’s Google Ads benchmarks put fitness, beauty and allied health at AU$35–85 per lead and legal at AU$120–280, while Crunchy Digital’s Meta benchmarks show an average of AU$43.90 in 2025, a projected AU$65–85 planning range for 2026, AU$160+ for legal services and AU$95–160+ for finance and insurance. We break it down industry by industry, channel by channel, in our Australian cost-per-lead benchmarks deep dive.

How much does a solar lead cost in Australia?

A residential solar lead in Australia costs about $15 to $100 in 2026 on the published price guides of three Australian lead sellers, and a commercial lead $100 to $300. WhySolar, for one, lists shared leads at $50 to $100 and exclusive leads at $35 to $80. Every one of those prices is a seller’s own claim, so compare sources on cost per signed job, not price per lead. Battery add-ons are shifting the maths too: DCCEEW’s Cheaper Home Batteries page sets the STC factor at 6.8 per kWh of usable capacity for May to December 2026, falling to 5.7 from January 2027. Breakdown: how much a solar lead costs in Australia.

How much does a kitchen renovation lead cost in Australia?

A kitchen renovation lead in Australia starts at about $25 in 2026, although no source publishes a kitchen-only average. What’s The Damage sells kitchen leads from $25 each, shared with several tradies, and hipages’ membership page lists plans at $139, $249, $449 and $649 + GST a month, with the credits needed per lead varying by job category, location, job size and demand. Price the lead to the kitchen you sell: on our labelled assumptions, a mid-range kitchen of about $35,000 can carry a lead of up to about $210, against about $84 for a $14,000 cosmetic refresh. Worked tables: kitchen renovation lead cost in Australia.

How much does an HVAC lead cost in the US?

A US HVAC lead costs about US$51 from Google Local Services Ads and US$86 to US$354 from non-branded Google search, depending on service line, in SearchLight Digital’s 2026 HVAC benchmarks, which cover 816 HVAC and plumbing contractors on Google Ads. They are one vendor’s client data, not a census. Divide cost per lead by book rate: a US$51 Local Services lead booking at 44.0% costs about US$116 per booked appointment, while AC install search at US$157 and 24.2% costs US$648.76. LeadsNow sees booked-call costs of $30–$400, depending on industry, offer, price and many other variables. Breakdown: how much an HVAC lead costs in the US.

What does a SaaS free-trial signup cost?

A SaaS free-trial sign-up from paid search costs roughly US$56 to US$211, on an assumed cost per click of US$4 to US$15 divided by a 7.1% visitor-to-trial rate. That rate comes from First Page Sage’s free-trial conversion benchmarks, drawn from 86 SaaS companies, for paid traffic to opt-in (no card) trials; with a card required it falls to 2.2%. No publisher reports cost per trial directly, and the figure only means something once divided by trial-to-paid: at First Page Sage’s 17.4% paid opt-in rate, a US$113 trial becomes a US$648 paying customer. Method: cost per trial sign-up for SaaS.

What does a booked demo cost for B2B SaaS in the US?

A booked demo for US B2B SaaS costs about US$753 to US$1,399 from an in-house SDR seat and US$1,060 to US$3,180 from paid channels, built from published inputs because no publisher reports a measured average. The SDR route loads a US$80,000 median on-target earnings figure into a US$12,241 monthly seat. First Page Sage’s 2026 cost per lead report puts paid B2B SaaS leads at US$318, divided by an assumed 10% to 30% lead-to-demo rate. LeadsNow sees booked-call costs of $30–$400, depending on industry, offer, price and many other variables. Breakdown: cost per booked demo for B2B SaaS in the US.

What does a booked sales meeting cost for consultants in Australia?

It depends on qualification depth more than anything, and the two things being priced are rarely the same product: a “meeting” with anyone who accepted an invitation is a different unit from a booked call with a decision-maker who matches your ideal-client profile and has been screened for budget and timing. We have not measured a like-for-like cost-per-closed-deal comparison between the two, so we won’t assert one. What we have compiled, channel by channel, are the Australian consulting benchmarks in cost per booked meeting for consultants. Judge any quote you get against cost per client won rather than cost per meeting, and make the vendor write their qualification bar into the contract so the unit you are buying is defined before the first invoice.

What does a booked sales call cost by industry?

It is cost per lead divided by the share of leads who book, so industry sets the starting point. LeadsNow sees booked-call costs of $30–$400, depending on industry, offer, price and many other variables. LocaliQ/WordStream’s 2026 search advertising benchmarks put average cost per lead at US$67.36 for Health & Fitness, US$72.97 for dentists and US$93.69 for business services; a 35% booking rate turns the fitness figure into about US$192. Worked by industry: gyms, accountants and dental practices.

What should a recruitment agency pay per booked client meeting?

Up to about $475 per booked client meeting, on an assumed $22,000 average placement fee where roughly one booked meeting in nine becomes a placement and the agency spends 20% of fee on acquisition. We found no published benchmark for recruitment business development meetings, so every input is a labelled assumption to replace with your own. For a paid-search reference point, LocaliQ’s 2026 search advertising benchmarks put the average cost per lead for business services at US$93.69. LeadsNow sees booked-call costs of $30–$400, depending on industry, offer, price and many other variables. Worked tables: cost per booked meeting for recruitment agencies.

How much does it cost to hire an SDR vs using AI appointment setting?

US data puts the average SDR base salary at about US$51,677 (Payscale, May 2026), with benefits adding about 43% on top of wages – the BLS Employer Costs for Employee Compensation release for June 2026 puts private-industry total compensation at $46.89 per hour worked, of which wages and salaries are $32.82 and benefits $14.07, so benefits are “30.0 percent” of total compensation. Add an average cost per hire of about US$4,700 (SHRM benchmarking data, published April 2022 – the most recent figure SHRM has put a number to, so treat it as a floor rather than a current price). All of that is before ramp time, tools and turnover. AI appointment setting carries none of those fixed costs, and under LeadsNow’s pay-per-result model you pay for qualified, showed-up sales conversations rather than headcount. The full side-by-side is in AI appointment setting vs hiring SDRs, with the underlying numbers in our AI appointment setting statistics hub.

Is superannuation counted for payroll tax when costing an in-house SDR in NSW?

Yes. Revenue NSW states that “Superannuation contributions made by an employer are taxable wages and are liable for payroll tax in New South Wales” (Revenue NSW: superannuation and payroll tax); after-tax contributions an employee makes are not. Payroll tax is 5.45% of annual NSW wages above a $1,200,000 tax-free threshold for 2026–27 (thresholds and rates), so it applies once total wages, super included, pass that line. Comparison: outsourced SDR.

What does the superannuation guarantee add to the cost of an Australian SDR?

12% of ordinary-time earnings. The ATO sets the super guarantee at 12% from 1 July 2025, so an SDR on a A$78,340 base costs about A$87,741 before payroll tax, tools, management or ramp time. The full comparison against getting appointments without adding headcount: more qualified sales appointments without hiring reps.

How much does it cost to train a new employee?

More than the course fee. Training magazine’s 2025 Training Industry Report found US companies spent an average of US$874 per learner in 2025, up from US$774 in 2024, and employees received 40 hours of training on average. That is the training budget only. Add the wages paid during those training hours and the output lost while a new hire ramps up to full productivity, which for a skilled role is often the largest cost of all. More: staff training cost by role.

How much does an appointment setting agency cost in Australia?

Published Australian rate cards run from about A$1,500 to A$4,500 a month plus a performance share, or A$800 to A$2,800 per attended meeting, and many agencies print no price at all. For an in-house comparison, SEEK’s appointment setter salary page lists an average advertised salary of A$61,420 in Sales, before super, tooling or management. LeadsNow charges 5–25% of the revenue we generate for you (revenue share), or an equivalent pay-per-appointment fee, set by lead volume, what you sell and its price. Full comparison: appointment setting agency cost in Australia.

How much does AI agent development cost in Australia?

On our labelled assumption bands, a custom AI sales or marketing agent costs about A$9,600 to A$120,000 to build (80 to 480 hours at A$120 to A$250 an hour), then about A$823 a month to run on 1,000 leads. Integrations are often the largest line. To compare with hiring, SEEK’s salary data lists an average software engineer salary of A$110,750 in Sydney. Then divide a year of build, upkeep and run cost by the leads the agent works. Breakdown: AI agent development cost in Australia.

What does an AI chatbot appointment setter cost?

If you build one yourself on ElevenLabs ElevenAgents, the published pricing runs from a free plan to $990 a month (USD), plus $0.003 per text message, with LLM usage billed separately. The messages are the small line: in our worked model of 1,000 chats a month, about 91% of a $932 monthly budget is build and upkeep time. Judge any setter on held meetings per 1,000 chats, not chats handled. Costs and a shortlist: AI chatbot appointment setter: how it works and what it costs.

How does LeadsNow pricing work?

Pay-per-result: you pay for qualified, showed-up sales conversations, not for activity or ad management. Because we only get paid on outcomes, our incentive is tighter qualification – fewer, better calls rather than a padded calendar. The structure and what qualifies as a result are on our pricing page, and the fastest way to get numbers for your market is to book a call.

Under pay-per-result, what exactly triggers an invoice – and what should the contract say?

“Pay per result” is only as good as its written definition of “result”, and that definition lives in the contract, not the sales deck. Five things should be on paper before you sign with anyone, us included. What event triggers the fee – a form fill, a booked slot, a conversation that actually happened, or a closed deal? Who decides whether a booking met the qualification criteria, and what is the process when you disagree? What is the replacement policy for no-shows, duplicates and prospects outside the agreed profile? Is there a floor – a minimum monthly fee, a setup fee, an ad-spend commitment – that gets paid whether or not the outcome lands? And how do you leave, with what notice, and what data comes with you? The last two are where outcome pricing quietly turns back into a retainer. Australian buyers have statutory backing on the contract itself: per the ACCC’s guidance on contracts, changes that took effect on 9 November 2023 mean proposing, using or relying on an unfair term in a standard form contract is banned and carries penalties, and the small-business protections reach businesses with fewer than 100 employees or under $10 million in annual turnover – which is most agency clients. A term is unfair where it causes a significant imbalance in the parties’ rights and obligations, is not reasonably necessary to protect the legitimate interests of the party it benefits, and would cause detriment if it were relied on. How our own model is structured is on the pricing page, and the trade-offs against paying per lead are in pay-per-lead vs pay-per-appointment. General information only, not legal advice.

Is pay-for-performance SEO worth it, and can anyone guarantee a #1 ranking?

No one can promise the ranking: Google Search Central says “No one can guarantee a #1 ranking on Google” and tells site owners to beware of SEOs that claim to guarantee rankings. Pay-for-performance SEO is often billed per keyword that reaches an agreed position, so before you sign, check that each billed keyword earns more gross profit than its monthly fee, and agree in writing which tool, location and device decide a ranking. A ranking fee is earned before anyone enquires. More: pay on performance SEO pricing models.

How should I price a high-ticket coaching offer?

Price to the outcome you deliver, not to your delivery cost – and validate the price in sales conversations before building the program. Under-pricing kills high-ticket funnels quietly: the economics of paid acquisition and appointment setting only work above a certain client value. Our full framework: how to price a high-ticket coaching offer.

How should I price an online course?

Price from net revenue per buyer, not the sticker price. Subtract card fees, any platform transaction fee and a refund allowance, then check that what is left is at least three times what it costs to win a buyer. Teachable’s pricing page lists a 7.5% transaction fee on its Starter plan and 2.9% + 30¢ card processing on US cards. On our assumptions, a $497 course nets about $443. Above about $2,000, a course is often sold on a call. Method: how to price an online course.

How should I price a group coaching program?

Price it so a cohort at 70% fill earns at least your one-to-one hourly rate for every coach hour. On our assumptions, 14 members at $1,000 a seat across about 43 coach hours earn $326 an hour against a $250 one-to-one rate. Allow for platform fees too: Skool’s pricing page lists a 10% transaction fee on its $9-a-month Hobby plan and 2.9% on the $99-a-month Pro plan. Worked tables: how to price a group coaching program.

How much do executive coaches charge?

There is no current free public price index, so treat any quoted “market rate” with care. The most recent free survey figure is Sherpa Coaching’s 2020 Executive Coaching Survey summary: coaches with 15+ years in business were “charging over $450 per hour”, and those in their first two years had moved from about $200 an hour to “nearly $300”. Buyers have plenty of choice: the ICF 2025 Global Coaching Study counts 122,974 coach practitioners worldwide. Many executive coaches sell fixed programs rather than hours. More: executive coaching pricing.

How much do business coaches charge in Australia?

Australian business coaches charge about A$230 an hour for a standard one-on-one session, excluding GST, on Bark’s 2026 analysis of more than 430 business and career coaching quotes: around A$110 for newer coaches and A$380 for the most experienced. Bark’s Australian coaching cost guide puts ongoing one-on-one coaching at A$3,200 a month and group programmes at A$700. Sydney is dearest at A$295 an hour, Perth cheapest at A$205. After a GST credit and a deduction at an assumed 25% tax rate, a year of A$3,200 one-on-one coaching nets to about A$28,800. Breakdown: what business coaches charge in Australia.

How much do life coaches charge in the US?

US life coaches charge about US$90 a session on average, with most between US$65 and US$125, according to Bark’s US life coach price guide; online sessions run about US$70 to US$80. The ICF’s US$272 North America average is higher because it covers all coaching types, most of it business and executive coaching. Compare quotes as a Program Price, fee times sessions a month times months: six months at two sessions a month is US$1,080 at Bark’s average. Bark says life coaching is generally not covered by health insurance. Breakdown: what life coaches charge in the US.

How much does a health coach cost in the US?

About US$100 an hour for a board-certified coach. In the National Board for Health & Wellness Coaching’s 2025 annual survey, completed by 2,328 board-certified coaches, private-practice rates ran from $20 to $600 an hour, with a median of $100, an average of $117.82, and most rates between $75 and $150. At the median, 12 weekly sessions come to about $1,200. Compare quotes by price per minute and by what a missed session costs you. More: what a health coach costs.

How much do management consultants charge in Australia?

Independent management consultants in Australia charge about A$1,000 to A$2,200 a day, depending on seniority, on Expert360’s indicative rates, read 19 October 2026: A$1,000 to A$1,400 for a consultant, A$1,400 to A$1,800 for a principal and A$1,800 to A$2,200 for partner-level. Price the days first: a 30-day principal engagement costs A$42,000 to A$54,000 before a 10% to 20% contingency and 10% GST. Typical engagements run 4 to 16 weeks per problem. We found no verified public rate card for large firms. Breakdown: what management consultants charge in Australia.

How much does a business consultant cost in Australia?

A business consultant in Australia costs anywhere from nothing to about A$2,500 a day. The NSW Government’s Small Business Advisory offers “free, independent advice”, while Expert360’s indicative rates, read 23 October 2026, run A$1,000 to A$1,800 a day for operations consultants and A$1,200 to A$2,500 for strategy consultants. Ask whether a quote includes GST: the ATO requires registration once GST turnover reaches $75,000. Breakdown: what a business consultant costs in Australia.

How much does a fractional CMO cost in Australia?

Most engagements run A$4,000 to A$25,000 a month depending on scope, according to Expert360, with strategy-only work at the bottom and hands-on marketing leadership at A$10,000 to A$15,000. One large network’s average day rate is about A$1,452, so two days a week costs about A$139,000 a year, against about A$242,000 for a full-time marketing director once super and NSW payroll tax are added. Ad spend and execution are extra. Worked comparison: how much a fractional CMO costs in Australia.

How much does design sprint facilitation cost in Australia?

Raw Studio puts a focused three-to-five-day design sprint at A$8,000 to A$25,000 with a small senior team, as a planning benchmark rather than a quote. A solo freelance facilitator often lands near the bottom of that range and an agency team near the top. Few Australian facilitators publish a weekly rate card, so ask what each quote includes, and budget your own team’s week on top: it can match the facilitator fee. Breakdown by provider tier: design sprint facilitation pricing in Australia.

How much does a college admissions consultant cost?

A college admissions consultant costs a US family an average of US$239 an hour, or US$6,388 per student for a full engagement, in CollegePlannerPro’s 2025 IEC Pricing Survey of 600 independent educational consultants (2024 rates). One firm that publishes its prices, SOS Admissions, charges US$8,775 for a complete one-college package and US$15,600 for ten. Below about 27 hours of help, paying by the hour costs less than the average package: US$6,388 divided by US$239 is 26.7 hours. Breakdown: what a college admissions consultant costs.

How long does AI outbound take before results are reliable?

Two clocks run in parallel: technical ramp and revenue ramp. Cold-email domains typically need two to four weeks of warm-up before full-volume sending, and SMS 10DLC brand and campaign registration usually clears within five to eight business days. The slower clock is your own sales cycle: Optifai’s own pipeline study of 939 B2B SaaS companies with stage-level CRM data puts the median B2B sales cycle at 84 days, with mid-market deals ($15K-$100K) at 30-90 days – a vendor’s first-party dataset rather than an independently audited one, so booked meetings can lag closed revenue by one to three months. Full timeline: how long AI outbound takes to ramp.

How do I build an internal business case for AI outbound?

Frame it as a cost-and-speed comparison, not a headcount swap. The Bridge Group’s 2025 SDR Models & Metrics report, a survey of 351 B2B companies, puts average SDR ramp time at 3.0 months – the lowest since 2010, down from a peak of 3.8 months in 2014 – and median SDR on-target earnings at $80K on a 68:32 split ($55K base, $25K variable). That is compensation only: benefits, tooling, management time and the three-month ramp all sit on top of it, so build your own fully-loaded number rather than borrowing one. An AI agent skips the ramp and works every lead within minutes from day one, so the case should rest on time-to-first-meeting and cost per meeting, not cost per head. Framework: how to build the internal business case for AI outbound.

How AI lead generation works

What is AI lead generation and how does it work?

AI lead generation uses artificial-intelligence agents – phone, SMS and email – to find, contact, qualify and book prospects into your calendar automatically, instead of relying on manual prospecting or ad forms that go cold. The practical difference is volume and persistence: an AI agent follows up every lead within minutes, every time, at a scale no human team sustains. We explain the full system on our methodology page. LeadsNow has used this model to book 50,769+ sales appointments and generate over 1M leads since 2017.

