Cost per lead is one of the most popular numbers in marketing. It is easy to calculate, easy to compare and easy to celebrate. It is also one of the easiest numbers to improve while making the business worse.
The arithmetic nobody puts on the slide
Imagine two campaigns. The figures are illustrative, but the pattern is one I have seen many times.
- Campaign A produces leads at ₹300 each. One in ten turns out to be genuinely interested.
- Campaign B produces leads at ₹500 each. Four in ten are genuinely interested.
On a CPL report, Campaign A wins comfortably. It is 40% cheaper. But the business does not sell to leads; it sells to interested people. Divide each CPL by its quality rate and the picture flips: Campaign A costs ₹3,000 per interested prospect, Campaign B costs ₹1,250.
The "expensive" campaign is less than half the price of the "cheap" one once you measure the thing the business actually needs.
A lead is not an outcome. It is a cost the sales team has to absorb before an outcome is possible.
Why platforms drift toward cheap leads
Advertising platforms are extraordinarily good at finding what you ask them to find. If the conversion you report is "form submitted", the algorithm will hunt for the people most likely to submit forms — which is not the same as the people most likely to buy.
Over time, an account optimised purely for CPL tends to discover audiences that click everything, fill in every form and answer no phone calls. The CPL graph looks beautiful. The sales team quietly stops trusting marketing.
The hidden costs of a bad lead
A low-quality lead costs more than its media price:
- Sales time. Every call, follow-up message and site visit invitation has a cost.
- Morale. Teams that chase dead leads start ignoring new ones — including the good ones.
- Corrupted learning. The platform learns from the wrong signal and gets better at finding more of the wrong people.
What to measure instead
I am not arguing that CPL is useless. It is a useful input. It is a poor objective. Instead, I want to see a short chain of numbers:
- Cost per qualified lead — CPL divided by the share of leads that meet an agreed quality definition.
- Lead-to-meeting or lead-to-visit rate — evidence that intent survives the first conversation.
- Cost per sale or booking — the number the finance team cares about.
The most important step is the least technical: sit down with the sales team and write down what a qualified lead actually is. Once that definition exists, it can be scored in the CRM and — crucially — fed back to the platforms as the conversion they optimise for.
A better question
The question is never "how cheap can we make a lead?" It is "what is the most we can afford to pay for a customer, and how do we get there with room to spare?"
When teams change the question, cost per lead often rises slightly while cost per customer falls dramatically. That is not a failure of performance marketing. That is performance marketing working the way it should.