The run ends and the number looks great: 100 sign-ups for $400. That is $4.00 a lead. It is easy to nod and call it a price. It is a price for an email address.
What you measured
A lead is a person who read your page and left a contact. We count it, we know what the clicks cost, and the division is exact. Cost per lead is a measurement.
What you are guessing
A customer is a person who pays. Nobody in the run paid, so the price of one depends on a rate you have not seen: how many leads would. At $4.00 a lead, here is what the same run implies:
| Leads who would pay | Cost per customer |
|---|---|
| 10% | $40 |
| 5% | $80 |
| 2% | $200 |
The spread is the point. The same $400 supports a comfortable business or a ruinous one, and the run cannot tell them apart. Cost per customer is an estimate, and it is only as good as the share you assume.
Why leads are a weaker signal than they feel
A sign-up costs a visitor ten seconds and nothing else. Paying costs money, trust and a decision. The two behave differently, so the second cannot be read off the first. Treat leads as interest and mark any customer cost as a projection.
What to do with the number
- Work in a range. Pick a cautious share and an optimistic one, and see whether the idea survives the cautious one.
- Price it early. A page that names what the product costs filters for people who would pay. Its leads are fewer and worth more.
- Ask the leads. Write to the first 10 and ask for a call or a pre-order. Their answers are the first real number you will have.
Every report we write keeps the two costs on separate lines. The measured one is a fact. The estimated one is labelled, so you do not spend money on a guess that looks like a fact.