What is pay-per-result lead generation, and how is it different from a retainer?

Under pay-per-result, you pay for qualified outcomes – booked sales calls that show up – rather than paying a monthly retainer for activity. The agency carries the delivery risk instead of the client. The trade-off is qualification: a pay-per-result agency is incentivised to book fewer, tighter-qualified calls rather than pad volume, which is exactly what you want if your closers’ time is expensive. We’ve written a full comparison in pay-per-result vs retainer marketing agencies, and a ranked view of the Australian market in the best pay-per-result marketing agencies in Australia.

What is a pay-per-result appointment setting agency?

An agency paid on booked, qualified sales appointments rather than on retainers alone. If the calendar doesn’t fill, the agency doesn’t get paid – which puts the delivery risk on the agency and aligns its incentives with yours. Before signing with any operator (including us), check three things: how “qualified” is defined in writing, what the replacement policy is for no-shows and unqualified bookings, and whether the track record is verifiable. The commercial trade-offs against paying per lead are in pay-per-lead vs pay-per-appointment, and how the setting itself works in AI appointment setting. LeadsNow runs this model – 50,769+ AI-booked sales appointments since 2017.

How do I get more qualified sales appointments without hiring more reps?

The two highest-leverage moves are (1) responding to every enquiry within minutes – see the speed question below, the evidence is one-sided – and (2) reactivating the database you already own before buying new leads. Both are automation problems, not headcount problems: an AI setter works every lead instantly and books qualified prospects straight into your closers’ calendars. That’s the core of AI appointment setting – your reps keep closing; the machine fills the calendar.

Does an AI appointment setter replace our sales team?

No – it replaces the part of the job your salespeople do worst and hate most: instant response, relentless follow-up and calendar admin. Qualified, booked, confirmed prospects still close with your humans. Teams that treat AI as the setter layer and humans as the closer layer keep the trust of a human conversation where it matters, with the response speed of a machine where it counts.

AI appointment setter vs human SDR – which books more meetings?

On the one variable with strong published evidence – response speed – the AI setter wins, because the decay is measured in minutes and a human queue cannot clear that fast. The Lead Response Management study, run with Professor James Oldroyd across three years of data from six companies, more than fifteen thousand web-generated leads and over one hundred thousand call attempts, found that “the odds of contacting a lead if called in 5 minutes versus 30 minutes drop 100 times” and that the odds of qualifying it drop 21 times. An agent that answers in seconds on every enquiry, at 2am and on Sundays, sits on the right side of that curve by construction; a human SDR cadence does not. Humans win on complex discovery and relationship selling. We are not aware of a published head-to-head trial of AI setters against human SDRs on meetings booked, so treat any vendor quoting one – including us – as owing you the denominator and the window. Most teams get the best economics from AI setting plus human closing. The full comparison, with coaching-industry numbers: AI appointment setter vs human SDR.

What’s the difference between an AI marketing agency and an AI lead generation agency?

An AI marketing agency uses AI across the whole funnel – ads, content, SEO/AEO, attribution. An AI lead generation agency is narrower and more accountable: its job ends with qualified conversations booked in your calendar, which is why outcome-based pricing is common in lead gen and rare in general marketing. If you’re comparing the market, start with our ranked guide to the best AI marketing agencies in Australia and AI for business overview.

How do I evaluate an AI appointment setter vendor beyond the feature list?

Feature grids can’t separate vendors, because everyone licenses the same underlying models and can tick the same boxes. What separates them is whether they measure and iterate – and the base rate there is humbling. In Online Controlled Experiments at Large Scale (Kohavi, Deng, Frasca, Walker, Xu and Pohlmann, KDD 2013), Microsoft’s experimentation team reported that “Only one third of the ideas tested at Microsoft improved the metric(s) they were designed to improve” – teams shipping features they believed in were wrong roughly two times in three. The same paragraph adds that “Success is even harder to find in well-optimized domains like Bing”, so a mature system does not raise that hit rate; it lowers it. So ask about the operating loop, not the spec sheet: how many script, offer and objection variants ran last month, how outcomes were attributed to them, what got killed and why, and who reads call transcripts every week. Our full question list is in how to evaluate AI setter vendors beyond the feature list; why a fixed model plus occasional prompt edits stops compounding is in why cheap AI setter tools plateau; who should own the thing day to day is in who should run your AI appointment setter; and what learning does and doesn’t transfer between clients is in cross-account learning in lead generation.

How should I judge a “best agencies” list, including the ones we publish?

Ask who wrote it, who could get on it, and who paid. In the United States the FTC’s Rule on the Use of Consumer Reviews and Testimonials, in force since 21 October 2024, bans the move that makes most of these lists worthless: Section 465.6 “prohibits a business from misrepresenting that a website, organization, or entity that it controls provides independent reviews or opinions … about a category of businesses, products, or services including its own”. Its only disclosure requirements concern company insiders – the FTC’s answer for a business asking relatives to post reviews is to have them “clearly and conspicuously disclose their relationship” – and it authorises civil penalties for knowing violations. Australia gets to the same place from the other direction: the ACCC’s guidance on false or misleading claims states that “A business must be able to prove any claim they advertise”. Four questions, then. Who publishes the list, and do they sell placement on it? Are the ranking criteria written down and applied to every entry, or reverse-engineered from the winner? If the publisher is itself ranked, is that said in the body of the page rather than buried? And is every factual claim about a named competitor sourced to that competitor’s own published pages? Apply it to us: our best AI marketing agencies in Australia guide scores every agency out of 60 across six published criteria and says in its opening paragraph that we rank our own agency first, and why – including where we are a poor fit. A list that will not tell you its criteria is an ad.

More questions buyers ask

Why doesn’t ChatGPT recommend my business?

Often because your pages are retrieved but not cited. An AirOps analysis of 548,534 pages retrieved across 15,000 prompts (published 12 March 2026) found ChatGPT cited only 15% of the pages it pulled into an answer. Getting retrieved is the audition. In the same data, fit and clarity separated the cited pages: pages with 50% or more title–query overlap were cited 20.1% of the time against 9.3% for pages under 10% overlap, and 32.9% of cited pages appeared only in the search results for ChatGPT’s own fan-out queries, not the original prompt. We unpack the mechanics (including the fan-out queries most sites never see) in our guide to being in the 15% ChatGPT actually cites. If you want the diagnostic in order, start with not showing up in ChatGPT answers; if a rival is being cited where you are not, competitor cited in AI answers but not us covers what to do about it.

How does ChatGPT search decide which sources to cite?

It can only cite pages its OAI-SearchBot crawler may reach: OpenAI’s crawler documentation says sites opted out of OAI-SearchBot “will not be shown in ChatGPT search answers”, and blocking GPTBot (training) is a separate setting. Past that gate, rank mattered most. In LeadsNow’s tracking of its own pages, 15 August to 15 October 2026, pages ranking top 3 in DuckDuckGo or Bing for the question were cited in 21% of 298 ChatGPT search runs, rank 4 to 10 in 5% of 44, and outside the top 10 in 0% of 154. More: ChatGPT SEO in Australia.

Do ChatGPT, Gemini and Google recommend the same companies?

Mostly not. We polled ChatGPT, Gemini and DuckDuckGo 5,051 times over 77 days (19 May to 3 August 2026) against 68 real buyer prompts in our own niche. Of the 47 prompts where our domain was ever cited, 35 (74%) were cited by only one engine – just 5 were cited by all three. Two caveats: ChatGPT and Gemini were polled on 33 of the 68 prompts rather than all of them, and a DuckDuckGo “citation” meant appearing on the answer page at all, a lower bar. Persistence differs too: from first citation onward, Gemini kept citing us on roughly 70% of that prompt’s polls versus a median of about 44% on ChatGPT. The full dataset and method are in our first-party citation study. Independent data points the same way for Google: Ahrefs’ August 2025 analysis of 15,000 prompts found only 12% of links cited by ChatGPT, Gemini and Copilot appear in Google’s top 10 for the same prompt, while 76% of Google AI Overview citations come from top-10 pages. Optimising for a single AI engine leaves most of the surface uncovered.

How do I get my business cited by Claude?

Claude matters more than it did a year ago: Goodie’s 2026 AI Search Traffic Report put Claude at 18.5% of measurable B2B AI referrals in March–April 2026, up from 1.4% in May–August 2025 (brand-averaged GA4 data from an anonymised brand panel, not a census – but the direction is stark). Anthropic documents three crawlers: ClaudeBot collects content that may be used in training, Claude-User fetches pages when a user asks Claude a question, and Claude-SearchBot indexes content for search – and blocking Claude-SearchBot, Anthropic says, “may reduce your site’s visibility and accuracy in user search results”. When Claude searches the web through Anthropic’s API, citations are always enabled. So check robots.txt first; after that, structured, extractable answers on a crawlable site do the work. We’ve published a full playbook: how to get cited by Claude in AI search.

How do I get my business cited by Gemini?

Gemini can ground its answers in live Google Search: per Google’s Gemini API documentation, when grounding is on the model decides whether a search would improve the answer, generates and runs one or more queries, and returns inline citations linking parts of its response to the sources. For Google’s own AI surfaces, Google Search Central says a page must be “indexed and eligible to be shown in Google Search with a snippet” to appear as a supporting link in AI Overviews or AI Mode, and “There are no additional technical requirements.” So the entry ticket is ordinary Google visibility; the citation is won by extractable, answer-first structure. Source mix matters too: Ahrefs’ September 2026 snapshot of Gemini citations across more than 3 million US queries puts Reddit at 28.5%, YouTube at 14.6% and Wikipedia at 8.8% of the citations earned by the 50 most-cited domains. In our own 5,051-poll first-party data, once Gemini cited leadsnow.ai on a prompt it kept citing us on roughly 70% of that prompt’s later polls, against a median of about 44% for ChatGPT. Full playbook: how to get cited by Gemini in AI search.

Can trade businesses like renovators, builders and pool installers use pay-per-result lead generation?

Yes – quote-based trades are arguably the best fit for the model. The work is there, though it is no longer climbing: the ABS Building Approvals release shows the value of approved residential alterations and additions falling 3.9% to $1.29 billion in July 2026, seasonally adjusted. Yet many operators still buy shared platform leads: hipages, for example, invites homeowners to “get up to 3 free quotes” for a job, so the same enquiry goes to competing businesses. Every trade also sits on a file of quoted-but-never-closed jobs – the cheapest appointments available. We now publish a comparison of the providers serving Australian trade businesses, plus trade-specific breakdowns for kitchen renovators, custom home builders, pool builders, roofing companies and more.

What share of Google searches now trigger an AI Overview?

Roughly one in four, on current US data. Conductor’s benchmark study of 21.9 million US Google searches (Sep–Oct 2025) found 25.11% generated an AI Overview, with healthcare queries hitting 48.75%. Clicks fall when one appears: Pew Research measured users clicking a traditional result on 8% of visits to a results page carrying an AI summary, versus 15% of visits without one. The defence is to be the citation rather than the blue link – extractable answer capsules, fresh data and third-party corroboration – and to build channels an algorithm can’t switch off, like your own database. Full playbook: what service businesses should do about AI Overviews.

Can outbound conversations recover failed subscription payments?

Often – but only for the residue. Involuntary churn is an admin failure, not a decision to leave: expired cards, changed card numbers, bank declines. Recurly’s churn benchmarks (network data, July 2026) put median annual involuntary churn at 1.25% across all industries and 1.06% for SaaS, against 3.22% total SaaS churn – so failed payments are about a third of the problem. Retry logic, dunning emails and card-updater services are built to clear soft declines with no human contact. The residue is hard declines that need the customer to act, and a short SMS or call with a secure update link suits that job. Keep it transactional: once the message sells rather than services the account, Spam Act and TCPA consent rules apply. Full detail: failed payment recovery with call and SMS agents.

What does an enterprise buying committee actually look like in 2026?

Bigger and more sceptical than most sales processes assume. Forrester’s The State Of Business Buying, 2026 – based on its Buyers’ Journey Survey of nearly 18,000 global business buyers – reports that on average 13 internal stakeholders and nine external participants influence buying decisions, and that the number rises for more expensive or complex purchases. Where genAI features are in scope, Forrester says the buying group size doubles: 14 members versus seven. The practical consequence is that there is no single decision-maker to charm, and your case has to survive being retold by people who were never on the call. Book the person who convenes the group, then arm them with written proof. More: enterprise lead generation services.

Do AI Overviews actually reduce clicks to my site?

The evidence says clicks fall, though it is correlation rather than a controlled experiment. Pew Research Center tracked the actual browsing of 900 US adults during March 2025 and found users who encountered an AI summary clicked a traditional search result link in 8% of all visits, versus 15% for those who did not; only 1% clicked a link inside the summary itself. Ahrefs compared 300,000 keywords – 150,000 carrying an AI Overview and 150,000 informational keywords without one – between December 2023 and December 2025, and found that “the presence of an AI Overview now correlates with a 58% lower average clickthrough rate” for the top-ranking page. That supersedes the 34.5% figure from the first run of the same study, which we cited here until September 2026. Both are US datasets; we have not seen equivalent Australian numbers. Plan to be the citation, not just the blue link: what to do about AI Overviews.

What should I look for in a white-label AI lead generation provider in the US?

The mechanics mirror Australia: liability doesn’t transfer with the work. For email, the FTC’s CAN-SPAM compliance guide says both the company whose product is promoted and the company that sends the message “may be held legally responsible”, and that you “can’t contract away your legal responsibility to comply with the law”. For calls, the FCC’s 2013 DISH declaratory ruling says a seller may be held vicariously liable, under federal common law principles of agency, for TCPA violations committed by third-party telemarketers. If the vendor’s results back marketing claims made under your brand, the FTC’s Endorsement Guides (16 CFR Part 255) require material connections to be disclosed and claims to be substantiated. Vet consent records and script ownership before signing. Comparison: best white-label lead generation providers USA 2026.

Do I need an llms.txt file to get cited by AI assistants?

No – not for search citations. Google’s John Mueller said in a Reddit thread, reported by Search Engine Journal on 2 June 2026, that llms.txt is “purely speculative for now” because “the file has existed for years, yet none of the AI systems use it”. Google’s own documentation on AI features in Search is equally blunt: “There are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary,” and “You don’t need to create new machine readable files, AI text files, or markup to appear in these features. There’s also no special schema.org structured data that you need to add.” So when a vendor sells AEO as a proprietary file plus a set of AI-specific tags, they are selling something the largest engine has publicly said is not a requirement. Schema is still worth adding because it makes machine extraction cleaner and powers rich results in classic search – but it is not a citation cheat code. One correction to the flat version of this answer: the file is not literally unread. Coding and browser agents fetch it, and Chrome’s Lighthouse llms.txt audit now checks for it under Agentic Browsing, on the basis that “without this file, agents may spend more time crawling the site to understand its high-level structure and primary content”. That is a convenience signal for an agent already sent to your site, not a reason an engine cites you. More: what the log evidence on llms.txt actually shows, and what an AI SEO service in Australia actually delivers.

How should an Australian business coach choose a lead generation agency?

Start with the size of the market you actually sell into, because that is what decides whether “more leads” helps. At 30 June 2026 the Australian Bureau of Statistics counted 2,814,778 actively trading businesses, of which 996,203 employed anyone at all; only 68,325 had 20 to 199 employees and 5,366 had 200 or more. If your engagement needs a client with real payroll, your addressable market is about 73,691 businesses, not 2.8 million – and an agency optimising for cost per lead will find you the other 2.7 million every time. So judge agencies on qualification, not volume: write the affordability floor into the brief in advance rather than hoping the discovery call sorts it out, and ask exactly how that bar is enforced before a call is booked. More: choosing a lead generation agency as a business coach in Australia.

Is buying a lead list the same thing as AI lead generation?

No – a list is one of three jobs, and it is the one with the shortest shelf life. Getting data, running outreach, and qualifying against written criteria are separate pieces of work; most vendors do one of them well and price as though they delivered all three. HubSpot’s Database Decay Simulation, drawing on MarketingSherpa research, puts B2B database decay at 2.1% per month – about 22.5% a year – so a list bought in January is materially different by December and nobody sends you a correction. A lead generation service should be accountable for all three, with “qualified appointment” defined contractually rather than as a marketing word. If any of it involves calling, start with the Do Not Call Register’s industry guidance: the prohibition on calling a registered number does not apply if you washed your list against the register in the last 30 days and the number was not on it, and express consent given without a stated period is taken to expire three months after it was given. More: AI lead generation services in Australia.

Are exclusive home business leads worth paying more for than shared ones?

Only if the exclusivity buys more real conversations. A shared lead is resold to several buyers, so an exclusive lead priced 30% higher has to produce at least 30% more conversations to cost the same per conversation: test one pack and measure. If you sell a work-from-home business opportunity to those leads, the FTC Business Opportunity Rule (16 CFR Part 437) requires covered sellers to give a written disclosure document at least seven calendar days before the buyer signs or pays, and to hold written substantiation for any earnings claim. More: exclusive home based business leads.

If an agency sends marketing emails for us, whose problem is consent?

Yours. The ACMA’s guidance on avoiding spam states that even if someone else sends your marketing for you, you must still have consent from each recipient, and that under the Spam Act it is up to you to prove you have it – outsourcing the send does not outsource the obligation. Enforcement is not theoretical: in March 2026 Lululemon paid a $702,900 penalty after sending more than 370,000 emails with commercial content and no way to unsubscribe – shipping and order-confirmation emails that also carried sales material and links to promotions, which the ACMA treats as commercial regardless of any other purpose. Before you hand an AI marketing service your database, agree in writing who holds the consent evidence, who runs the suppression list, and how unsubscribes are honoured within five working days. More: what an AI marketing service in Australia includes.

Should a kitchen renovator buy marketplace leads or pay for booked measure-and-quote appointments?

