How Agencies Add AI Appointment Setting to Their Offer

An agency adds AI appointment setting in one of three ways (build it on AI voice tools, white-label a fulfilment partner, or refer the client), and on a worked example of 40 booked appointments a month at $250 each, white-label leaves about $3,100 of monthly margin while an in-house build only earns more above 25 appointments a month.

At a glance:

• Three models: build in-house on AI voice platforms, white-label a done-for-you partner, or refer and take a fee.
• Margin on the worked example (40 booked appointments, $250 each): in-house $5,350, white-label $3,100, referral $850 a month, before the in-house build cost of $6,000.
• The Half-Volume Test: at 20 appointments, in-house drops to $350 and white-label to $1,100.
• AI voice running cost: Retell AI lists $0.07–$0.31 a minute on its pricing page (read 27 October 2026).

Rules of thumb

• At 40 appointments, white-label earns about $77.50 per appointment: $250 price less $150 fulfilment is $100, less account time of $22.50 per appointment (12 hours at $75 over 40 appointments) (worked example below).
• An in-house build beats white-label only above 25 appointments a month for one client on these inputs ($3,750 of extra fixed cost divided by the $150 per-appointment saving).
• Payback on a $6,000 in-house build is about 1.1 months at 40 appointments and about 17 months at 20 (worked example below).
• US AI voice calls need prior express consent (FCC 24-17); in Australia, the ACMA says “ultimately, the business is responsible” even when marketing is outsourced (ACMA).

What margin does an agency make reselling AI-booked appointments?

An agency reselling 40 AI-booked appointments a month at $250 each keeps about $5,350 building in-house, $3,100 white-labelling or $850 referring, on the stated assumptions below. No credible public benchmark exists for agency resale prices or wholesale fulfilment rates, so every input is a labelled assumption, not a market rate, LeadsNow’s terms or any provider’s price list.

In-house keeps the most per appointment at full volume. The catch is the 50 hours: they assume someone in the agency can test scripts, review recordings, tune qualification and keep consent records every month. Figures are illustrative and work the same in USD or AUD.

Frequently asked questions

Can an agency white-label AI appointment setting under its own brand?

Yes. In a white-label arrangement the agency contracts with the client and a fulfilment partner runs the calling, texting and booking behind the agency’s brand. See white label lead generation for agencies and the best white-label lead generation providers in the USA.

Do AI voice calls need consent in the US?

Yes. The FCC’s February 2024 declaratory ruling (FCC 24-17) confirmed that AI-generated voices count as an “artificial or prerecorded voice” under the TCPA, so those calls need the prior express consent of the called party (FCC 24-17, released 8 February 2024). For telemarketing, the consent must be written and must authorise the seller, which is the agency’s client.

Is an Australian agency responsible if its fulfilment partner sends the SMS?

The business whose marketing it is stays responsible. The ACMA says businesses “cannot outsource their obligations under the spam and telemarketing laws” and that “ultimately, the business is responsible” (ACMA, common issues and mistakes). Put who holds the consent record into the contract between agency, client and partner.

How many booked appointments a month does an in-house AI build need to beat white-label?

On the assumptions in the margin table, an in-house build earns more than white-label above 25 booked appointments a month for one client: the build carries $4,650 of fixed monthly cost against white-label’s $100 margin per appointment less $900 of account time.

Should an agency charge per appointment or a retainer for AI appointment setting?

Per-appointment pricing ties the agency’s revenue to the client’s result, which suits a white-label partner who also bills per appointment, because cost and revenue rise and fall together. A retainer suits an in-house build with fixed costs. The trade-offs are compared in pay-per-result vs retainer for a marketing agency.

The Half-Volume Test: which model survives a slow month?

The Half-Volume Test reruns the margin at half the booked appointments, and at 20 a month white-label keeps $1,100 while an in-house build keeps $350. Judge a model on a weak month, not the launch forecast.

