Home · Lead quality
Your leads aren't bad. Your signal is.
Almost every "lead quality problem" is one of seven things, and only one is targeting. The underlying fault is the signal gap - the platform never learns which leads were any good, so it keeps finding more of the wrong ones. This is where we look first on every account.
Have us run it on your account, freeStart here
The platform is finding exactly who you asked for
Meta and Google are extraordinarily good at finding more of a defined thing. Tell Meta success is a submitted form and it finds people most likely to submit forms - a real population, and not the one that buys.
Bad lead quality is rarely the algorithm failing. It's the algorithm succeeding at a goal nobody meant to set.
So the fix isn't better targeting. It's redefining success, adding friction where friction filters, and closing the loop. Below, in order of impact.
The lead chain audit
Seven causes, in order of impact
Seven places it usually breaks, in the order they cost you most. We work through all of them on the audit call.
- 01
No outcome data ever reaches the platform
The big one, and it's missing in most accounts. Your CRM knows who closed. The ad platform never finds out, so it keeps optimizing on form submissions. Close that loop and cost per lead usually rises while cost per customer falls.
- 02
The form asks nothing that filters
Name, email, phone is a volume machine. Every qualifying question costs leads and buys back sales hours. Which way that trade should go depends on whether your team can clear the volume they already have.
- 03
The ad promises something the page doesn't
An offer in the ad that the landing page doesn't repeat isn't producing bad leads. Those people are answering a promise correctly — it just wasn't kept.
- 04
Nobody calls fast enough
A lead contacted within minutes behaves like a different person from the same lead contacted tomorrow. Slow follow-up manufactures "bad leads": the only ones still available by then are the least motivated.
- 05
The offer attracts the wrong person
Free attracts everyone, including everyone who intended to pay nothing. A small paid step is the cheapest qualification there is, and it works before your team spends a minute.
- 06
Targeting reaches people you can't serve
The real targeting problem, and rarer than people assume. Geography set at the wrong grain guarantees waste — a catchment is a travel-time question, not a distance one.
- 07
You're measuring the wrong number
If cost per lead is the only number anyone sees, every decision pushes it down — and pushing it down is exactly what breaks quality. The gap between cost per lead and cost per qualified lead is the real diagnostic.
What changes look like
Expect the cost per lead to go up
- 01
Weeks 1–2
Tracking and conversion feedback set up, form qualification added, follow-up timing fixed. Lead volume usually falls. This is intended.
- 02
Weeks 3–4
Connect rate rises first - it's the fastest-moving metric, because it responds to speed and filtering rather than to the algorithm.
- 03
Weeks 5–8
The platform has enough closed-outcome data to shift who it targets. Lead mix changes. Cost per lead settles higher than before.
- 04
Beyond
Cost per customer becomes the operating metric. Scaling decisions get made on it, and the CPL number stops mattering.
If someone promises better quality and a lower cost per lead in the same sentence, ask how - the two usually pull in opposite directions.
Follow-ups
Questions this usually raises
- Our agency says the leads are fine and sales isn't working them.
- Both can be true, and the argument is unresolvable without data. Connect rate and time-to-first-contact settle it: if follow-up is fast and leads still can't be reached or don't qualify, it's the account. If contact takes a day, it's the process. Measure before assigning blame.
- We don't have a CRM. Can we still send conversion data back?
- Yes. A maintained spreadsheet with lead identifier, date and outcome is enough for offline conversion upload on both platforms. It's manual, but it works and it's far better than no signal at all.
- Won't adding form fields destroy our volume?
- It will reduce it, sometimes substantially. That's the mechanism working. The question isn't whether volume drops - it's whether qualified volume drops, and usually it doesn't. Test one field at a time and measure qualified leads, not total leads.
- How long before quality actually improves?
- Filtering and speed changes show up within days. Algorithmic changes from conversion feedback take six to eight weeks, because the platform needs a meaningful volume of closed outcomes before it can act. Anyone promising a quality turnaround in a fortnight is describing filtering only.
- Can you do this without taking over the account?
- The free audit is exactly that - a diagnostic you can hand to your existing team. Plenty of people take the findings and fix it themselves, which is a fine outcome.
Want us to run this on your account?
Book a free 30-minute audit. We'll work through all seven on your actual campaigns and tell you which one or two are costing you the most. Yours to keep, whoever implements it.