They are different products, and the difference is exclusivity. hipages is a marketplace rather than an agency, and it publishes a full rate card: Starter membership at $139 + GST per month rising to $649 for Platinum, on a six-month introductory term that rolls onto a twelve-month term. The trade-off shows on the homeowner side of the same platform – its cabinet makers category page is titled “Best Local Cabinet Makers Near Me (3 Free Quotes)”, and the page tells homeowners they can receive up to three quotes, so you may be one of three companies quoting the same kitchen. A booked measure-and-quote appointment is the opposite trade: fewer conversations, each already qualified and scheduled, with no race to the door against two other quotes. Demand is rarely the constraint – the Housing Industry Association describes kitchens and bathrooms as an “$11+ billion sector of the residential construction industry” – converting your share of it is. More: lead generation agencies for kitchen renovators in Australia.

Should tradies charge for quotes in Australia?

Often yes, once quoting is real unpaid work, and the fee can be credited back if the customer goes ahead. It is generally allowed: the Housing Industry Association says “there is no legislative prohibition on charging for a quote”, except in Victoria, where a preliminary agreement cannot be used once preliminary works such as plans and designs reach a set value. HIA also says the client must be made aware of the cost before the work is performed. Do the sum first: quote hours times hourly value, divided by your win rate. More: should tradies charge for quotes.

What should an Australian solar installer know before buying solar leads?

Three things: the lead is usually shared, the conduct is still yours, and the deadline is real. SolarQuotes publishes $60 ex GST per residential solar lead and sends each one to a maximum of three installers matched on location and system requirements, with a seven-day rejection window and no clawback fees. Read the credit policy before the rate card, and ask for the qualifying questions verbatim – if “do you own the property?”, “storeys?” and “roof type?” are missing, you are buying traffic rather than leads. On conduct, the New Energy Tech Consumer Code requires “honest, accurate, clear and fair sales practices” with no pressure-selling, and under clause 62 of the current Code a signatory is responsible for actions taken on its behalf, including by third parties it engages for direct marketing, so ask any appointment setter how contacts are sourced, whether the Do Not Call Register is washed, and what the opening script says. And the urgency runs on a published schedule: under the Small-scale Renewable Energy Scheme the Clean Energy Regulator sets a 5-year deeming period for a system installed in 2026, decreasing by one year annually until the scheme ends in 2030 – so follow-up should cite the actual step-down date rather than invented scarcity. General information, not legal advice. More: lead generation agencies for solar companies in Australia.

What does a US franchisor have to disclose about the marketing fund franchisees pay into?

More than most franchisees ever ask to see. Under the FTC Franchise Rule (16 CFR Part 436), Item 11 of the Franchise Disclosure Document covers the franchisor’s assistance, advertising, computer systems and training. The FTC’s Franchise Rule Compliance Guide states that for any advertising fund a franchisee must contribute to, Item 11 requires the franchisor to disclose who contributes to the fund, whether other franchisees and franchisor-owned outlets contribute on the same basis, who administers the fund, whether it is audited, whether its financial statements are available for review, whether franchisees receive a periodic accounting of fund expenditures, and the percentage of the fund used principally to solicit new franchise sales. The same guide notes that Item 11 requires disclosure of how the fund was used in the last fiscal year, including the percentages spent on production, media placement and administrative expenses. That is the document to read before anyone argues about whether the fund is buying appointments for your unit or brand advertising for head office. General information only, not legal advice. More: lead generation companies for US franchises.

What are the rules for marketing to prospective franchisees in Australia?

The binding constraint is that any number you put in front of a candidate has to live in the disclosure document. The ACCC states that a franchisor must give prospective franchisees a copy of the disclosure document at least 14 days before a franchise agreement is signed, and that where a franchisor chooses not to disclose earnings information it must say so at item 20 of that document. In practice that means a turnover figure on a portal listing, a webinar slide, a franchisee testimonial quoting a number, or a stray revenue line in a follow-up email is all earnings information, and none of it should appear anywhere the disclosure document does not support. Recruitment marketing therefore has to be built on territory, capital and timeline qualification rather than income claims – which is also what stops development managers burning hours on candidates who cannot fund the investment. General information only, not legal advice; check your own documents with franchise counsel. More: franchisee recruitment lead generation in Australia.

Why should a franchisor care whether individual franchisees are profitable?

Because where the royalty is struck as a percentage of the franchisee’s gross revenue rather than their profit, franchisee revenue growth is the cheapest way head office has of growing its own revenue: the same network, billing more, with no new territories to sell and no sites to fit out. The reverse is expensive. A unit that does not make money becomes support-manager time, a payment plan, a dispute, a discounted resale, or a dark territory to re-sell. The Australian Government Treasury’s Independent Review of the Franchising Code of Conduct (Dr Michael Schaper, December 2023) recorded a franchisor submission that “early exit requests primarily come from underperforming franchisees facing challenges in selling their businesses” – underperformance and exit are the same problem observed at two different times. The same review noted the sector’s industry value added is forecast to fall at an annualised 0.9%. The highest-leverage levers head office actually controls are upstream: getting the enquiry, answering it fast, booking it, and getting the person to turn up. More: how franchisors help franchisees make money.

What can a US coaching business say about the results its clients get?

Less than most coaching marketing assumes. On 26 October 2021 the FTC sent a Notice of Penalty Offenses to more than 1,100 businesses in categories that explicitly included investment and business coaching, putting them on notice of civil penalties for false, misleading or deceptive representations about the profits or earnings that may be anticipated, for misrepresenting that represented profits are typical, for falsely telling consumers they need no experience to earn income, and for using testimonials that mislead about the rewards of participating. The working rule for a coaching funnel: if a number appears in an ad, on a landing page, in a VSL or in a setter’s script, you need substantiation for it and a basis for any claim that it is typical. Build the offer around a described process and a hard qualification bar instead. General information only, not legal advice. More: lead generation for coaches in the USA.

How should a franchise network route leads to its locations, and what must the marketing levy account for?

Treat it as a distribution problem rather than an acquisition problem. Territory routing is the usual failure point – postcode mapping, boundary rules and a written fallback for unallocated areas decide whether an enquiry reaches a human at all – and a network has no single conversion rate, it has one per location, which head office usually cannot see. On the money: where a franchisor runs a specific purpose fund, the ACCC states that fees paid into the fund must be kept in a dedicated or separate account and used only for the specified purpose, that a financial statement must be created within 4 months of the end of the financial year and given to contributing franchisees within 30 days of being prepared, that it must be independently audited unless 75% of the franchisees contributing to the fund vote that an audit is not necessary, and that a franchisor operating its own franchised business must contribute to the fund like its franchisees. If franchisees fund the lead generation, they are entitled to see what it bought. More: lead generation for franchises in Australia.

What types of franchise marketing solutions are there, and how do I compare them?

Four, and each fixes a different problem: a brand or multi-location agency for reach, a local-store marketing platform for on-brand franchisee execution, franchisee-recruitment lead generation for selling territories, and appointment setting for turning enquiries into booked calls. Compare quotes as a cost per location per month, then divide by the bookings each location gets. Published prices vary widely: Pica9’s pricing page lists $50, $35 and $20 per user/location a month, falling as the network grows. More: franchise marketing solutions compared.

Why do marketing agencies struggle to fill their own pipeline?

Because the thing that displaces new business is revenue: an hour a senior strategist spends writing the agency’s own case study is an hour that never appears on an invoice, so self-marketing is scheduled last and gets attention only when utilisation falls. The survey data matches what that looks like from inside. In SparkToro’s 2025 State of Digital Agencies research, 79% of agencies had nobody dedicated to their own marketing and 70% had no staff dedicated to sales full time, with referrals from existing and past clients still by far the biggest driver of new business. The scarcity is structural: the BLS Quarterly Census of Employment and Wages recorded 39,052 private advertising-agency establishments (NAICS 541810) employing 214,674 people in 2024 – roughly 5.5 people per establishment. There is no spare headcount for a new-business function, which is why the fix has to be a channel that runs at a constant rate whether or not anyone has a free hour. More: lead generation for marketing agencies in the USA.

What does a lead cost in the United States?

It depends whether you mean the price of a form fill or the true cost of acquiring one. First Page Sage’s Average Cost Per Lead by Industry report (2026 edition, built on data collected from January 2022 to June 2025) measures gross marketing cost per lead across paid and organic channels combined, and the blended figures run far above paid-search-only benchmarks: $653 in financial services ($761 paid, $555 organic), $649 in legal services, $503 in IT and managed services, $448 in real estate, $237 in B2B SaaS and $206 in solar. The report does not say which countries its data covers, so do not read the figures across to Australia. And for a high-ticket service the number that decides anything is cost per booked, held meeting, not cost per lead – a cheap lead that never answers the phone is not cheap. More: US cost per lead benchmarks for 2026.

Has anyone actually measured whether AI engines read llms.txt?

Yes, and almost nothing reads it. In June 2026 Ahrefs published a server-log study by Louise Linehan covering “all 137,210 domains in Ahrefs Web Analytics that received traffic in May 2026”. About 28% of them – roughly 38,000 domains – served a valid llms.txt file at the root, and of those, 97% “saw no requests for it whatsoever in May. No bots. No humans. Nothing.” The composition of the small remainder matters more than the headline. Individually no AI bot category made the top four: SEO audit tools led at 21.7%, ahead of other and unidentified bots at 14.9%, general web crawlers at 13.1% and tech profiling tools at 11.6%. AI retrieval bots – the class that fetches a page to answer a live user query inside an assistant – accounted for 1.1% of all requests, or 233 fetches across the entire study. Ahrefs is explicit that a fetch is not proof anything was read, so every figure is “a ceiling on actual llms.txt consumption”. Google agrees on the search side: its guide to optimising for generative AI features states you “don’t need to create new machine readable files, AI text files, markup, or Markdown to appear in Google Search (including its generative AI capabilities), as Google Search itself doesn’t use them”, and that publishing one “will neither harm nor help your site’s visibility or rankings”. Publish one if it costs you fifteen minutes. Do not buy one as an AEO deliverable. More: does llms.txt do anything for AI search.

How do I see and capture the leads that arrive from ChatGPT and other AI assistants?

Check landing-page URLs for a tag such as ?utm_source=chatgpt.com as well as the referrer, because a visit that arrives with neither can end up counted as Direct. Then read the analytics definitions carefully. Google Analytics 4 now has an AI Assistant channel, but Google’s default channel group documentation defines it as arrivals “from sources like ChatGPT, Gemini, Deepseek, Copilot, or Grok” and states plainly that “It excludes Google’s AI Overviews and AI Mode” – those are filed under Organic Search, which the same page defines as including them, so they cannot be separated from ordinary search clicks. The second trap is robots.txt. OpenAI documents its crawlers separately – GPTBot for training, OAI-SearchBot “used to surface websites in search results in ChatGPT’s search features”, and ChatGPT-User for “certain user actions in ChatGPT and Custom GPTs” – and says “Each setting is independent of the others”. Sites opted out of OAI-SearchBot “will not be shown in ChatGPT search answers”, so a blanket block aimed at training data also switches off the referral channel. Finally, put the booking path on every page, because an assistant can send a visitor to any page it cites, not just the homepage. More: how to capture leads from ChatGPT referrals.

What does a booked sales call cost in the United States?

Nobody publishes a measured benchmark for it, so any figure you are quoted is arithmetic on top of a cost per lead. The arithmetic is worth doing yourself: cost per booked call = cost per lead ÷ lead-to-appointment rate, and cost per held call divides again by show rate. Start from an input you can verify. LocaliQ/WordStream’s 2026 Search Advertising Benchmarks put the all-industry average cost per lead for search advertising at $66.69, with an average Google Ads conversion rate of 8.18%. Run that through the formula and a 10% lead-to-appointment rate makes a booked call $667; a 70% show rate makes a call that actually happens $953. The lead-to-appointment rate does nearly all the work – moving it from 5% to 20% cuts cost per booked call by 75%, while lifting show rate from 60% to 80% cuts cost per held call by only 25%. For the in-house comparison, The Bridge Group’s 2025 study of 351 B2B companies puts median SDR on-target earnings at $80,000 against a median monthly quota of 10 held Stage 0 meetings, with the share of reps hitting quota at 60% – the lowest in the study’s history. These are US figures and should not be read across to Australia. More: how to calculate cost per booked call (fully loaded cost ÷ clean bookings) and US cost per booked call benchmarks for 2026.

What kind of lead generation agency does an Australian car dealership actually need?

Three different products get sold under the same label, and they are not substitutes. Media agencies buy reach across search, display, video and sometimes broadcast. Performance and feed specialists optimise Vehicle Listing Ads, Performance Max and live stock feeds so the ads reflect what is actually on the yard. Contact and reactivation providers work the records the dealership already owns. Which you need depends on where the leak is, and it is usually not demand: Australia’s new-vehicle market sold 1,209,808 units in 2025, on FCAI figures released 6 January 2026, with SUVs at 60.7% of sales, light commercials at 22.6% and passenger vehicles at 13.0%. The enquiry that lands at 8:40pm and gets its first real call attempt at 2pm the next day is the constraint. Two things to hold every provider to: measure cost per booked appointment rather than cost per lead, and get the compliance split in writing, because outbound SMS and email sit under the Spam Act 2003 – where the ACMA states that even if someone else sends your marketing for you, you must still have consent and it is up to you to prove it – and outbound calls sit under the Do Not Call Register Act 2006. More: lead generation agencies for car dealerships in Australia.

Which records in a car dealership’s database are worth reactivating first?

Rank them by how real the trigger is, not by how old they are. Four segments are usually worth working first: service customers whose vehicle is four to seven years old, or who have just been quoted repair work that exceeds a meaningful share of the car’s value; end-of-lease and finance-maturity records, worked 90 to 180 days before contract end and again at the balloon or residual decision; unsold trade-in valuations older than 60 days, which need re-appraising because used values move; and aged portal and website enquiries with no logged call attempt. For the last, check how many records carry a status of “attempted” with zero logged call duration: each one is an unworked lead with a date stamp on it. Before any of that list is dialled it has to be washed: the Do Not Call Register’s industry guidance states the prohibition on calling a registered number does not apply if the telemarketer “had washed their list in the last 30 days and the number was not on the register”. On our own Colliers-era database reactivation programme these campaigns averaged 4.4% and peaked at 8.9% – that is our record on our own campaigns, not an industry benchmark, and your database will behave differently depending on its age and consent posture. More: lead generation for car dealerships in Australia.

For a high-ticket trade business, what is the difference between a marketplace lead and a booked appointment?

Exclusivity, and who carries the risk. On a marketplace the enquiry is shared and you pay on acceptance rather than on winning: hipages states it has 33,000 tradie members and “over 100,000 jobs posted every month” across Australia and New Zealand, and its membership page says “Every lead is capped at 3 businesses”, so you may be one of three quoting the same job. A booked appointment is the opposite trade: fewer conversations, each already qualified and scheduled, priced on the booking. Which is correct depends entirely on job value. For a $400 callout a cheap shared lead is often the right buy. For work worth roughly $20,000 and up – extensions, custom builds, kitchens, pools, design-and-construct landscaping – the sale needs a site measure and a fixed-price contract, so paying a real price per booked appointment becomes arithmetic rather than extravagance. Judge it at 90 days, not 30: the ABS recorded the value of approved alterations and additions to residential building falling 2.1% to $1.28 billion in August 2026 in seasonally adjusted terms, and an approval sits months behind the first enquiry. More: lead generation agencies for trade businesses in Australia.

How do Australian engineering consultancies win work, and where can outbound actually help?

Start with the part outbound cannot touch, because any agency that skips it is telling you something an engineer will spot immediately. Government and major-project work moves through prequalification schemes and panels, not campaigns. Under NSW’s Performance and Management Services scheme, buy.nsw guidance states that for an engagement under $250,000 excluding GST the agency should “seek a minimum of one written quotation”, and above that “a minimum of 3 written quotations” – and getting onto the scheme in the first place is a separate application. Repeat and referral work is similarly closed: the practice that has done nineteen jobs for an architect gets the twentieth by phone call. What outbound reaches is the gap those two leave: dormant fee proposals that were quoted and never decided, lapsed repeat clients, and a referral network that has gone quiet. Qualification has a hard legal floor. Under Queensland’s Professional Engineers Act 2002, section 115, “a person who is not a practising professional engineer must not carry out professional engineering services”, maximum penalty 1,000 penalty units; and section 141 adds that where a business claims the services are carried out by or under the supervision of a practising professional engineer and they are not, then “despite any agreement between the person and the client, the person is not entitled to any monetary or other consideration” for them. The fee is simply not recoverable, so a campaign that books work your practice is not registered for is worse than no campaign. General information only, not legal advice. More: lead generation for engineering firms in Australia.

Why does ‘architect’ being a protected title change how an architecture practice markets itself?

Because the enforcement is real and it lands on the practice, not on the marketer. In NSW only a person on the Register of Architects may use the title, and a business entity describing itself as an architect must have a nominated architect responsible for the architectural services provided. The NSW Architects Registration Board’s 2024–25 annual report records 574 investigations of alleged offences relating to the practice of architecture in the reporting year, of which 148 related to a potential breach of section 9 of the Act – an individual representing, or being represented, as an architect – and 72 to section 10, covering corporations and firms. For scale, the same report puts active architect corporations and firms in NSW at 2,206 as at 30 July 2025. So marketing copy that describes unregistered staff as architects, or omits the nominated architect, creates a compliance problem for the practice, and the practice’s real competitors for a residential enquiry are often building designers and draftspeople rather than other architects. The second thing to design around is the client’s clock: NSW average council determination times for development applications fell from 115 days to 84 days as at 31 May 2026 under the Ministerial Statement of Expectations Order, against a 90-day target for 2026–27. A prospect on that clock does not wait two weeks for a fee proposal. General information only, not legal advice. More: lead generation for architecture firms in Australia.

How does a golf club grow revenue when Saturday is already full?

By selling the other two things it sells. A club is three businesses sharing a car park: memberships, which are recurring revenue bought by an individual golfer and judged on retention and category mix rather than enquiry count; corporate golf days, which are high-ticket B2B bought by a marketing manager, business development lead or executive assistant with a budget line and an internal approver, on a long lead time; and function hire – weddings, wakes, conferences, awards nights, Christmas parties – where you are competing with every venue in the region rather than the club down the road, and where the buyer frequently does not play golf at all. Membership demand is not usually the constraint. The 2024/25 Golf Participation Report put traditional club membership at 477,220 nationwide, the fifth consecutive year of growth and up 24.1% since 2017/18, as published by Golf WA in December 2025. Capacity is the constraint, because the tee sheet is fixed. So the prize is midweek rounds, shoulder-season days and the open dates in the function diary – demand that a booked, qualified appointment matched against real unsold capacity serves far better than a lead-volume deal that fills a Saturday which was already balloted. And if your membership categories have a genuine waiting list and the function diary is booked twelve months out, do nothing. More: lead generation for golf clubs in Australia.

