A

Journal · 01 · Measurement

Cost per lead is a vanity metric

The cheapest number in your dashboard is usually the most misleading one — and the reason so many clinics can describe their marketing in perfect detail while having no idea whether it works.

Published — August 2026 Reading — 6 minutes Topic — Measurement, funnel

A twelve-dollar lead who never walks through the door is more expensive than a forty-five dollar lead who books and buys. Everyone agrees with this sentence and almost nobody runs their account as though they believe it.

Why the number is so seductive

Cost per lead is available immediately, it sits at the top of every dashboard, and it moves. You can change it this afternoon. Broaden the audience, loosen the form, shift budget to a cheaper placement, and the number goes down before dinner. It feels like progress because something visibly improved.

Every other number in the chain takes weeks and requires talking to someone in operations.

What the platform is actually doing

If your campaign objective is set to leads and your conversion event is a form submission, the platform has been given one instruction: find people likely to submit a form. It will do that superbly. It will find the people most inclined to tap a button, which is not the same population as people inclined to spend two thousand dollars on a treatment package.

The part worth sitting with

Bad lead quality is almost never a targeting failure. It is the account doing precisely what it was told, toward a finish line nobody moved.

The chain the number hides

Between an impression and money there are eight steps, and cost per lead reports on three of them.

A clinic can halve its cost per lead and lose money doing it, if the leads it gained convert at a third of the rate of the ones it replaced. This happens constantly and is nearly invisible, because the metric that got worse lives in a different system from the metric that got better.

What to look at instead

Cost per arrived customer. Cost per sale. Those are the two numbers that decide whether a campaign is worth running, and both require lead source to survive the journey into whatever system holds your bookings.

If that connection doesn't exist, building it is more valuable than any optimisation you could make this quarter — because until it does, every budget decision you make is a guess wearing a spreadsheet.

One diagnostic worth running this week

Take your two best-performing ads by cost per lead. Then find out, for each, how many of those leads actually arrived. In my experience two ads at an identical cost per lead routinely produce completely different downstream conversion, and the gap between them is the most useful thing in the account.

You can optimise toward cheap, or you can optimise toward revenue. The platform will do whichever you ask, and it will never tell you that you asked for the wrong one.

None of this means cost per lead is useless. It's a fine input. It is a terrible objective.

← All writing Next piece →

Bring me the number you cannot explain. I'll tell you where it went.

An hour of looking is usually enough to find the leak — and roughly what it has been costing you.

Write to me