“My leads suddenly dried up” — is it noise or a trend?

Most sudden lead droughts are arithmetic, not emergencies. At 10 leads a week, a fortnight down 40% happens about once a year with nothing changed at all; at 100 a week, the same warning sign is a 12% drop. Check your volume band before you change a single campaign.

• Tonight, free: submit your own enquiry form from a phone on mobile data, check the notification inbox and its spam folder, check the ad account for a declined card or policy notice, and count raw CRM records rather than a dashboard number.
• The noise band: at 10 leads a week, two silent business days in a row and a fortnight as low as 12 leads are both normal. At 50 a week, a single silent day is already a 1-in-85 event.
• The trend signal: six consecutive weeks below your median week. That happens by chance in about one year in three; eight consecutive weeks happens in fewer than one year in ten.
• Change one thing at a time. Moving budget, offer and copy together destroys the only measurement that could have told you what happened.

“Our leads dried up” — is it noise, or did something break?

“My leads suddenly dried up” describes three different situations that need opposite responses. A mechanical break — a form, a tracking tag, a card declining, a mail rule — removes leads instantly and completely, so the daily count has a visible edge: a flat zero starting on a specific day. A trend removes them gradually, so the count sags over weeks without ever hitting zero. Normal variation removes nothing; it just clumps the same leads unevenly.

The distinguishing test costs nothing: plot daily lead counts for the last 60 days and look for an edge. A flat zero starting on a nameable day is broken plumbing, not a market. Markets sag; plumbing stops.

How many quiet days are normal at my lead volume?

This is the part almost everybody gets wrong: two owners can look at the same percentage drop and only one of them is right to worry. The fewer leads you normally get, the wider the normal band, because small counts are dominated by chance. At 10 a week, a fortnight down 40% is ordinary noise; at 100 a week, a fortnight down 12% is the same event.

Method, so you can rebuild it. Leads modelled as Poisson arrivals at a constant rate across five trading days a week and a 260-day year; the silent-run column is the longest run of zero-lead days with at least a 50% chance of occurring in one year, from 20,000 simulated years. Computed by LeadsNow, September 2026. The caveat matters: real lead flow is more variable than Poisson, not less — campaign bursts, weekday effects and referral clumps widen the spread. Treat the table as a floor: inside the band it is definitely noise, outside it, it still might be.

The six-week rule: counting weeks below my median

The volume table needs a stable average. If you do not have one, there is a simpler test that needs no distribution at all: take your median week from the last eight weeks, then count consecutive weeks below it. By definition half your weeks are below your median, so a run of quiet weeks is a run of coin flips, and coin flips clump.

Method: 60,000 simulated 52-week years of fair coin flips, LeadsNow, September 2026. The rule worth writing on the wall is the one that keeps people from burning budget in week three: two quiet weeks is noise at any volume, six weeks below your median happens by chance in one year in three, and eight weeks is a trend.

What to do in the next 24 hours

None of this involves buying anything, and none of it should be skipped for being too obvious. The obvious one is usually the broken one.

1. Submit your own enquiry form from a phone, on mobile data, not on the office wifi and not logged in. Time how long until the record appears in the CRM and until a human replies. A form that posts to a dead address is the single most common cause of a genuinely sudden stop.
2. Check the inbox that receives notifications — its spam folder, any forwarding rule changed this month, and whether a staff mailbox was deactivated when someone left.
3. Check billing and account status on every paid channel. A declined card pauses delivery silently. So does a policy disapproval on an ad set that used to carry most of your volume.
4. Count raw records, not dashboard numbers. Open the CRM and count rows created in the last 14 days. A tracking tag can break without a single lead being lost, and it produces exactly the same panic.
5. Write down the last eight weekly counts and the median. That baseline is what every decision after tonight depends on, and you cannot reconstruct it later.
6. Change nothing else tonight. If you move budget, offer and copy at once, you will never know which one was the problem or which one was the fix.