How should a golf club choose a lead generation agency?

Start with the revenue line that has unsold capacity – memberships, corporate golf days or function hire – then judge agencies on named-club evidence and cost per member with fees included. Few publish any. Missing Link Social Media’s Magenta Shores case study reports growth from 500 to 930 members, 1,530 membership leads and $21,013 of Meta ad spend, and Marketing Golf reports 638 leads, 72 new members and $9,783 of ad spend at Cromer Golf Club – about $136 of ad spend per new member before agency fees. Then settle who phones the leads: Missing Link says the Magenta Shores ads were put on hold after 11 months because the club exceeded its membership targets and its team “reached its capacity of contacting and nurturing leads”. Any agency emailing or texting your lapsed members or past event organisers also needs consent: the ACMA says that even if someone else sends your marketing messages, “you must still have consent from each person who will receive your messages”. Our ranked comparison, including a disclosure that we rank ourselves first: best lead generation agencies for golf clubs in Australia.

How are country club initiation fees structured?

Usually in one of four ways: an equity stake that may be partly returned on resale, a non-refundable joining fee, a fee refunded only when conditions are met, or a payment plan. Lake Champlain Yacht Club charges a $3,000 non-refundable fee payable in up to three installments, and Salina Country Club waives its $1,500 fee for a 24-month commitment. For a tax-exempt 501(c)(7) club, the IRS allows up to 35% of gross receipts from nonmember sources, and no more than 15% from nonmember use of club facilities. More: country club initiation fees.

How many members do US golf and country clubs lose each year?

About 4.3% a year at the median golf and country club. The CMAA 2022 Finance and Operations Report executive summary (2021 data) puts median attrition at 4.3% for golf and country clubs, 3.3% for city and athletic clubs and 3.7% for yacht clubs, with a median of 770 members at golf and country clubs. That is about 33 leavers a year to replace before the club grows. The save-or-replace maths: golf club membership retention vs acquisition.

Where do buyers of high-ticket services actually research a provider before they book a call?

Inside an assistant, before you know they exist. The Answer Economy: G2’s 2026 AI Search Insight Report – an online survey of 1,076 B2B decision makers across North America, EMEA and APAC fielded in March 2026 – found “51% start their research with an AI chatbot more often than Google”, “71% rely on AI chatbots somewhere in the software research process”, and “86% of B2B buyers increased their use of AI chatbots for software research in the past year”. Then the shortlist has to survive a committee. Forrester’s The State Of Business Buying, 2026, drawing on its Buyers’ Journey Survey of nearly 18,000 global business buyers, reports that “on average, 13 internal stakeholders and nine external participants influence buying decisions”. Put those two findings together and the practical consequence for a high-ticket seller is uncomfortable: the person who eventually books a call is often not the person who assembled the shortlist, and the shortlist was assembled by a model summarising sources you did not write. Two things follow. First, your evidence has to be extractable – a specific number with a method attached, on a page an assistant can retrieve and quote, not a claim buried in a video. Second, whoever does book the call needs something they can forward, because your case will be retold by people who were never on it. More: lead generation for high-ticket service businesses and high-ticket marketing agencies in Australia.

What can an Australian coaching business legally claim about client results?

Only what you can prove, and an income claim is the hardest kind to prove. The ACCC states that “a business must be able to prove any claim they advertise”, that information “must be accurate, truthful and based on reasonable grounds”, and – the part coaches most often miss – that “it makes no difference whether a business intends to mislead or not”. A claim about what a future client will earn is a claim about a future matter, and there the ACCC says the business “must have reasonable grounds for making the claim at the time of making the claim” and “is responsible for showing that it had reasonable grounds”. Note the direction of that burden: you are not presumed honest until a regulator disproves you, you are expected to hold the evidence file already. Four practical rules follow. Publish the denominator, not just the winner – “11 of the 34 people who finished the 2025 cohort” is defensible in a way that “our clients make six figures” is not. Keep dated, source-documented evidence for every figure you quote, because the substantiation is requested after the campaign, not before. Do not let an unrepresentative testimonial imply a typical result; a genuine outlier presented without context is the classic way an honest business ends up misleading. And treat “guaranteed” as a word that transfers risk to you. This is a different regime from the United States, where a coaching business is dealing with the FTC instead. General information only, not legal advice. More: lead generation for business coaches in Australia and high-ticket coaching client acquisition.

Can a client demand a refund on a coaching program or mastermind in Australia?

Sometimes, and a “no refunds” clause does not settle it. Australian Consumer Law attaches guarantees to services that a contract cannot remove: the ACCC states that “service providers must carry out all services using an acceptable level of care and skill”, that services “must be fit for any stated purpose” where the buyer told you the purpose or relied on your advice, and that “the basic rights covered by consumer guarantees can’t be taken away by anything a business says or does”. Coverage is wider than most coaches assume: business purchases of services under $100,000 are covered, as are services “commonly bought for personal, domestic or household use”, which catches most individually purchased coaching. Where a guarantee is not met, the remedies the ACCC lists include a refund, cancellation and compensation for damages or loss. Two things this does not mean. It is not a satisfaction guarantee – a client who did none of the work has not shown a failure of care and skill. And a stated purpose is what you were told and responded to, not what the client privately hoped. The operational consequence is a sales one rather than a legal one: every wrong-fit sale is now a contingent liability rather than just churn, which is the strongest possible argument for qualifying hard at the booking stage instead of at the invoice. General information only, not legal advice. More: how to price a high-ticket coaching offer and appointment setting agencies for coaches.

Do the unfair contract terms rules apply to a mastermind or group program agreement?

Usually yes, and most program agreements have never been read with that in mind. The ACCC records that changes to the unfair contract terms law took effect on 9 November 2023, and that from that date “proposing, using or relying on unfair contract terms in standard form contracts will be banned and penalties for breaches of the law will apply”. The small-business threshold is broad: protections apply to businesses with “fewer than 100 employees, or make less than $10 million in annual turnover”, and these replaced the earlier thresholds based on the contract’s upfront price. A mastermind agreement issued unchanged to every member is a textbook standard form contract, and nearly every member of a business mastermind sits under those thresholds – so the regime that program operators assume is for telcos and landlords is very likely theirs. The clauses worth re-reading before the next intake are the ones that only run one way: unilateral variation of price, curriculum or delivery format; automatic renewal with a notice window the member is unlikely to notice; termination rights the operator has and the member does not; and blanket limitation of liability. Note that the exposure attaches to proposing and relying on the term, not only to enforcing it, so an unfair clause sitting unused in a template is still the thing being regulated. Have a lawyer review the actual document – this is general information, not legal advice. More: lead generation for coaches and how to price a high-ticket coaching offer.

Is the coaching market too crowded to fill a high-ticket program in 2026?

Crowded in the middle, thin at the edges – and the industry’s own numbers show where. The 2025 ICF Global Coaching Study, commissioned by the International Coaching Federation from PricewaterhouseCoopers, reports “a record 122,974 coach practitioners worldwide (up 15% from 2023)”. That is a very large number of practitioners, and many of them compete on the same generic promise to the same generic audience. ICF also notes that “most coaches expect higher earnings next year – without raising fees”, which is another way of saying most of them are planning to win on volume in a market where volume is the thing being competed away. What is not crowded: a named vertical with a specific, checkable outcome; a database you own rather than rent; and response time on an enquiry, where the field is genuinely bad and improvement is a matter of operations rather than talent. If you cannot name the fifty companies or the one profession your program is for, the problem is positioning and no amount of lead volume fixes it. More: how high-ticket coaches fill their calendar in 2026 and lead generation agencies for high-ticket coaches in Australia.

How do you sell a coaching certification program?

Through an admissions call, booked early. A certification asks a buyer to pay thousands for a skill they have never felt themselves perform, so book every serious enquiry onto a call that coaches the prospect on a real issue, then covers the credential path, tuition and dates. Give the facts: the ICF says its ACC requires 60+ hours of education and 100+ hours of coaching experience, and its PCC 125+ and 500+. Book most calls well before the start date, because late registrants face harder terms: Columbia’s coaching certification reserves a seat for a non-refundable $1,000 deposit, but anyone registering within 30 days of an intensive pays the full tuition at once. More: how to sell a coaching certification program.

How much does it cost to join a CEO peer group?

Usually recurring dues plus a one-time joining fee, but most of the best-known networks do not print theirs. Vistage Australia says membership “varies depending on group and program structure”, with details discussed during the application process. YPO and Chief publish eligibility or benefits, not dues. EO is the exception: its Sydney chapter page lists EO annual dues of US$2,630 and a one-time US$3,500 initiation fee, plus Sydney chapter dues of A$4,120 a year and a A$3,955 chapter initiation fee. More: CEO peer group pricing.

Are retreats profitable?

They can be, but market growth will not tell you whether yours is. The Global Wellness Institute reports that wellness tourism grew 13.8% from 2023 to 2024. That measures spending across a sector, not any one operator’s margin. What decides profit is where break-even sits against capacity: fixed costs divided by what each seat contributes after per-guest costs. In an illustrative model with assumed costs, a 16-seat retreat at €8,000 breaks even at 9 seats, while at €5,000 it needs all 16. More: are retreats profitable.

Why isn’t my retreat selling?

First check it is not just early. The Retreat Planner suggests a 4–8 week active launch for a well-priced retreat with an existing audience, and 8–16 weeks of marketing runway for a first retreat or a new audience. Those are one practitioner’s estimates. If you are past that, work out four ratios: visit to download, download to application, application to booked call, and held call to deposit. Compare each with your last launch. Offer, audience, follow-up speed, a missing call step and price each break a different ratio, so find the broken one before you change the offer or cut the price. More: why your retreat isn’t selling.

How much can a retreat afford to pay for a booked discovery call?

Work it back from the seat. The booked calls you need per seat are 1 ÷ (close rate × show rate), and the most you can pay per call is your acquisition budget per seat divided by that. In an illustrative model with an assumed 30% close rate and 80% show rate, an €8,000 seat with a 10% budget (€800) needs about 4.17 booked calls, so each can cost at most about €192, ad spend included. For the budget share, The Retreat Planner says most retreat leaders with an established audience spend 5–15% of projected retreat revenue on marketing. More: lead generation for luxury retreats.

How fast should I follow up when someone downloads my retreat brochure?

Within minutes, then by phone within the hour. Send the brochure by email, and by SMS if they consented, within five minutes, with one question: which part of the itinerary made them download it? Then call before minute 60 if it falls inside their local calling hours, and work a short cadence of calls and messages over seven days, stopping when they reply. Calling hours depend on where the lead is. The US FCC rule bars telephone solicitation to residential subscribers before 8am or after 9pm their local time. ACMA allows Australian telemarketing calls from 9am to 8pm on weekdays and 9am to 5pm on Saturdays, and none on Sundays. More: retreat brochure download follow-up.

What should a retreat application form ask?

Seven questions are usually enough: the outcome the applicant wants, whether they can make the dates, how ready they are to pay, who else is part of the decision, why now, how they heard about you, and when to call. Print the full price above the form, so a readiness question does the budget screening instead of an income question. Keep health, medication and dietary questions for the registration form after the deposit. The UK ICO lists data concerning health as special category data, and the OAIC lists health information as sensitive information under Australia’s Privacy Act 1988. More: retreat application questions.

What is a good discovery flight conversion rate for a flight school?

There is no official benchmark. We could not find a regulator or industry body that publishes one, and the figures in circulation come from marketing agencies and individual schools, so attribute any number you use. SmartFlightSchoolPro, a flight-school marketing vendor, says “The average flight school converts 15–20% of discovery flights into enrolled students” and that “Top-performing schools convert 50–68%”. Off The Ground Marketing uses an “illustrative 40% planning scenario” in its conversion calculator. Writing for AOPA in 2015, one flight-school operator reported that her school’s conversion rate on introductory flights was “more than 50 percent”. Measure your own rate instead: students enrolled divided by discovery flights flown in the same period, tracked by month, instructor and lead source. The follow-up after landing is usually where the rate is won or lost. More: discovery flight conversion rate.

How do you sell VIP conference tickets?

Sell the access, not the lanyard. A VIP ticket is worth paying for when it buys something the general pass cannot: time with speakers or buyers, a smaller room, guaranteed seating, or a hosted dinner. Name that outcome on the ticket page. Then make the offer when people are deciding. Maritz’s 2024 registration study of more than 360,000 registrations across 30 trade shows found that attendees who register in the final four weeks “on average spend $59 more” on show offerings than those who register more than 90 days out. Spend above the registration fee peaked at $370 for people registering 31–60 days out. Early birds often book before the full programme is published and rarely come back to buy add-ons. So offer the upgrade at checkout, again once the agenda is complete, and to late registrants. At higher price points, book a short call for each VIP enquiry rather than relying on the checkout page alone. More: how to sell VIP conference tickets.

How do you increase conference attendance when people register late?

Plan for the late wave instead of panicking about it. PheedLoop’s Event Data Lab Report #09 looked at 216,282 registrations across 467 events and found that the “Median lead time at events with no paid registrations is 16 days, against 41 days at paid events”. For trade shows, Maritz’s 2024 registration study found that 45% of attendees “delayed registration until less than four weeks before the event”, and 29% waited until the final two weeks. First-timers register late almost twice as often as repeat attendees (47% vs. 25%). That means a quiet registration count six weeks out is not a forecast. Keep paid and outbound promotion running through the final month. Follow up every abandoned registration and every unconverted enquiry by phone. Give first-time prospects a concrete reason to commit now, such as a named session, a meeting or a seat limit. More: how to increase conference attendance.

What is a good back-of-room sales conversion rate?

No rigorous public benchmark exists. The published figures are self-reported by stage-selling trainers, from 25% of “most rooms” to a claimed 60% average, so treat 10% to 25% of attendees as a working range. Divide buyers who pay at the event by attendees present when the offer is made, and recount after cancellations: in the US, the FTC Cooling-Off Rule lets buyers cancel sales of $130 or more at temporary locations such as hotels until midnight of the third business day. More: back-of-room sales conversion rate.

What can a charity gala sponsor receive without it counting as advertising?

In the US, under 26 CFR 1.513-4, a qualified sponsorship payment can buy acknowledgment: the sponsor’s name, logo and value-neutral product listings. Messages with qualitative or comparative language, price information, an endorsement or an inducement to buy are advertising. Benefits such as seats and dinners are disregarded only if they total no more than 2% of the payment. If a payment is partly a gift and partly for goods or services and exceeds $75, the IRS requires a written disclosure statement. More: how to get sponsors for a charity gala.

Can I use WhatsApp to follow up high-ticket leads?

Yes, but Meta’s rules shape the cadence. Free-form messages are only possible inside a customer service window. Meta’s WhatsApp pricing documentation says “Non-template messages can only be sent within an open customer service window”, and the window closes once “24 hours have passed since the user last messaged you”. Outside it you can only send approved templates. US leads are a special case: Meta’s developer documentation states “WhatsApp does not currently deliver marketing template messages to WhatsApp users with United States phone numbers (numbers composed of a +1 dialing code and a US area code)”. In practice, get the lead to message you first, for example from a click-to-WhatsApp ad or a form that opens the chat. Answer within minutes and book the call while the window is open. Use calls, SMS or email to re-open the conversation with a US lead rather than a marketing template. More: WhatsApp follow-up for high-ticket leads.

Can a UK retreat keep a deposit if a guest cancels, and how fast must refunds be paid?

If the retreat is a package under the Package Travel and Linked Travel Arrangements Regulations 2018 (for example accommodation combined with transport, or with tourist services that make up a significant part of the value), a non-refundable deposit has to be justifiable as a termination fee. Regulation 12 lets a traveller terminate “at any time before the start of the package” and allows “an appropriate and justifiable termination fee”. The contract may set reasonable standard fees based on how close to the start the cancellation falls and the expected cost savings and resale income. You must justify the amount on request. If “unavoidable and extraordinary circumstances” at the destination significantly affect the package, the traveller can cancel without a fee and gets a full refund. Regulation 14 requires refunds and reimbursements “without undue delay and in any event not later than 14 days after the package travel contract is terminated”. So set a deposit and a staged balance schedule that track your real costs, write the fees into the contract, and keep a refund float. legislation.gov.uk lists amendments to these Regulations that it has not yet applied, so check the current version. General information, not legal advice. More: retreat deposits and payment plans.

How do you start a private members club in the UK?

Choose the legal shape before you sign a lease, because it decides the licence. The Home Office’s section 182 guidance says proprietary clubs, run commercially for profit, need a premises licence. Only a qualifying members’ club can use a club premises certificate. Under section 62 of the Licensing Act 2003 it must have at least 25 members and a gap of at least two days between application and admission. General information, not legal advice. More: how to start a private members club.

How many businesses compete for the same consulting work in Australia?

More than you can name, and the churn matters more than the count. The ABS release Counts of Australian Businesses, including Entries and Exits (published 18 August 2026, covering July 2022 to June 2026) records that “at 30 June 2026 there were 2,814,778 actively trading businesses in the Australian economy, with 996,203 of these businesses being employing”. In 2025–26 there was a “3.1% or 85,130 increase in the number of businesses”, a “16.9% entry rate, with 460,461 entries” and a “13.8% exit rate, with 375,331 exits”. Professional, Scientific and Technical Services – the division most consultancies sit in – grew 3.6% over the year. Two things fall out of those numbers for a consulting practice. First, roughly two in three Australian businesses employ nobody, so most of your competitors are one person with a laptop and a network, which means you are rarely losing on capability and often losing on being findable at the moment the need appears. Second, and more importantly for pipeline: 375,331 businesses exited the market in a single year. A referral network is not a stable asset, it is a decaying one – the client who sent you three jobs in 2024 may not be trading in 2027, and the buyer who championed you has moved. That is the argument for having a second channel that runs at a constant rate: not because referrals stop working, but because the population generating them turns over by roughly a seventh every year. More: how consultants get new clients in Australia and lead generation for consultants in Australia.

