The first number is not the final outcome
A campaign can produce more leads while producing fewer customers. Cost per lead answers what it cost to generate a recorded enquiry. It does not tell you whether those enquiries were suitable, contacted, or won.
Follow the same period and lead cohort through four counts: spend, leads, qualified leads, and customers. Calculate cost per lead, cost per qualified lead, and cost per customer. If the later counts are missing, improve the handoff and measurement before declaring the ads good or bad.
Find the point where volume loses value
Suppose a fictional campaign spends $6,000 and records 120 leads: $50 per lead. If only 30 are qualified, cost per qualified lead is $200. If six become customers, this simple cohort shows $1,000 per customer. None of these numbers alone proves profitability; customer value, margin, timing, and attribution still matter.
The useful question is not “Can we make cost per lead lower?” It is “Which transition loses commercial value, and is that caused by targeting, message, offer, response time, sales process, or tracking?” Each cause calls for different work.
Model a change, then test the mechanism
If qualified share rises, a simple model can show the equivalent spend needed to produce the same number of qualified leads at today’s cost per lead. That is a sensitivity calculation, not money already saved. Better qualification may change volume, auction cost, or customer mix.
The Ad Spend Efficiency Explorer makes the arithmetic visible. Use it to frame a practical test: change one audience or message, preserve a comparable baseline, define qualification consistently, and watch both qualified leads and customers—not the headline lead count alone.
Questions worth asking next
- Can we trace one lead cohort from spend to customers?
- Where does qualified volume fall away most sharply?
- What single handoff will we change and measure next?