The formulas, so you can rerun them: in-house margin = $250 x appointments minus $4,650; white-label margin = $100 x appointments minus $900; referral margin = $25 x appointments minus $150. The in-house formula assumes the same leads are called whether or not they book, so its $4,650 cost stays fixed (an assumption). The rule: sell the model that still clears your hourly cost at half volume.

Which three models can an agency use to add AI appointment setting?

An agency can add AI appointment setting by building it on AI voice and SMS platforms, white-labelling a done-for-you partner, or referring the client to a specialist for a fee. Each model changes who contracts with the client, whose brand is on the calls and who takes the blame in a bad month.

Contract structure and failure modes for each are set out in how white label lead generation for agencies actually works; this page adds the margin maths and launch plan.

How long does it take an agency to add AI appointment setting?

An agency can plan on roughly two to six weeks to sell its first AI-booked appointment through a partner, and longer to build in-house, with the consent and data audit the stage most likely to slip. The durations below are planning assumptions, not measured averages.

The audit slips because consent wording lives on the client’s forms and old lists, which only the client’s developer or CRM owner can change. Start it in week one.

What compliance does an agency inherit when it resells AI calls and SMS?

An agency reselling AI calls and SMS inherits the consent rules of every market it calls into: in the US the TCPA treats AI voices as artificial voices, and in Australia the business being marketed stays responsible. General information, not legal advice.

• US, AI voice consent. The FCC’s declaratory ruling FCC 24-17 (adopted 2 February 2024, released 8 February 2024) confirms that the TCPA’s restrictions on “artificial or prerecorded voice” cover AI technologies that generate human voices, so such calls need the prior express consent of the called party (FCC 24-17).
• US, who the consent names. For telemarketing, prior express written consent is a signed written agreement that “clearly authorizes the seller” and names the phone number (47 CFR 64.1200(f)(9)). The seller is the agency’s client, so the client’s form wording has to cover the calls the agency resells.
• US, exposure. The TCPA lets a person sue for actual loss or $500 per violation, whichever is greater, and a court may increase the award for wilful or knowing violations (47 U.S.C. 227(b)(3)).
• Australia, SMS and email. Under the Spam Act 2003 you need consent before sending marketing messages, must identify the sender and must make unsubscribing easy; the ACMA adds that consent is still required “even if someone else is sending out your marketing messages for you” (ACMA, avoid sending spam).
• Australia, calls. Once a number has been on the Do Not Call Register for 30 days, telemarketers can only call it with consent or an exemption; a business-only number cannot be registered (ACMA, Do Not Call Register).

The practical step is a three-party clause: the client warrants its consent records, the partner follows calling and messaging rules, and the agency keeps a copy of both.

How should an agency price and package AI appointment setting?

An agency should price AI appointment setting in the same unit its cost moves in: per appointment when a partner bills per appointment, and a retainer when an in-house build carries fixed costs. A per-appointment price funding a fixed in-house cost is where the Half-Volume Test bites.

Three packaging choices matter more than the headline price: define a qualified booked call in writing, report booked and shown separately, and decide whether the offer serves existing clients or wins new ones, the pipeline question covered in how marketing agencies get their own clients. Churn drivers are in why agency client churn is so high.

For agencies that want booked calls for their own pipeline rather than a resale line, the LeadsNow page for marketing and creative agencies covers that, and AI appointment setting explains how the calling, SMS and DM follow-up works. LeadsNow, a Melbourne-based agency, has delivered 50,769+ AI-booked appointments and 1M+ leads.

Sources

1. FCC 24-17 Declaratory Ruling (released 8 Feb 2024)
2. 47 CFR 64.1200 (Cornell LII)
3. 47 U.S.C. 227 (Cornell LII)
4. ACMA: Avoid sending spam
5. ACMA: common issues and mistakes
6. ACMA: Do Not Call Register
7. Retell AI pricing (read 27 Oct 2026)