How do Australian independent and private schools grow enrolments when the school-age population barely is?

By winning switchers, because that is where all of the growth now is. The ABS Schools release for 2025 records 4,160,918 students enrolled in 9,673 schools nationally: 2,613,404 in government schools (62.8%), 831,692 in Catholic schools (20.0%) and 715,822 in independent schools (17.2%). Overall enrolment growth was 0.7%, or 28,912 more students. The composition of that growth is the whole story: “government school enrolments recorded a decrease of 0.2% (6,109 less students)” while “non-government school enrolments recorded an increase of 2.3% (35,021 more students)”. Non-government schools added more students than the country added in total. So an independent school is not competing for a growing pool of children; it is competing with the other non-government schools in its catchment, and with the local government school, for a family that has decided to move. That changes what marketing is for. The decision usually turns on the tour, not the brochure – which makes tour bookings, not enquiries, the metric that actually predicts an enrolment. The constraints are almost always operational: enquiries that arrive after hours and wait days for a first human response; sibling and past-family lists that are never systematically worked; and applications that stall between offer and acceptance with nobody assigned to them. Fixing response time on an enquiry costs little. More: lead generation for education providers and education marketing agencies in Australia.

Can we pay an education agent commission to recruit international students?

Not for onshore transfers – that changed this year, and any agent contract written before it is now out of date. The Department of Education’s summary of changes to the legislative framework for overseas students states that the National Code of Practice for Providers of Education and Training to Overseas Students 2018 “was amended in January 2026 to introduce a ban on the payment of education agent commissions in relation to onshore transfers (i.e., the recruitment of overseas students who have already commenced studying with another registered provider)”, on the basis that it “removes incentives for unscrupulous education agents to facilitate unnecessary transfers that may not be in the student’s best interests”. Two carve-outs matter operationally. The ban “does not apply where the relevant student has been accepted for enrolment by the relevant provider on or before 31 March 2026”, and because it reaches transfers only, it “will not impact students enrolling in further study after they complete their principal course or where a student is progressing through the package of courses for which their visa was granted”. Alongside that, the department says the ESOS Act changes made with the passage of the Education Legislation Amendment (Integrity and Other Measures) Bill 2025 replace the definition of ‘agent’ with a new definition of ‘education agent’, introduce a definition of ‘education agent commissions’ with collection of commissions information, and improve transparency of education agent information for providers. The commercial consequence is straightforward: a recruitment channel whose economics depended on moving students already in Australia between providers has had those economics removed, and the providers who cope best are the ones with genuine offshore demand generation and a direct enquiry pipeline they own. Check your own agent agreements against the fact sheets rather than against this summary – general information only, not legal advice. More: CRICOS provider international student recruitment and student recruitment in Australia.

Our rankings haven’t moved but our organic traffic is down – what changed?

The clicks behind the ranking moved, not the ranking. Ahrefs re-ran its clickthrough study on 300,000 keywords – 150,000 carrying an AI Overview and 150,000 informational keywords without one – comparing December 2023 with December 2025, and found that “the presence of an AI Overview now correlates with a 58% lower average clickthrough rate” for the top-ranking page, up from 34.5% when the same study was first run. The detail underneath is the part worth internalising: average position-one clickthrough on AI Overview keywords fell from 0.073 to 0.016, and on informational keywords with no AI Overview it fell from 0.076 to 0.039. Even the control group roughly halved. So position one in 2026 is not the asset position one was in 2023, and a flat rankings report next to a falling traffic graph is not a contradiction or a tracking bug – it is the expected result. It is correlation across a large sample rather than a controlled experiment, and it measures desktop clickthrough only. Three things to do about it. Separate impressions from clicks in your reporting so you can see the decoupling instead of arguing about it. Read your own server logs for assistant fetches, because a page being retrieved by an assistant is real distribution that analytics will file as nothing. And shift the goal from ranking to being the thing quoted – extractable answers, a hard number with a method attached, and third-party corroboration. More: AI referral traffic as a lead channel and AI SEO and AEO agencies in Australia.

Can a school, RTO or coaching business email its own past enquirers and students?

Often yes, on inferred consent – but it is narrower than most education marketers think, and you carry the burden of proof. The ACMA recognises two kinds of consent, express and inferred, and states plainly that “under the Spam Act, it’s up to you to prove that you have a person’s consent”. On the inferred kind: “you may infer that you have consent to send marketing messages if the recipient has knowingly and directly given their address and it is reasonable to believe they would expect to receive marketing from your business. This is usually when a person has a provable, ongoing relationship with your business, and the marketing is directly related to that relationship.” The two limits in ACMA’s own examples are where education providers get caught. Relevance is judged narrowly – ACMA’s illustration is that a bank telling a customer about a higher-interest savings account is covered, but “it would not cover the bank trying to sell them insurance products”. And a completed transaction is not an ongoing relationship: inferred consent “does not cover sending messages after someone has just bought something from your business”. Applied to an education database, a parent who enquired about a Year 7 place is a defensible inferred-consent audience for that year’s open day; the same parent is not a defensible audience for an unrelated holiday program, and a graduate from four years ago is a weak case for anything. ACMA also notes that “inferred consent is not as reliable as getting someone’s express consent”, and that with a marketing list you buy or use “you are still responsible for making sure you have consent for any addresses you use”. Every message must identify you as the sender, carry contact details and make it easy to unsubscribe. The practical answer is to convert inference into evidence: collect express consent at enquiry, log when and how, and re-permission the old records rather than assuming them. General information only, not legal advice. More: RTO lead generation in Australia and ASQA-compliant marketing for RTOs.

Speed, follow-up and conversion

How can a services business increase revenue without increasing ad spend?

Work the demand you already have harder. Three levers, in order of speed: (1) respond to existing leads faster – the Lead Response Management study found “the odds of contacting a lead if called in 5 minutes versus 30 minutes drop 100 times”, and Harvard Business Review’s 2011 study of 1.25 million sales leads across 29 B2C and 13 B2B US companies found firms that tried to make contact within an hour were nearly seven times as likely to qualify the lead as those that tried an hour later – Australian numbers in our lead response time benchmarks. (2) Reactivate the dormant database you already own – past clients and old enquiries carry zero media cost: database reactivation services. (3) Lift show rates on the appointments you already book, with confirmation sequences and tighter qualification. None of the three needs an extra ad dollar.

How fast should you follow up on a new lead?

Within five minutes. The Lead Response Management study, run with Professor James Oldroyd across three years of data from six companies, more than fifteen thousand web-generated leads and over one hundred thousand call attempts, found that “the odds of contacting a lead if called in 5 minutes versus 30 minutes drop 100 times” and that “the odds of qualifying a lead if called in 5 minutes versus 30 minutes drop 21 times”. A separate analysis reported in Harvard Business Review – 1.25 million sales leads received by 29 B2C and 13 B2B US companies – reached a compatible conclusion: firms that tried to contact a prospect within an hour were nearly seven times as likely to qualify the lead as those that tried an hour later, and more than 60 times as likely as those that waited 24 hours or longer. The same HBR article separately audited 2,241 US companies and found an average response time of 42 hours among companies that responded within 30 days, with 23% never responding at all – a different study, and worth keeping straight, because the 7x and the 42 hours get quoted as if they were one finding. Both datasets are more than a decade old; we cite them because nothing better-powered has replaced them, not because buyer patience has stayed still. This is still the single best-documented lever in lead generation, and it’s why every LeadsNow campaign runs instant AI response. More detail: the 5-minute rule in Australia.

Is speed-to-lead more important than lead scoring?

For most SMEs, yes. Lead scoring optimises which leads get attention; speed-to-lead optimises whether the lead is still reachable at all – and the decay curve above says reachability expires in minutes. Scoring earns its keep once volume genuinely exceeds capacity. We compare the two approaches, with the research, in speed-to-lead vs lead scoring.

How do you calculate contact rate correctly?

Contact rate = unique leads reached in a right-party, two-way conversation within a fixed window of arriving (14 days is a sensible default) ÷ unique leads in that same intake cohort. Voicemails, ringouts, gatekeepers and delivered SMS are attempts, not contacts; an SMS reply or an inbound callback is a contact. The common mistake is dividing a calendar month’s logged conversations by that month’s new leads, which counts last month’s leads in the numerator and judges late-month leads on a few days of effort – in our illustrative worked example, one September reports anything from 14.2% to 47.3% against a true 41.0%. None of this is new: survey researchers formalised it in the AAPOR Standard Definitions (10th edition, 2023), which say temporary, attempt-specific codes “should be replaced with final disposition codes” before rates are reported, and which define two bases worth copying: CON1 “assumes that all cases of indeterminate eligibility are eligible”, while CON3 “includes in the base only known eligible cases”. Report both; the gap between them is your data-quality problem, measured. Full method and worked example: how to calculate contact rate.

How do you reduce no-shows for booked sales appointments?

Confirm fast, remind more than once, and qualify tighter. The best-powered evidence on reminders comes from healthcare rather than sales, so read it as a mechanism, not as a benchmark for your calendar. A systematic review of 29 studies of telephone and SMS reminders for hospital appointments found that “the weighted mean relative change in non-attendance was 34% of the baseline non-attendance rate” – and, against the usual assumption, that “Automated reminders were less effective than manual phone calls (29% vs 39% of baseline value)”. Content matters as much as cadence: in the first of two randomised controlled trials at Barts Health NHS Trust (19,959 participants across both), an SMS stating the specific cost of a missed appointment produced a did-not-attend rate of 8.4%, against 11.1% for the standard message, and the second trial replicated the effect – at no extra cost. Read across to a sales calendar and the instruction is: automate the cadence so nothing is missed, but put a human call on the appointments you care most about, make the reminder say something rather than merely notify, and shrink the gap between booking and meeting date. Tighter qualification does the rest, because loose qualification manufactures no-shows. The full playbook is in how to improve sales appointment show rates.

Do text message reminders actually reduce no-shows?

Yes, in the best evidence available. A Cochrane review of mobile-phone reminders (8 trials, 6,615 people) found text reminders raised attendance against no reminder in 7 trials with 5,841 people (risk ratio 1.14), with overall attendance of 78.6% with texts against 67.8% with no reminder, and in 3 trials with 2,509 people texts performed about the same as phone-call reminders (risk ratio 0.99). The trials were healthcare appointments, so treat the size of the lift as indicative for sales calls. The copy and timing we use: the SMS reminder sequence that cuts no-shows.

Does messaging a no-show to reschedule get them to turn up?

It roughly doubled attendance in one randomised trial. In a 2024 trial in the American Journal of Ophthalmology, every patient who missed an appointment got the standard mailed letter, and those also sent a patient-portal message to reschedule within one business day attended within 30 days at 22.2% (42 of 189), against 11.6% (20 of 173) for the letter alone. It was a healthcare setting, not sales calls. The rebooking process for missed sales calls: how to rebook leads who no-showed.

What is a good show rate for sales calls?

It depends most on how far out the call is booked, and public sales data is thin. The largest sales-meeting dataset we could verify at source is RevenueHero’s December 2024 no-show benchmark: of 6,428 booked B2B meetings, 76.1% were completed and 6.5% were outright no-shows, with no-show rates running from 0% in healthcare (179 meetings) to 15.1% in real estate and 18.1% in education and e-learning. Count cancellations and reschedules against your show rate as well, because a no-show rate leaves them out. Lead time and reminders move the number: a systematic review of 105 no-show studies found high lead time and prior no-show history were the most commonly reported predictors of a missed appointment, and a randomised trial of 54,066 primary-care patients (Steiner et al., 2018) found reminders three days and one day before the visit cut missed appointments to 4.4%, against 5.8% for a three-day reminder alone. Those are healthcare figures, not sales calls, but the mechanics carry over. On our own accounts, show rate varies by offer and reminder cadence, up to 93% on our best-performing accounts – a ceiling from our record, not an average. More: sales call show rate benchmarks.

What is the average no-show rate for events?

About one in five. Across 1,070+ live events with check-in data, PheedLoop’s Event Data Lab Report #05 found a median no-show rate of about 20% (mean 29%). Free events had a median of about 28% (N=389) against about 17% for paid events (N=683), and small events of 10–49 expected attendees ran at about 32%. PheedLoop’s figure does not separate advance cancellations from people who simply did not turn up. For booked sales calls rather than events, see calendar link vs call-back booking rates.

Is it harder to get conference sponsors now?

For associations it was in 2023: an Eastside Rooms survey reported by ABPCO (fieldwork October 2023, 125 respondents) found 45% of associations and not-for-profits finding sponsorship harder and 41% seeing sponsor funds fall. In 2026, IDC’s sponsor survey of 150 B2B tech marketers found 67% rank cost per opportunity as a top success metric. More: why conference sponsorship got harder.

What is database reactivation, and what results can it get?

Database reactivation contacts the leads, past clients and dormant enquiries already sitting in your CRM – people who already know you – with AI calling and SMS, and books the warm ones back into your calendar. It’s typically the fastest ROI in lead generation because the media cost is zero. Before calling or texting an old US list, check it against the Reassigned Numbers Database, which identifies “numbers that were reassigned from a previous consumer”, and honour opt-outs: under 47 CFR § 64.1200(a)(10), a request to revoke consent made in any reasonable manner must be honoured “within a reasonable time not to exceed ten business days”. Our own benchmark from reactivating Colliers’ commercial property databases: 4.4% of contacted records converted to appointments on average, with an 8.9% peak on the best segment. The five ways to run one, with the campaign maths: database reactivation; step by step: database reactivation campaign guide, the buy-vs-reactivate maths in database reactivation vs buying new leads, and the full write-up of the Colliers-era numbers in 4.4% conversion on dormant leads.

Is it worth following up with leads that are months old?

Yes – in most CRMs the months-old leads vastly outnumber the fresh ones, and the buyers among them haven’t gone anywhere. The LinkedIn B2B Institute and Ehrenberg-Bass’s 95-5 rule found that about 95% of potential buyers aren’t ready to buy today – they will be “in-market” sometime in the future. And MarketingSherpa found that “79% of marketing leads never convert into sales”, with lack of lead nurturing given as the common cause (as compiled by HubSpot). That’s exactly what our long-term AI lead nurture service is built for – the mechanics of how the follow-up works are in long-term lead nurture with AI follow-up.

Why does our sales conversion rate fall as we scale?

Usually because the mix changed, not because your closers got worse. Adding spend, channels or markets adds segments that convert at different rates, and a blended average can fall even when every individual segment holds or improves. It is the same arithmetic trap documented in Bickel, Hammel and O’Connell’s 1975 paper in Science on Berkeley graduate admissions, where “Examination of aggregate data on graduate admissions to the University of California, Berkeley, for fall 1973 shows a clear but misleading pattern of bias against female applicants” – a pattern that disappeared once each department was examined on its own. So before you retrain the sales team, split conversion by source, offer, deal size and time-to-first-contact; there is usually one fast-growing, low-intent segment dragging the average down. Then fix the two levers that survive scale: response time and a qualification standard that doesn’t loosen as volume rises. More: how to increase sales conversion rate at scale, how to increase free trial opt-in rate, lead generation for scale-ups in Australia and the enterprise lead generation companies US buyers shortlist.

What is a good discovery call conversion rate?

About half of held discovery calls ending in a dated, accepted next step is a reasonable reference point – but no independent study measures discovery calls directly. The closest published proxy is First Page Sage’s sales funnel benchmarks (updated 10 August 2026), where SQL-to-opportunity conversion runs from 40% (real estate) to 66% (eCommerce) across 30 industries, a median of 49% by our count. It is a proxy, not the same measure: First Page Sage’s SQL has already “received a description of services and pricing information” and chosen to continue, its opportunity is a prospect with “a proposal or contract in hand”, and it publishes no sample size, so a raw every-held-call rate will usually sit below it. Measure yours as calls ending in a next step the buyer accepted and dated ÷ discovery calls actually held (no-shows belong to show rate), and don’t compare it with anything below about 40 held calls, where the 95% margin of error is roughly ±15 points. On what happens inside a converting call, Gong’s analysis of 519,000 recorded discovery calls found the most effective ones uncover three to four business problems. Benchmarks and method: good discovery call conversion rate benchmarks.

What is a good proposal acceptance rate?

Roughly a third to a half, depending on what you count. Proposify’s analysis of 742,137 proposals puts the average close rate at 34%, from 14% (hotel and venue) and 21% (construction) up to 53% (non-profit). RFP teams in Loopio’s 2026 benchmarks (1,500+ teams) won 39% of bids, or 45% averaged across 2019–2026. The 46th Deltek Clarity A&E study of nearly 700 US and Canadian firms (May 2025) reports a median win rate of 50%; Deltek’s 47th study, released in May 2026, gives no win rate in its public summary. Those are different denominators – quotes sent through software versus tenders a firm chose to enter – so do not average them. Deltek also shows why the rate alone misleads: proposal volume dropped 38% in the same study in which the median win rate rose to 50%. Measure proposals accepted ÷ proposals sent on a send cohort with a fixed window, and always report it next to quotes ÷ enquiries. We found no credible Australian proposal-acceptance benchmark. Full table and threshold rule: good proposal acceptance rate benchmarks.

Why do Instagram DM ad leads message us but never book a call?

Usually because the reply comes too late or the conversation never asks for a specific time. Meta’s Messenger Platform and IG Messaging API policy, which covers messages sent through Meta’s API and the inbox tools built on it, says “Businesses have up to 24 hours to respond to a user”, and its Human Agent tag “allows businesses to manually respond to user messages within a 7-day period”. So answer within minutes while the lead is still in the app, qualify in two or three questions, and offer two concrete call times inside the 24-hour window rather than a bare booking link. More: Instagram DM ads leads not booking.

How many AI DM setter conversations end in a booked call?

About 4.1% in the largest published dataset, a vendor study. SetSmart’s AI DM conversation study covers 828,761 conversations across 391 businesses from July 2024 to March 2026: 34,091 ended with the lead agreeing to book, which is about 10.2% of leads who replied. “Booked” there means the lead agreed in the chat, not that a call was held. How to read it against your own funnel: AI DM setter booking rate benchmarks.