What to do in the next 7 days

Split the count by source first. One source falling to zero while the rest hold is a channel failure; every source drifting down together is demand, seasonality or a tracking change. Then work the causes with public evidence behind them:

• Email. If outbound or nurture email is a source, check deliverability before anything else. Google requires bulk senders to keep the spam rate reported in Postmaster Tools below 0.3%, with SPF, DKIM and DMARC in place; crossing that line moves mail to spam without a bounce, so nothing looks broken.
• Organic search. Check whether a ranking update overlaps your drop, and remember the rollout windows are long — the May 2026 core update ran 11 days 21 hours and the March 2026 core update 12 days 4 hours. A dip that starts and ends inside 48 hours is not an algorithm update.
• Seasonality. Compare the same weeks last year, not last month. Month-on-month comparisons across a holiday period manufacture a crisis every year.
• Leads arriving but not reached. Volume can look flat while usable volume collapses because response time slipped. Median minutes from enquiry to first human contact, and the share never reached at all, are the two numbers to pull; our pages on why contact rate falls and the speed-to-lead response window cover the tests.
• Work the warm list before buying cold volume. It is the cheapest inventory you own. Our own record on dormant databases is 4.4% average conversion with an 8.9% peak, from the Colliers-era campaigns on our database reactivation results page, with the mechanics in how to run a reactivation campaign — our number on our data, not an industry benchmark.

When lead flow dropped suddenly and the cause is not marketing

Some of these drops are not a marketing problem, and no amount of budget touches them.

• An account suspension or policy action. The fix is the platform’s appeal process, not a new campaign. Spending elsewhere while the appeal sits is how a two-week problem becomes a two-month one.
• A referral partner or contract that ended. That is a commercial conversation, and sometimes a legal one. A lawyer or your accountant is the right first call, not an agency.
• A genuine sector contraction. If published data for your industry is down and you are down by less, you do not have a lead problem — you have a market, and the decision is about the cost base. That is an accountant’s conversation.
• A regulatory change. Check the regulator’s own site for your industry before you accept anyone’s interpretation of it, including ours.

The weekly board that stops this happening again

Four numbers, one screen, updated every Monday: leads by source for the week, the median of the last eight weeks, a counter of consecutive weeks below that median, and median minutes from enquiry to first contact. Ten minutes in a spreadsheet. The board is what turns “it feels quiet” into a number you can act on or ignore.

What it honestly costs: somebody has to do it every week, and it is worthless until it has eight weeks of history, so starting it now does not rescue this drought. It also lies the moment sources stop being tagged consistently — below about 20 leads a week you can reconcile that by hand; above that, tagging has to be automatic or the board reports a channel failure that is really a labelling failure. If it eventually shows leads arriving fine and nobody working them the same day, that is a response-capacity problem, not a volume one — the problem AI appointment setting addresses, and what a pay-per-result engagement is priced against.

Questions people ask when their leads stop

Why did my leads stop suddenly, overnight?

An overnight stop with a clean edge is almost always mechanical: a form posting to a dead mailbox, a declined card pausing ad delivery, a changed mail-forwarding rule, or an email domain that crossed a deliverability threshold. If email is a source, check that first — Google’s sender guidelines require bulk senders of more than 5,000 messages a day to Gmail accounts to keep the spam rate reported in Postmaster Tools below 0.3% and to have SPF, DKIM and DMARC in place, and mail that crosses that line is filtered rather than bounced, so nothing in your own system looks broken.

How many quiet days should I wait before I worry?

It depends entirely on your normal volume. At 10 leads a week, two silent business days in a row will happen in most years and a fortnight as low as 12 leads is inside normal variation. At 50 a week, a single completely silent day is roughly a 1-in-85 event and is worth investigating the same day. The threshold table above gives the bands for 2 to 100 leads a week.

Could a Google update have caused my lead flow to drop?

It can, but the timing has to fit. Google publishes every ranking update with its start time and duration on the Google Search Status Dashboard: the May 2026 core update ran 11 days 21 hours, the March 2026 core update 12 days 4 hours, and the August 2026 spam update 2 days 16 hours. If your drop began on a day with no update in progress, or began and ended within 48 hours, look at your own plumbing before you blame the algorithm.

My enquiries stopped but my traffic is unchanged — what does that mean?

Steady traffic with no enquiries points at the form or the tracking rather than the market, in that order. Submit the form yourself on a phone, then count rows in the CRM instead of trusting a dashboard. A broken tracking tag produces zero recorded leads while real ones keep arriving, and a broken form produces real silence — the CRM row count tells you which one you have in under five minutes.

Should I increase ad spend when lead flow drops suddenly?

Not before you know whether the drop is real. Adding budget during normal variation is how a noise event becomes an expensive one, because the recovery that was going to happen anyway gets attributed to the spend and the higher budget stays. Run the two tests first: does the daily count have an edge, and how many consecutive weeks are below your median?