There is an argument that happens in almost every lead-generation business, usually in a Monday meeting, usually unresolved.
Sales says the leads are rubbish. Marketing points at cost per lead, which is down, and lead volume, which is up, and says sales is not working them properly. Both sides have numbers. Neither number settles it, so the meeting ends with a decision to generate more leads, which makes the problem worse.
The problem
An advertising platform optimises toward whatever you tell it to count. Tell it to count form submissions and it will find you the people most likely to submit a form — which is not the same population as the people most likely to buy. It is frequently the opposite population.
So the loop goes like this. Cost per lead falls. Volume rises. Marketing reports a win. Close rate quietly drops, because the incoming mix has shifted toward people who fill in forms easily. Revenue stays flat or falls. Nobody connects the two, because the two numbers live in different systems and are reported by different people.
The tell is specific: volume up, cost per lead down, revenue flat. If that is your last two quarters, this article is about you.
Diagnose it in an afternoon
Three numbers. You probably have all of them already, in different places.
Qualified lead rate. Of last quarter's leads, what share matched your actual buyer — right company size, right budget, right authority, real need? Take a sample of a hundred if you have no tagging. Under 30% means the account is buying the wrong people, not too few of them.
Close rate by source. Split it by channel and campaign. In almost every account I have run this on, one source has a close rate several times another, and budget is allocated as though they are equivalent.
Cost per qualified lead. Cost per lead divided by qualified rate. This is the first figure that compares channels honestly, and it routinely reverses the ranking that cost per lead produced.
That third number is the one to put in front of the Monday meeting, because it ends the argument without anyone having to be wrong.
Four fixes, cheapest effort first
1. Qualify in the ad and on the form
The fastest change and the least popular, because it reduces lead volume on purpose.
Put the disqualifying information in the ad: price, minimum engagement, who it is for. "From $2,000 a month" in the copy costs you the people who were never going to buy, and you stop paying for their clicks.
On the form, add one question that only a real prospect can answer comfortably — budget band, team size, timeline. Each field costs completions; this one costs you the completions you did not want.
Expect volume to fall by a third and qualified volume to rise. If nobody internally panics at the volume drop, you have not briefed them properly.
2. Feed the outcome back to the platform
This is the fix with the largest effect and the one almost nobody does.
Both Meta and Google accept offline conversion imports: you send back which leads became qualified, and which became customers, with the value. The platform then optimises toward the people who look like buyers instead of the people who look like form-fillers.
It needs your CRM to record the source and keep it through to close, and it needs someone to push that data back on a schedule. It is an afternoon of setup and a weekly export. In my experience it moves qualified lead rate more than any targeting change available to you.
Google's version pairs with enhanced conversions; the measurement groundwork is the same as for e-commerce.
3. Change what the account counts as a conversion
Once qualified-lead data is flowing back, stop optimising for form submissions. Optimise for the qualified event, or for a value-weighted conversion if your deal sizes vary.
One caution: the algorithm needs volume to learn from. If you only get fifteen qualified leads a month, optimising directly for them will starve the account. In that case optimise for a mid-funnel action that correlates with quality — a booked call, a pricing page view followed by a form — and use the qualified data for reporting and exclusions instead.
4. Fix the follow-up before blaming the leads
Response time moves close rates more than channel choice does in most lead-gen businesses. A lead contacted within an hour converts dramatically better than the same lead contacted the next day.
Before spending anything on fixing lead quality, find out how long your average first response takes. If it is measured in days, that is the cheapest fix available and it is not a marketing fix.
What to expect
Volume falls. Cost per lead rises — often sharply, which is why this needs agreeing with whoever reads the report before you start. Qualified lead rate rises, cost per qualified lead falls, and close rate recovers because the mix changed.
Somebody will look at the cost per lead going up and call it a regression. Have the cost per qualified lead ready in the same table, and agree beforehand that it is the number you are managing to.
The same pattern elsewhere
This is the mechanism behind cheap Google Ads leads that never close in high-ticket categories, and it is why LinkedIn's expensive leads are often cheaper per customer. It is one problem wearing three different costumes: the account is optimising toward a proxy, and the proxy has drifted away from the thing you actually sell.
Upstream of all of it, if your messaging reads like every competitor, you pay more per lead on every channel at once before any of this is even in play.
The short version
More leads and worse revenue is not a traffic problem. Measure qualified lead rate and cost per qualified lead, qualify harder in the ad and the form, feed closed-won data back to the platforms, and check your response time before blaming the channel.
The free growth audit includes the qualified-lead analysis and the conversion-action review, which together are usually the first time a lead-gen account has been looked at past the form. Our Google Ads and Meta Ads work runs on these numbers rather than on cost per lead.