Does Meta’s 24-hour messaging window stop me replying to Instagram DMs by hand?

Not as Meta documents it. The window is written as a rule for apps sending through Meta’s APIs: “Your app has 24 hours to respond to any message sent from an Instagram user to your app user” (Instagram Messaging API documentation), and the Messenger Platform policy adds a Human Agent tag allowing manual responses “within a 7-day period”. Neither page describes a limit on a person typing a reply in the Instagram app. How to work those conversations: qualifying leads in Instagram DMs.

Which books more calls: SMS, email or a phone call?

Use all three in sequence, because each fails differently on its own. A call lets you qualify and book in one conversation, but in Pew Research Center’s July 2020 survey of US adults only 19% said they generally answer their cellphone when an unknown number calls. So text first to say who you are and why you are calling, call straight after, and use email for the detail and the booking link. More: SMS vs email vs phone follow-up.

What should a sales cadence look like after a demo request?

Dense and short, because the buyer asked. A workable inbound template is 9 touches over 10 days: an email within 2 minutes, a call within 5 and an SMS at 15 minutes, then follow-ups on days 1, 3, 5, 7 and 10. Cold outbound needs wider spacing, such as 8 touches over 18 days. Stop on a reply, a booking or an opt-out: in the US, 47 CFR 64.1200(a)(10) treats a “stop” reply as revoking consent, to be honoured within ten business days at most. Five templates: sales cadence templates.

How much ad spend is wasted on leads nobody works?

Your own CRM has the answer: count the leads from a period with no logged contact attempt and multiply by your cost per lead for that period. The best-known audit shows how common the gap is – for Harvard Business Review in 2011, researchers sent a web-generated test lead to 2,241 US companies and found that 37% responded within an hour, while 23% never responded at all. That is a share of companies rather than of leads, and the data is old, so treat it as a reason to measure, not as your number. More: how much ad spend is wasted on unworked leads.

Why don’t leads reply to our follow-up texts?

Often because the text looks like a scam, or never arrived. People have reason to be wary: the FTC says people reported $470 million in losses to text scams in 2024, more than five times the 2020 number. Carriers filter as well – Twilio’s error 30007 means a message was “filtered (blocked) by Twilio or by the carrier”. So name your business and the enquiry in the first line, check delivery reports for filtered messages, and register your sender wherever the rules require it. More: why leads don’t reply to SMS.

What is a good lead-to-booked-call rate?

We know of no independent benchmark; the published figures come from scheduling vendors measuring their own customers, so attribute any number you use. Chili Piper’s analysis of nearly 4 million form submissions, mostly B2B, found 14.1% were not qualified and 66.7% of qualified submissions booked a meeting, while RevenueHero calls “around 50–60% conversion” from qualified lead to meeting “a common benchmark”. Both describe inbound demo requests booked through instant scheduling tools, so they are not a fair yardstick for paid-social or reactivated leads; measure yours as booked calls ÷ qualified leads from the same intake period. More: good lead-to-booked-call rate benchmarks.

What is a good inbound lead conversion rate by channel?

Inbound conversion rates run from 1.1% for paid social to 8.1% for organic search in professional services, against a 5.13% average across 13 industries, in Ruler Analytics’ 2026 conversion rate benchmarks, which cover 110 million+ sessions. Across all industries, paid search converts at 5.4% and organic search at 4.9%, while organic and paid social sit at 2.23% and 2.11%. The bigger swing often comes after the form: Chili Piper’s demo form benchmark report found 66.7% of qualified form submissions booked a meeting with an instant scheduler, against what it calls an industry average of 30%. Detail: inbound lead conversion rate by channel.

AI voice agents and compliance

Do AI voice agents actually work for outbound sales calls?

Yes, within honest limits. In Deepgram and Opus Research’s State of Voice AI survey of 400 business leaders (North American, 82% US-based, 83% at companies above US$100M revenue), 80% of organisations reported using some form of voice agent – a span the report defines as reaching from traditional IVR through to AI-powered solutions, so it is not a count of modern AI agents. The direction of travel is not in question; the size of the modern-agent share is. Modern agents handle structured calls – qualification, booking, reminders, reactivation – very well. Speech recognition is still not uniformly accurate across speakers: a PNAS study of five commercial ASR systems from Amazon, Apple, Google, IBM and Microsoft transcribed structured interviews with 42 white and 73 black US speakers and found an average word error rate of 0.35 for black speakers against 0.19 for white speakers. Test any agent against the accents in your actual market, not a demo script. Our full, vendor-neutral capability audit: what AI voice agents can and can’t do in 2026.

Are AI outbound sales calls legal in Australia?

Yes, when run correctly. Outbound sales calls – human or AI – fall under ACMA’s telemarketing rules: permitted calling hours, caller identification, and immediate action when someone asks not to be called. Numbers on the Do Not Call Register generally can’t be cold-called without consent or an existing relationship – an AI agent gets no exemption for not being human. The rules are enforced: in July 2026, TAB paid more than $2.7 million in penalties for breaches that included calls to Do Not Call numbers and calls outside permitted hours. Reputable operators also announce call recording up front in every state. For the full Australian rulebook – permitted hours, the three-month consent clock and when cold AI calling is worth doing at all – see our AI cold calling guide. Selling into the US? The rules are stricter – see our TCPA compliance guide for AI voice and SMS.

Is it legal to use AI to call existing customers in Australia?

Yes, if each call meets the same rules as a human telemarketing call. The Do Not Call Register’s industry guidance says consent can be “reasonably inferred” from a business relationship, and that it “ends immediately” when the customer says no. A call that only reminds or reschedules an appointment is not telemarketing; add an offer and the sales rules apply, including 9am to 8pm weekday and 9am to 5pm Saturday calling hours. General information, not legal advice. The full test: is it legal to use AI to call existing customers?

Are AI outbound sales calls legal in the United States?

Yes, with consent – and the bar is higher than Australia’s. The FCC has ruled that AI-generated voices are “artificial” under the TCPA, so an AI voice calling without the required consent is treated like an illegal robocall, and the statute (47 U.S.C. §227) carries private rights of action of $500–$1,500 per call. Text messages count too. The practical rule: call or text only leads who gave prior express consent (your own form fills, your own CRM), identify who’s calling, and honour opt-outs immediately. Full detail in our TCPA compliance guide for AI voice and SMS agents.

What happens when an AI agent gets something wrong on a call?

Design for it, because the liability is real: in Moffatt v Air Canada (2024), British Columbia’s Civil Resolution Tribunal held the airline liable for its chatbot’s wrong advice that a bereavement fare could be claimed retroactively, ordering it to pay $812. The tribunal found that “Air Canada did not take reasonable care to ensure its chatbot was accurate”, and rejected the airline’s submission that the chatbot was “a separate legal entity that is responsible for its own actions”. The mitigations are boundaries (the agent qualifies and books; it doesn’t invent pricing or terms), instant human handoff on anything emotional or contractual, and full call logging so every conversation is auditable. Ask any vendor you evaluate to show you all three.

Do I have to register my SMS sender ID in Australia?

If you send SMS under a branded alphanumeric sender ID (your business name rather than a phone number), yes. Under the ACMA’s SMS Sender ID Register rules, from 1 July 2026 participating telcos must over-stamp unregistered sender IDs with the word “Unverified”, and messages carried by non-participating providers are blocked. Two scope limits matter and are easy to get wrong: ACMA states plainly that the rules do not apply to text messages sent from phone numbers instead of a branded sender ID, and they do not apply to messaging apps such as WhatsApp, iMessage and Facebook Messenger. Registration is lodged through a participating telco, and in its register FAQs ACMA states that “most applications are approved within 24 hours”. Registration is a separate question from whether a message is lawful: marketing texts must still meet the Spam Act’s consent, identification and unsubscribe rules. Full detail in our guide to SMS sender ID registration in Australia, and the full sending checklist in our AI SMS outbound automation guide.

How fast must an Australian business honour an SMS or email unsubscribe?

Within five working days. The ACMA’s guidance on avoiding spam says every commercial message must carry an unsubscribe option that “honours a request to unsubscribe within 5 working days”, works for at least 30 days after the message is sent, and costs nothing beyond the usual charge for using the address. Separately, from 1 July 2026 texts sent under a branded sender ID need that ID on the ACMA’s SMS Sender ID Register. Broader context: lead generation in Australia.

What is A2P 10DLC, and do I need it to send SMS in the United States?

A2P 10DLC is a carrier-level registration regime, not a law. If you send application-to-person SMS from a US 10-digit long code, you register a Brand and then a Campaign with The Campaign Registry through your provider. Twilio documents that unregistered US-bound 10DLC traffic has been fully blocked since 1 September 2023, returning error 30034, and that messaging rates still apply to the blocked messages. It is separate from the TCPA: registration does not make a message lawful, and consent does not get you throughput – you need both, and they fail independently. Full detail in our guide to A2P 10DLC registration for outbound SMS in the USA.

How quickly must a US business honour a STOP reply to a text?

Within a reasonable time not exceeding ten business days. The FCC’s TCPA rule at 47 CFR 64.1200(a)(10) treats replies such as “stop”, “quit”, “end”, “revoke”, “opt out”, “cancel” and “unsubscribe” as a revocation of consent, and the request must be honoured within that limit. This is general information, not legal advice. How opt-outs affect your SMS numbers: SMS follow-up benchmarks.

Can a business in a regulated industry use an AI sales agent in Australia?

It can, but a sector layer sits on top of the general rules and it is the sector layer that trips people up. In financial services, ASIC’s Regulatory Guide 38 states at RG 38.36 that the hawking prohibition is technology neutral and extends to real-time interactions “through media that use artificial intelligence such as chat-bots”, and at RG 38.37 that it cannot be circumvented by contracting a third party to build the tool. Health information is sensitive information under the Privacy Act, and the OAIC’s guidance on communications with patients treats a patient’s contact details as part of their health information – so a list exported from a practice-management system is sensitive information, not a mailing list. Registered training organisations carry marketing and transparency obligations that follow the work out to contractors: ASQA’s information and transparency practice guide applies the standard to advertising published by “the organisation, a third party or an expert engaged by the organisation”, and requires materials to identify where a third party is “recruiting prospective VET students on behalf of the organisation”. This is general information, not legal advice – confirm your own obligations. More in AI outbound for regulated industries in Australia.

What do Google, Yahoo and Microsoft require for cold email at scale in 2026?

Senders of 5,000 or more daily messages must authenticate with SPF, DKIM and DMARC and support one-click unsubscribe under RFC 8058. Google’s bulk sender guidelines set a hard ceiling of keeping spam complaint rates under 0.30% in Postmaster Tools, and Microsoft’s Outlook requirements for high-volume senders apply the same SPF, DKIM and DMARC bar above 5,000 messages a day. Note what actually happens there, because it is widely misreported: Microsoft first said non-compliant mail would go to Junk, then updated its post on 30 April 2025 to say it would instead reject it from 5 May 2025, with the error “550; 5.7.515 Access denied, sending domain [SendingDomain] does not meet the required authentication level.” Failing these requirements does not merely deprioritise your mail; it stops it being delivered. Full checklist: email and SMS deliverability for outbound at scale.

Why has my LinkedIn outreach stopped getting replies?

Rule out a restriction first, because it is the one cause that fixes itself. LinkedIn’s help page on invitation restrictions names triggers including sending many invitations in a short time and having many invitations “ignored, left pending, or marked as spam”. It says most restrictions are removed automatically within one week, that you should wait up to one month if you were restricted for too many outstanding invitations, and that LinkedIn won’t remove a restriction on request. Automation carries its own risk: LinkedIn’s User Agreement prohibits using “bots or other unauthorized automated methods” to add or download contacts or send messages. If the account is not restricted, split the drop into two rates: acceptance (invitations accepted ÷ invitations sent) and reply-after-acceptance (connections who replied ÷ connections you messaged). A fall in the first points to targeting, your profile or blank invitations; a fall in the second points to the first message. We know of no LinkedIn-published reply-rate benchmark, so compare four weeks of your own counts with your own baseline. More: why LinkedIn outreach stops getting replies.

Does the CCPA apply to B2B and employee contact data used for outbound sales?

Yes. The CCPA’s carve-outs for employment-related and business-to-business personal information expired on 31 December 2022, per the California Attorney General’s official CCPA guidance, so B2B leads and cold-outbound contact data are now full ‘personal information’ under the Act, triggering opt-out, deletion and Global Privacy Control obligations. Data brokers selling that data must also register. California is the strictest: CalPrivacy requires registration between 1 and 31 January each year and an annual fee of “$6,000 plus an associated third-party processing fee for electronic payments”, and a broker that fails to register by 31 January “may be liable for administrative fines and costs” under Civil Code § 1798.99.82(d). The California Lawyers Association records that Texas, Vermont and Oregon have each enacted their own data broker registration laws, differing in scope and design. Full breakdown: US state privacy laws and outbound sales lists.

Does CAN-SPAM apply to business-to-business email?

Yes. The FTC’s CAN-SPAM compliance guide says the law makes no exception for business-to-business email, so cold B2B email in the US needs accurate headers, a non-deceptive subject line, a valid postal address and a working opt-out honoured within 10 business days. Each separate email in violation can cost up to $53,088. This is general information, not legal advice. How it shapes outbound to a free tier: outbound for product-led growth SaaS.

Getting found in AI search

Are buyers really using ChatGPT instead of Google to find providers?

Increasingly, yes. The Answer Economy: G2’s 2026 AI Search Insight Report – an online survey of 1,076 B2B decision makers across North America, EMEA and APAC, fielded in March 2026 – found that half (51%) “now start their software research with an AI chatbot more often than Google” and that “71% say they now rely on AI chatbots for software research”. The direction of travel is the more useful number: G2 reports that “86% of B2B buyers increased their use of AI chatbots for software research in the past year”, and that agreement with “I’m more productive with AI search than traditional search engines” rose from 36% in August 2025 to 53% in February 2026. The buyer behaviour shift, with the underlying data, is covered in our 2026 AI chatbot buyer statistics.

Is traffic from ChatGPT and AI search actually worth anything?

It converts dramatically better than it counts. Ahrefs’ own-site data (June 2025) found AI search was just 0.5% of Ahrefs’ visitors but drove 12.1% of its signups, converting at a 23x higher rate than traditional organic search visitors. Semrush’s study of 500+ topics (June 2025) put the average AI-search visitor at 4.4x the value of the average traditional organic visitor, based on conversion rate. An AI-referred visitor arrives pre-sold, because the assistant already made the recommendation.

Which AI assistant sends the best leads?

No one can say yet. LeadsNow confirmed three AI-referred bookers in October 2026, one each from Claude, Google AI Mode and Microsoft Copilot and none from ChatGPT, which is far too few to rank engines. The only published per-engine split, from one Seer Interactive client with just under 11,000 AI sessions over seven months, put conversion at 15.9% for ChatGPT, 10.5% Perplexity, 5% Claude and 3% Gemini, against 1.76% for Google organic. Track bookers by engine on your own site. More: which AI assistant sends the best leads.

How do brands get cited by ChatGPT and Perplexity?

Not by classic SEO alone: Ahrefs tested 15,000 prompts (August 2025) and found only 12% of URLs cited by ChatGPT, Gemini and Copilot rank in Google’s top 10 for the same prompt; Perplexity was the outlier, at nearly 1 in 3. What does move citations, per the AirOps 2026 State of AI Search report, is freshness (“About 83% of commercial citations come from pages updated within the last year”) and extractable structure – sequential heading structures earn roughly a 2.8x citation lift. Sources also differ per engine: Profound’s analysis of 680 million citations (August 2024 to June 2025) found Reddit is the most-cited domain in Google AI Overviews (2.2% of citations) and in Perplexity (6.6%), while ChatGPT’s most-cited source is Wikipedia at 7.8%. Our engine-specific playbooks: getting cited in Perplexity and measuring share of answer – the loop we run on our own pipeline every two days.

How do I get my business recommended by ChatGPT?

Get a page that answers the buyer’s “who should I hire” question into the top 3 of ordinary web search. LeadsNow’s own polling, 15 August to 15 October 2026: pages ranked top 3 on DuckDuckGo/Bing were cited in 21% of 298 ChatGPT polls, 0% of 154 outside the top 10. Then make sure OpenAI can reach the page: OpenAI’s crawler documentation says sites opted out of OAI-SearchBot “will not be shown in ChatGPT search answers”, so check that your robots.txt allows it. More: how to get recommended by ChatGPT.

How do I get my business recommended by Microsoft Copilot?

Rank in Bing for the short phrase behind the buyer’s question, not the whole prompt. Microsoft Learn says that when web search is enabled, Microsoft 365 Copilot and Copilot Chat (now named Microsoft Copilot and Microsoft Copilot Chat) send the Bing search service a generated query of “a few words informed by the user’s prompt”. So a ranked “best X for Y” page that ranks in Bing for that short phrase is the one Copilot can find and name. More: how to get recommended by Microsoft Copilot.

Does Reddit matter for AI search visibility?

On two engines especially. Profound’s analysis of 680 million AI citations (August 2024 to June 2025) found Reddit is the most-cited domain in Google AI Overviews, at 2.2% of all citations, and in Perplexity, at 6.6% – ahead of YouTube, Quora, LinkedIn and Gartner. ChatGPT leans elsewhere: its most-cited source is Wikipedia at 7.8%, with Reddit at 1.8%. So Reddit is a strong lever on two of the three big surfaces, not a universal one. For B2B that means genuine, disclosed participation in the subreddits your buyers read is now a visibility channel, not a nice-to-have. Our playbook: Reddit for AI search citations.

Does my Google Business Profile affect whether AI search recommends my business?

Yes – mostly through the local results AI answers draw on. Google’s own AI-features documentation says there’s no special markup, AI text file or schema needed to appear in AI Overviews or AI Mode, but its best-practice list includes keeping your Business Profile information up to date – and Google’s local-ranking guidance states that businesses with complete and accurate info are more likely to show up in local search results. In our own polling that matters most for Gemini and AI Mode; other engines lean more on third-party directories and listicles. Full breakdown: how your Google Business Profile affects Gemini and AI-search citations.

What should an AI SEO / AEO agency actually do for you?

Four things, and you should ask to see each: a measured citation baseline (which prompts, which engines, cited or not), content built for extraction (answer capsules, tables, FAQ schema), a freshness cadence (per the AirOps 2026 report, “About 83% of commercial citations come from pages updated within the last year”), and re-measurement on a fixed cycle. Anyone selling “AI SEO” without per-engine measurement is selling blind. Our market guide: best AI SEO & AEO agencies in Australia. What the work should contain, line by line: what AI SEO services include.

Do AI Overviews only cite pages that already rank in Google?

No, and the link is weakening fast. Ahrefs’ March 2026 analysis of 863,000 keyword SERPs and 4 million AI Overview URLs found 37.9% of AI-Overview-cited URLs also appeared within the first 10 SERP blocks, with 31.2% at positions 11–100 and 31.0% beyond the top 100. Ahrefs contrasts this with roughly 76% in July 2025. Read the methodology before quoting the drop: Ahrefs says it has “improved our parsing methodology” since the earlier study and analysed over double the URLs this time, so this is a substantial weakening rather than a precise 38-point collapse. The practical consequence is that rank tracking alone is no longer a valid proxy for AI visibility, and you have to measure citations directly.

How long does AEO take to work?

Longer than a content sprint, and it never really finishes – AI visibility behaves like a cadence rather than a milestone. Two published sources set honest expectations. Google’s own Search Central documentation states plainly that “Google doesn’t guarantee that it will crawl, index, or serve your page, even if your page follows the Google Search Essentials”, so nobody can sell you a date. And the AirOps 2026 State of AI Search report (AirOps with Kevin Indig, December 2025) found the surface is genuinely unstable: “Only 30% of brands stay visible from one answer to the next, and just 20% remain present across five consecutive runs”, and pages not updated quarterly are 3× more likely to lose citations. Our own repeat-polling of assistants found the same instability across engines – the dataset is published in our first-party citation study. So we plan engagements around a first re-measurement at four to six weeks, a meaningful read at about three months, and a permanent quarterly refresh after that. Anyone promising a fixed date for a citation is promising something no engine has committed to. More: how long AEO takes to work, and how we define every number we publish on our methodology page.

How long does it take to get recommended by ChatGPT?

Weeks, not days, and often never for a given page. In LeadsNow’s tracking of 307 pages published 3 July to 14 October 2026, 170 appeared in DuckDuckGo web results for their target question (median 4.8 days), but only 3 of 86 checked in ChatGPT were cited, at 12, 20 and 35 days. Access is the fast part: OpenAI’s crawler documentation says it can take about 24 hours from a robots.txt update for its search systems to adjust. More: how long it takes to get recommended by ChatGPT.

How do financial advisers get clients from ChatGPT and AI search?

By ranking in the top 3 for the “best financial adviser for…” questions prospects ask, then making the review meeting one click to book. LeadsNow’s own polling, 15 August to 15 October 2026: pages ranked top 3 on DuckDuckGo/Bing were cited in 21% of 298 ChatGPT polls, 0% of 154 outside the top 10. In Australia, ASIC’s RG 38 explains the hawking prohibition, which aims to protect consumers from unsolicited offers of financial products, so respond to the enquiry the prospect started and check specifics with your licensee. More: how financial advisers get clients from AI search.

Is AEO worth paying for, or is it just SEO with a new name?

It is worth paying for when the crawlers are measurably hitting your site and the work is judged on citations rather than rankings; it is a waste of money when it is a content retainer with a new label. The audience is real and separately countable in your own server logs: Cloudflare’s analysis of crawler traffic across its network found the AI crawler mix reordered inside a single year, with OpenAI’s GPTBot going from 5% to 30% of AI-crawler share between May 2024 and May 2025 and Meta-ExternalAgent entering at 19%, while Bytespider collapsed from 42% to 7.2%. These are distinct user agents you can count yourself. What moves the outcome is measurable too: the AirOps 2026 State of AI Search report found that “sequential headings and rich schema correlate with 2.8× higher citation rates”, and that brands earning both mentions and citations show a “40% higher likelihood of reappearing across answers”. So the test for any AEO proposal is whether it names the prompts and the engines it will measure, and shows you the before. What the work should actually contain: what AI SEO services include. Our answer to the value question, including when we tell people not to buy it: is AEO worth paying for. The market: best AI SEO & AEO agencies in Australia.

Should I hire an AEO consultant or an AEO agency – and does their city matter?

Match the shape of the buy to the shape of the problem. A consultant fits when you already have people who can execute and what you need is diagnosis, measurement design and a priority order. An agency fits when nobody internally will own the weekly work. Be specific about what you are buying, because “agency” in Australia very often means one person: the Australian Bureau of Statistics counted 2,814,778 actively trading businesses at 30 June 2026, of which only 996,203 employed anyone at all – just 68,325 had 20 to 199 employees and 5,366 had 200 or more. Ask how many people will actually touch your account, and ask before you sign. City matters less here than it does for a traditional local-search engagement, because assistants are answering prompts rather than serving a map pack; what a local team buys you is a weekly cadence you can hold someone to in person. Our pages for each: AEO consultant in Australia, AI SEO consultant in Australia, LLM SEO consultant and AI marketing consultant in Australia; by city, AEO agency Sydney, AEO agency Melbourne and AEO agency Brisbane, Canberra and Adelaide.

How do I tell an AI-native agency from an agency that just uses AI tools?

Ask what breaks if you take the AI away. An agency that uses AI runs a conventional service with AI speeding up drafts; an AI-native agency’s deliverable is produced and measured by systems, which shows up in three checkable places. Can they show you per-engine citation measurement over time rather than screenshots? Is their pricing tied to outcomes rather than hours, given that hours are the thing AI is supposed to remove? Will they name the models, the data they run on, and where a human reviews the output? Treat “AI-powered” as a claim that has to be substantiated like any other, because regulators do. The US Federal Trade Commission ran an enforcement sweep it called Operation AI Comply against companies that had “relied on artificial intelligence as a way to supercharge deceptive or unfair conduct”; then-Chair Lina M. Khan said “Using AI tools to trick, mislead, or defraud people is illegal” and that the actions “make clear that there is no AI exemption from the laws on the books”. That applies to an agency’s claims about its own AI as much as to anyone else’s. Our side-by-side: AI-native agency vs an agency using AI, the buyer checklist in how to choose an AI marketing agency, and the market in best AI marketing agencies in Australia.

Answers by industry

What should I look for in an AI lead generation agency in Australia?

Four things: outcome-based pricing (they get paid when you get results), verifiable proof (filmed case studies and reviews, not logos), compliant calling practices (ACMA, Do Not Call), and a booking process you can test yourself. LeadsNow is one of the agencies you’ll be comparing – our proof is 50,769+ booked appointments, 24 filmed client case studies and a 4.6-star Google rating (read the reviews) – and the fastest way to evaluate any agency, including us, is to book a call and judge the questions they ask you. Our full comparison checklist, including the questions that separate operators from resellers, is in how to choose an AI marketing agency.

How does lead generation work for coaches and consultants?

High-ticket coaching sells through conversations, so the whole funnel is engineered toward booked calls with people who can invest – AI setters qualify for fit and intent before anything reaches your calendar. Details, with our coaching-industry proof: lead generation for coaches and for consultants.

How do high-ticket coaches keep their calendar full in 2026?

By treating booked calls as an operations problem rather than a content problem. The field keeps getting more crowded: the 2025 ICF Global Coaching Study, researched by PwC for the International Coaching Federation, reports “A record 122,974 coach practitioners worldwide (up 15% from 2023)” and that “industry revenue soared to $5.34 billion USD”. More coaches competing for the same audiences means lower reply rates on the same content, so the leverage moves to what a competitor can’t copy in a weekend: a specific offer, a database you own, and instant response on every enquiry – in Harvard Business Review’s 2011 study of 1.25 million sales leads across 29 B2C and 13 B2B US companies, firms that tried to contact a prospect within an hour were nearly seven times as likely to qualify the lead as those that tried an hour later, and more than 60 times as likely as companies that waited 24 hours or longer. (The same article separately audited 2,241 US companies and found an average response time of 42 hours among companies that responded within 30 days, with 23% never responding at all. It is 2011 data, and we cite it because nothing better-powered has replaced it, not because buyer patience has stayed still.) Full playbook: how high-ticket coaches fill their calendar in 2026 and how to increase your sales call booking rate.

How long does executive coaching usually take?

Often 7 to 12 months, though engagements vary widely. In Harvard Business Review’s research report The Realities of Executive Coaching (January 2009), 45% of the 140 coaches surveyed said their engagements typically last 7 to 12 months, 27.1% said 2 to 6 months and 17.9% said 13 to 18 months. It is an older, self-reported coach survey, so treat it as a guide rather than a standard. For a coaching firm, engagement length sets how many new clients it must book each month. More: how long executive coaching takes.

What should a coach do with leads that never bought?

Sort them into four piles: rebook no-shows within 24 hours, give unclosed calls one new reason to talk, send ghosts one yes-or-no question, and move leads already called ten times to another channel or a 90-day pause (a LeadsNow rule). Under the Spam Act, the ACMA says “it’s up to you to prove” consent. Under LeadsNow’s “trash to cash” model, a coaching and consulting client closes around $180,000 a month in cash collected purely from leads it had given up on. More: coaching leads that never bought.

How do gyms get more members with AI?

Gyms sit on the best reactivation databases in small business – ex-members and old trial enquiries respond unusually well to a well-timed SMS-and-call sequence, and speed-to-lead on new enquiries does the rest. Our gym system: More Gym Members.

How does AI lead generation work for real estate agents?

Real estate lead costs span an order of magnitude by channel – Ylopo’s 2026 cost guide puts social leads at roughly US$5–$30 (cheap because intent is low) and search leads several times higher – so the winners are decided by follow-up speed and nurture persistence, not lead source. Our Australian guide: lead generation for real estate agents.

What does a vendor lead cost for a real estate agent in Australia?

On Facebook lead forms, about A$48.71 per vendor lead: that is the median Rex Software reports across 152 Australian Rex Reach accounts with A$285,658 of ad spend, 1 October 2023 to 30 September 2024. Referral platforms charge at the other end of the funnel, as a share of commission paid only on sale, so compare sources on cost per sold listing rather than cost per lead, and follow up new vendor leads the same day. More: vendor leads in Australia.

How should financial advisers approach lead generation?

Adviser numbers are still falling – Riskinfo reported the count dropping below 15,000 in July 2026, to 14,984, the lowest since the Financial Adviser Register began in 2015 – and lead generation in advice is now regulator-watched: in February 2026 ASIC commenced a review of advice licensees using lead generation services and began publishing a list of lead generators, referral partners and the licensees that acquired their leads. Choose partners accordingly – our honest ranking of the market: best lead generation agencies for financial advisers.

How do accounting firms in Australia get new clients?

Referrals still bring in most new work, but they cap out at whatever your existing clients happen to send – you can’t scale word of mouth on demand. The firms growing fastest layer channels deliberately: formal partner networks with lawyers, brokers and financial advisers; visibility in AI search, where prospective clients now ask ChatGPT and Gemini to recommend a firm; outbound and AI appointment setting to reach businesses unhappy with their current accountant; and reactivating the dormant enquiries already sitting in the practice’s inbox. Competition for attention is real – CPA Australia alone counts more than 176,000 professionals across 150 countries and regions. We rank every channel, with the evidence, in how accounting firms get new clients in Australia – and the done-for-you version is lead generation for accounting firms.

How do mortgage brokers generate leads without buying shared lists?

The two channels that consistently work for brokers are instant response on inbound enquiries (rate questions go cold in minutes) and reactivating past-settlement clients and old enquiries – a broker’s book is a reactivation goldmine at refinance time. Shared purchased leads are resold to multiple brokers, so you’re paying to race. Our ranked market guide: best lead generation agencies for mortgage brokers.

How do commercial finance brokers generate leads?

The highest-converting channel is accountant and adviser referral partnerships, because the introduction arrives with trust attached. Aggregator and purchased leads add volume but are shared and price-shopped, so response speed decides who wins them. The overlooked asset is the broker’s own book: past clients and previously worked enquiries come back into market at every facility renewal, and AI appointment setting can work that database – plus every new enquiry – within minutes, around the clock. The broker channel for commercial lending has been growing fast: the MFAA’s Industry Intelligence Service report, 19th edition, found brokers settled a record $22.68 billion in commercial loans in the six months from April to September 2024, up 31.20% year on year, with a record 7,023 mortgage brokers also writing commercial loans. The full playbook: lead generation for commercial finance brokers.

How can an Australian insurance broker generate leads without breaching the hawking prohibition?

Make the client the one who asks for the conversation. ASIC’s Regulatory Guide 38 explains that under s992A of the Corporations Act a person must not, in the course of or because of an unsolicited contact with a retail client, offer financial products or invite the client to apply for them, and that contact is unsolicited unless the consumer has first given consent that is “positive, voluntary, and clear”. Four paragraphs shape a broker’s lead generation. RG 38.35 allows communications that are not in real time, such as a letter or email, to encourage a consumer to request future contact. RG 38.36 says the prohibition is technology neutral and extends to instant messages and AI chat-bots. RG 38.37 says it cannot be circumvented by engaging a third party, and that the person commissioning a chat-bot, or under whose licence it operates, is responsible for any offer it makes. RG 38.111 says consent can be relied on only for the six weeks after it is given (up to 12 weeks where a medical examination is needed), and RG 38.114 says those time limits apply equally to insurance brokers. So an appointment setter, human or AI, should turn interest into a recorded, scoped request for contact and a booked review with a licensed person, and leave quotes and advice to that review. This is general information, not legal advice. More: lead generation for insurance brokers in Australia.

How does AI lead generation work for SaaS companies?

SaaS is where AI outbound shines, because the maths is transparent: trials and demos have measurable conversion and LTV, so cost per qualified demo maps straight to CAC. AI setters work signup lists, product-qualified leads and churned accounts around the clock. Start with AI lead generation for SaaS and the SaaS agency comparison.

How long does a B2B SaaS sales cycle take?

It lengthens with contract size, and the hard data is thinner than confident benchmarks suggest. The most-cited independent anchor is old: the Bridge Group’s 2017 SaaS AE Metrics report (384 SaaS companies, 89% headquartered in North America) found a 60-day median sales cycle at a $25K median annual contract value, and noted that higher contract values bring longer sales cycles and larger buying units. The direction since is longer: in the seventh edition of Salesforce’s State of Sales report, 57% of 4,050 sales professionals in 22 countries said customers take longer to decide than they used to – a perception survey, not a measured cycle. We found no independent, current, ungated dataset of cycle length by deal size, so measure your own: days from first meeting held to contract signed, closed-won deals only, reported as a median rather than a mean and split by contract-size band, because a blended number describes no real deal. More: how long a B2B SaaS sales cycle takes in 2026.

Is my SaaS trial-to-paid conversion rate actually low?

Only if it is low against trials with the same entry mechanic. ChartMogul’s SaaS Conversion Report (February 2026, a self-reported survey of 200 B2B software products, written by Kyle Poyar) puts the “good” band at 4–6% for free trials and 25–35% when a card is required upfront, with “great” at 10–15% and 50–60%; the median across all products is 8%. Check its fraction before comparing: leads or free signups that become paying customers within six months, so a rate measured on the last trial day will read low against it. If yours is genuinely low, look in order at junk signups in the denominator, trials from outside your target customer, trials that never activate, and activated trials that stall at the purchase. On activation, Lenny Rachitsky’s survey of over 500 responses found a median activation rate of 25%, and 30% for SaaS products alone. More: why a trial-to-paid conversion rate is low.

Should a SaaS company offer a free trial or a demo?

Decide from the deal, not from competitors: a free trial suits a product that shows value quickly on a small contract, a demo suits long setups and large buying committees, and most B2B SaaS sits in between, where a trial plus an assisted call fits. Trials are the common entry point: ChartMogul’s SaaS Conversion Report (February 2026, 200 software products) found 57% lead with a free trial, and 80% of free-trial products add human touchpoints when an enterprise user enters the trial. More: whether a SaaS company should offer a free trial or a demo.

What is a good freemium to paid conversion rate?

For a self-serve freemium product, 3% to 5% of new accounts paying within six months is good and 6% to 8% is great, on Kyle Poyar and Lenny Rachitsky’s 2023 survey of more than 1,000 products. With a sales-assist motion, good rises to 5% to 7% and great to 10% to 15%. Measure it the same way: accounts created in one month that start paying within their first six months, divided by all new accounts created that month. More: how to convert freemium users to paid.

What is a good SaaS trial activation rate, and how do I raise it?

Around 30% is a sanity check: in Lenny Rachitsky and Yuriy Timen’s 2022 survey of 500+ products, SaaS activation averaged 36% with a median of 30%. Test your milestone first: users who reach it should retain at least 2x better than users who don’t. Then split the trial into signup, setup, first value and habit, and fix the rung with the biggest absolute drop before buying more signups, adding a human or AI conversation only where contract value pays for it. More: how to increase SaaS trial activation rate.

How do I win back churned SaaS customers?

Sort churned accounts by why they left, then give each reason one matched offer on one channel: a smaller plan for price leavers, the shipped feature for feature leavers, never a blanket discount. Treat failed payments as a billing queue, not a win-back case. It matters because ChartMogul found a 2025 median net revenue retention of 82% for B2B SaaS, so the median customer base shrinks before new sales. Report reactivation MRR on its own line. The grid and a worked model: how to win back churned SaaS customers.

How big an annual discount should a SaaS business offer?

Size it on churn. On a $100 monthly plan, a 20% annual discount ($960 a year) only out-earns monthly billing in year one for customers who would churn faster than about 4.2% a month, and two months free (16.7%) breaks even near 3.4%. Customers who churn more slowly would have paid you more on monthly billing. Published norms sit close by, on pricing pages read 25 October 2026: Notion advertises “Save up to 20% with yearly” and monday.com an 18% yearly discount. Worked tables: how to move SaaS customers to annual plans.

How does AI lead generation work for education providers and RTOs?

Education enquiries are high-volume, deadline-driven and brutally sensitive to response time – an enquiry about a course intake answered next day is an enrolment lost to whoever answered first. AI setters handle the volume spikes around intake periods without hiring. See lead generation for education companies and the education marketing agency comparison.

Can an agency white-label AI lead generation for its own clients?

Yes – it is a common arrangement, and it is how a lot of PPC and SEO shops add appointment setting without building a calling team. Two things decide whether it works. First, liability does not transfer with the work: the ACMA is explicit that businesses “cannot outsource their obligations under the spam and telemarketing laws through these arrangements. Ultimately, the business is responsible” (telemarketing and e-marketing – common issues and mistakes), and its guidance on avoiding spam adds that “If someone else sends messages on your behalf, the message must still identify you as the business that authorised the message.” Consent records, sender identification and unsubscribe handling therefore have to be contractually assigned and auditable, not assumed. Second, anything you claim under your own brand is yours to prove: the ACCC’s guidance on false or misleading claims states that “A business must be able to prove any claim they advertise” and that “It makes no difference whether a business intends to mislead or not.” Detail: white-label lead generation for agencies and lead generation for PPC agencies in Australia.

Does LeadsNow work with businesses outside Sydney and Melbourne?

Yes – AI appointment setting is location-agnostic: the calls, messages and bookings work the same way for a business in Perth, Brisbane, Adelaide or regional Australia as in Sydney or Melbourne. Local pages: Perth, Brisbane, Adelaide, Gold Coast.

Does LeadsNow work with US businesses?

Yes. The system is the same; the compliance regime isn’t – US outbound falls under the TCPA, which imposes strict consent requirements for automated calls and texts, with per-violation statutory damages. We run US campaigns compliant-by-design: our TCPA compliance guide covers the rules, and best AI appointment setting services USA covers the market.

Can US colleges pay recruiters or agencies commission per enrolled student?

No. The Higher Education Act’s incentive compensation ban (34 CFR 668.14(b)(22)) bars any Title-IV-eligible institution from paying commission, bonus or other incentive tied ‘in any part, directly or indirectly’ to securing enrollments or financial-aid awards, for anyone engaged in recruitment or aid decisions. The distinction that decides it is not flat fee versus percentage. The Department of Education’s Dear Colleague Letter GEN-11-05 lists “tuition as a source of revenue from which compensation is paid to an unrelated third party for a variety of bundled services” as not incentive compensation, while it lists “‘tuition sharing’ as a measure of compensation when based on a formula that relates the amount payable to the entity to the number of students enrolled as a result of the activity of the entity” as compensation that is caught. So an unaffiliated vendor can be paid out of tuition for a genuine bundle of services, but the moment the amount tracks enrolments won it is prohibited, and recruitment must not be separately priced inside the bundle. (An earlier version of this answer described “narrow safe harbors” for flat fees. That was wrong on both counts, and we corrected it on 9 September 2026.) FERPA also limits what student-record data a vendor can see: under 34 CFR 99.31, an outside contractor counts as a school official only if it performs a service the institution would otherwise use employees for, is “under the direct control” of the institution for the use and maintenance of education records, and is bound by the redisclosure limits in §99.33(a). Detail: student recruitment for US education providers.

What is government-funded VET in Australia, and how does it change RTO lead generation?

Government-funded vocational training runs through state-contracted programs, such as NSW’s Smart and Skilled and Victoria’s Skills First, so a subsidised place is only available through a contracted provider and only to students who meet that state’s eligibility rules. Under the Standards for RTOs that took effect on 1 July 2025, Standard 2.1 requires that “all information provided to VET students by the organisation or any third parties is clear, accurate and current”, including “the availability of any relevant government training entitlements and subsidies”, and ASQA’s Information Practice Guide lists “False or misleading marketing” and weak oversight of “third parties that market to, or recruit, students” among the known risks – so a lead-generation partner’s ads are the RTO’s compliance problem. Misrepresenting funding is also consumer-law territory: in a case brought by the ACCC, the Federal Court ordered Phoenix Institute and CTI to pay $438 million in penalties, in part for representing courses as free. Lead generation has to qualify eligibility before enrolment, not after. Detail: government-funded RTO lead generation in Australia.

How does NSW’s Smart and Skilled program affect RTO enrolment marketing?

Smart and Skilled is NSW’s subsidised training market: only providers on the Provider List can deliver subsidised training (the current list covers the 2026–27 activity period), and only qualifications on the NSW Skills List attract a subsidy. Under the Smart and Skilled Prices and Fees schedule (v16.2), the concession student fee is $160 for a Certificate II and $240 for a Certificate III or IV, and concessions do not apply at Diploma or Advanced Diploma level. The disclosure obligation comes from the training standards themselves rather than from any ASQA marketing guide. Standard 2.1 of the National Vocational Education and Training Regulator (Outcome Standards for Registered Training Organisations) Instrument 2025, in force since 1 July 2025, requires an RTO to demonstrate that “all information provided to VET students by the organisation or any third parties is clear, accurate and current”, and that students can easily access “all fees, costs and charges associated with the provision of the training product”, including “the availability of any relevant government training entitlements and subsidies”. Advertising a conditional subsidised place as unconditionally free fails that test. Detail: Smart and Skilled RTO enrolment marketing NSW.

Is Victoria’s Skills First ‘upskilling’ rule still in effect?

No. The upskilling requirement, which previously blocked funding unless a course was a higher qualification level than one already held, was completely removed from 1 January 2023, per the Victorian Government’s official Skills First eligibility guidance. Students can now enrol in a subsidised course regardless of prior qualifications, subject to the standard course limits – the same Victorian guidance states Skills First funding is limited to “2 courses at a time; 2 courses in a year” – and to citizenship or residency requirements plus being physically present in Victoria or a designated border region during training. Any marketing or vendor claim citing the old rule is out of date. Detail: Skills First RTO enrolment marketing Victoria.

What compliance regime applies to recruiting international students for a CRICOS course?

CRICOS providers sit under the Education Services for Overseas Students Act 2000 and the National Code 2018, whose Standard 1 requires marketing, including through education agents, to be “not false or misleading” and consistent with Australian Consumer Law, bars a provider from claiming it will secure a migration outcome for a student, and requires its CRICOS registered name and number on written and online material. Visa processing is also rationed: Ministerial Direction 115 replaced MD111 from 14 November 2025 and gives higher processing priority to providers that manage their international enrolments broadly in line with their indicative allocations under the 2026 National Planning Level of 295,000 places. For VET and English-language providers the door is also shut for now: on 18 May 2026 the Government suspended new CRICOS registration and course applications to ASQA for 12 months, with public providers such as TAFEs and Table A universities exempt. Detail: CRICOS provider international student recruitment Australia.

Can allied health clinics in Australia use patient testimonials in marketing?

Not about clinical care. Ahpra’s advertising guidelines, which a court or tribunal may consider when hearing advertising offences under section 133 of the Health Practitioner Regulation National Law, treat a testimonial as a recommendation or positive statement about the “clinical aspects” of a regulated health service – a symptom, a diagnosis or treatment, or an outcome – and those are not permitted in advertising for any regulated health service, from physio and psychology to dentistry. Comments about customer service or communication style that do not mention clinical aspects are not testimonials for this purpose. Responsibility sits with whoever controls the advertising: advertisers are not responsible for removing testimonials on platforms they do not control, but the guidelines warn that engaging with reviews on a third-party site may itself count as using a testimonial. Recall, follow-up and appointment-reminder messages to existing patients are treated differently: the OAIC’s guidance on communications with patients says it is generally reasonable to imply consent to recalls, follow-up after screening and tests, and reminders to attend a scheduled appointment, while direct marketing is assessed separately under APP 7. Detail: lead generation for allied health clinics in Australia.

What rules govern how Australian universities market courses and admissions?

Universities registered under the Higher Education Standards Framework (Threshold Standards) 2021 answer to TEQSA on Domain 7 of the Framework. Section 7.1 requires a provider’s representations about itself and its courses, “whether directly or through other parties”, to be “accurate, ethical and not misleading”, and Section 7.2 requires enough public information for students to make informed choices, including the particular needs of international students. TEQSA says its main interest under 7.1 is marketing materials and claims about career outcomes, and it expects providers to control how agents and third-party websites represent their courses. The standards are moving: amendments published on 13 July 2026 add standards in Domain 7, among others, applying from 1 January 2027 (governance changes at providers other than Australian universities apply from 1 July 2027). Oversight is also shifting: the Australian Tertiary Education Commission was “Established under the Universities Accord (Australian Tertiary Education Commission) Act 2026 (ATEC Act), which came into effect on 29 April 2026”, and lists as a key priority supporting the Government’s goal for 80% of working-aged Australians to complete a tertiary education qualification by 2050. Detail: lead generation for universities and higher education in Australia.

AI for corporate and enterprise buyers

Why do most enterprise AI pilots never make it into production?

Because the tools don’t learn the business – not because the models are weak. The GenAI Divide: State of AI in Business 2025, written by Aditya Challapally, Chris Pease, Ramesh Raskar and Pradyumna Chari with Project NANDA out of MIT (July 2025), rests on “a systematic review of over 300 publicly disclosed AI initiatives”, structured interviews with representatives from 52 organizations and survey responses from 153 senior leaders. Its headline finding is that despite “$30–40 billion in enterprise investment into GenAI”, “95% of organizations are getting zero return”. The funnel underneath it is the part you can act on: “Sixty percent of organizations evaluated such tools, but only 20 percent reached pilot stage and just 5 percent reached production”, and the report attributes the losses to “brittle workflows, lack of contextual learning, and misalignment with day-to-day operations” rather than to model quality or regulation. Two consequences for a buyer. Scope the pilot to one workflow that already has a P&L owner and a number attached to it, so “did it work?” has an answer. And make “does this system improve on our data, or reset every session?” a written evaluation criterion rather than a demo question. The report labels itself “Preliminary Findings”, so read 95% as a direction of travel, not a constant.

Should we build our own AI sales system or buy one?

Decide it on where the learning loop has to live, not on licence cost. The same MIT Project NANDA report found that “external partnerships with learning-capable, customized tools reached deployment ~67% of the time, compared to ~33% for internally built tools” – while listing, among its own methodological constraints, “Build vs. buy percentages based on interview responses rather than comprehensive market data”. Treat it as directional, not as a forecast for your business. Building is defensible when the workflow is genuinely proprietary, the data it needs is yours and clean, and you already employ the people who will still be maintaining it in year three. Buying is defensible when the capability is a commodity that someone else improves weekly, and when your real constraint is calendar time rather than engineering budget. The question that usually settles it: who is accountable at 6pm on a Friday when the agent says something wrong to a customer – and is that person on your payroll or your vendor’s? Our own model sits at the far end of “buy”: you are buying booked, qualified conversations rather than software, which is why we carry the delivery risk. How our pricing works.

Where does AI actually produce a measurable efficiency gain?

In high-volume repetitive work where a good answer already exists somewhere in the organisation. The cleanest published measurement is Generative AI at Work (Erik Brynjolfsson, Danielle Li and Lindsey R. Raymond, NBER working paper 31161, April 2023, revised November 2023), which studied the staggered rollout of a generative AI assistant “using data from 5,179 customer support agents” and found that access “increases productivity, as measured by issues resolved per hour, by 14% on average, including a 34% improvement for novice and low-skilled workers but with minimal impact on experienced and highly skilled workers”. Two things follow for a sales or service team. The gain lands on your newest people, so the business case is faster ramp and fewer poor early conversations – not replacing your best closer. And the mechanism is diffusion: the authors offer “suggestive evidence” that the model “disseminates the best practices of more able workers”, which only works if your best practice is captured somewhere a system can read. A later version was published in the Quarterly Journal of Economics in 2025; the numbers quoted here are the working paper’s.

What AI governance actually applies to an Australian business using AI in sales and marketing?

Less than the headlines imply, and more than most teams have written down. Privacy law already applies: the OAIC’s guidance on privacy and the use of commercially available AI products (21 October 2024, updated 17 January 2025) states that “The Privacy Act applies to all uses of AI involving personal information”, and that “When looking to adopt a commercially available product, businesses should conduct due diligence to ensure the product is suitable to its intended uses”. The government’s practical framework has also moved. On 21 October 2025 the National AI Centre published Guidance for AI Adoption, which sets out six essential practices – decide who is accountable, understand impacts and plan accordingly, measure and manage risks, share essential information, test and monitor, and maintain human control. The Department of Industry, Science and Resources describes it on its Voluntary AI Safety Standard page as “updated and simplified guidance for industry” that “evolves the Voluntary AI Safety Standard” and its 10 guardrails, so an AI policy that still cites only the 2024 guardrails is a version behind. None of that displaces the rules that already governed the channel rather than the technology: consent, identification and unsubscribe under the Spam Act, the Do Not Call Register for outbound calling, and whatever your own regulator says about advertising claims. General information only, not legal advice.

What should we ask an AI vendor before signing?

Six questions, and most of them are about accountability rather than features. Where does our data go, is it used to train anything, and can we get it back? Who owns each part of the system – the National AI Centre’s Guidance for AI Adoption is explicit that when “AI systems involve multiple parties (vendors, developers, integrators etc), make it clear who is responsible for each part of the AI supply chain”. What happens when the agent gets something wrong in a live conversation, who finds out, and who is on the hook? Does the system improve on our feedback, or reset every session? What does success look like in writing, measured on our numbers, with a named review point? And how do we leave – with what notice, and what comes with us? The OAIC puts the first duty plainly: “When looking to adopt a commercially available product, businesses should conduct due diligence to ensure the product is suitable to its intended uses” (OAIC AI guidance). And the reason to be blunt in the room is in the MIT Project NANDA data: of the enterprises that evaluated AI tools, “only 20 percent reached pilot stage and just 5 percent reached production”. How we answer these about ourselves: our methodology and how we charge.

Can an AI agent call our existing customers without asking for fresh consent?

It depends on where the phone rings, and of the US, Australia, the UK and the EU, only Australia lets the customer relationship do the work. In the US, the FCC’s Declaratory Ruling FCC 24-17 confirmed that the TCPA’s restrictions on “artificial or prerecorded voice” “encompass current AI technologies that generate human voices”, and 47 CFR § 64.1200(a)(2)–(3) requires prior express written consent for artificial-voice telemarketing to mobiles and residential lines. The established-business-relationship exemption (18 months after a purchase, three months after an enquiry) lets a live caller reach an existing customer on the Do Not Call Registry; the FCC eliminated that exemption for prerecorded telemarketing calls to residential lines in 2012. In Australia, the Do Not Call Register’s industry guidance allows telemarketing calls where consent “can be reasonably inferred from their conduct and business or other relationships” – a bank’s credit-card holder can expect calls about that bank’s home loans – and that consent “ends immediately” when the customer says no. In the UK, the ICO says a number on the Telephone Preference Service needs specific consent “even if they are an existing customer”, and that for automated marketing calls “General consent for marketing, or even consent for live calls, is not enough”. The ICO does not say whether a voice generated live by AI counts as automated, so the cautious reading treats it as automated. In the EU, Article 13(1) of the ePrivacy Directive, as amended in 2009, allows “automated calling and communication systems without human intervention” for direct marketing only with prior consent, and the existing-customer exception in Article 13(2) covers electronic mail only. Service calls such as appointment reminders sit under looser rules until they start to sell. General information, not legal advice. The jurisdiction-by-jurisdiction detail, including which regimes require the call to say it is an AI: what the rules actually say about AI calls to your customers.

How much time does AI actually save employees?

For regular users, about two hours a week on self-reported US survey data, and much less averaged across a whole workforce. The St. Louis Fed’s nationally representative survey by Alexander Bick, Adam Blandin and David Deming found “an average time savings of 5.4% of work hours in the November 2024 survey” among generative AI users, which for someone working 40 hours “implies a time savings of 2.2 hours per week”; across all workers, including non-users, “workers saved 1.4% of total hours” (The Impact of Generative AI on Work Productivity, February 2025). Those are self-reported figures. Where the time goes matters more than the headline. In Anders Humlum and Emilie Vestergaard’s study linking Danish adoption surveys to administrative records (NBER working paper 33777, revised March 2026), “most chatbot users (85%) report reallocating time savings from AI chatbots to other job tasks” and “fewer than 10% report taking additional breaks or leisure”, and the authors estimate “precise null effects on earnings and recorded hours”. So budget against an output count that should rise, not against hours on paper. Five studies side by side, with their samples and a worked team calculation: how much time AI actually saves employees.

How long does AI take to show a return on investment?

Longer than most budgets assume. Deloitte’s AI ROI: the paradox of rising investment and elusive returns (22 October 2025), based on a survey of 1,854 senior executives in 14 European and Middle Eastern countries, found that “most respondents reported achieving satisfactory ROI on a typical AI use case within two to four years”, against “the typical payback period of seven to 12 months expected for technology investments”. “Only six percent reported payback in under a year, and even among the most successful projects, just 13 percent saw returns within 12 months.” The spread by use case is wide because time to ROI is three waits added together: the build to production, the ramp to a stable leading indicator, and then your own sales cycle or contract notice period – and no vendor can shorten the last one. Use cases that end in a booked appointment on leads you already hold sit at the short end; employee copilots and custom models sit at the long end. Payback worked month by month, by use case: how long AI takes to show ROI.

How long do AI vendors keep our customers’ call data?

Each company in the chain keeps its own copy on its own clock, so the honest answer is “as long as the slowest one”. At the model layer, OpenAI says API abuse monitoring logs, which “may contain certain customer content, such as prompts and responses”, are “retained for up to 30 days, unless longer retention is required by law”, and approved customers can apply for Zero Data Retention. Anthropic says it deletes API inputs and outputs “within 30 days of receipt or generation”, keeps them “for up to 2 years” if they are flagged as violating its Usage Policy, and “in all cases” may retain data as required by law. At the telephony layer, Twilio keeps a deleted recording’s metadata for 40 days. In a default setup, the copies with no end date are usually the ones you control: a recording nobody deletes, and the transcript and summary written into your CRM. Those summaries count under Australian privacy law too: the OAIC says privacy obligations apply to “the output data generated by AI (where it contains personal information)”. The full seven-store map, with each published default and who can compel it: how long AI vendors keep your customers’ conversations.

Last updated: 1 November 2026. Answers on this page are refreshed as the underlying studies and regulations change.