Customer acquisition
When a low CPA becomes an expensive customer
Put acquisition cost alongside qualification, activation, and customer value before deciding where the next budget increase belongs.
Two acquisition sources can hit the same campaign goal and create very different businesses. One produces a large number of inexpensive registrations. The other costs more per registration but brings people who activate, buy, or become useful sales conversations. Deciding between them requires following the customer beyond the event that appears in the campaign report.
The first step is to make the comparison fair. Define the action, the costs included, and the period customers have to complete it. Then ask what happens next. A cheaper initial event may still be a good opportunity, but its value needs to survive the rest of the journey.
Name the action behind the number
Write the event beside the CPA: cost per registration, accepted lead, first purchase, or another defined action. A single headline number can hide different goals, repeated events, or a change in how conversions are counted. Keep the definition consistent while comparing sources.
Google Ads defines average CPA using the cost of conversions divided by the number of conversions. The arithmetic is straightforward; selecting a conversion that helps answer the business question is the work that comes first. Google Ads: average CPA definition (opens in a new tab)
An earlier event can help a team learn while sales take time to arrive. Keep checking whether it predicts the later outcome. If the relationship changes, investigate the audience, offer, product experience, and reporting before allowing that early event to justify more spend.
Follow the same group through the funnel
Consider two hypothetical lead campaigns. Both spend $6,000, serve the same market, and have completed the same 30-day evaluation window. Count each inquiry once. The figures below are an illustration, not Digital Edge campaign results.
| Measure | Source A | Source B |
|---|---|---|
| Media spend | $6,000 | $6,000 |
| Submitted leads | 300 | 150 |
| Cost per submitted lead | $20 | $40 |
| Accepted leads | 60 | 90 |
| Cost per accepted lead | $100 | $66.67 |
| Completed sales | 12 | 18 |
| Media cost per sale | $500 | $333.33 |
Source B costs twice as much per submission but produces more accepted leads and sales from the same spend. That gives the team a reason to investigate B further. It does not establish that the result will hold at a larger budget, that every sale has equal value, or that the campaign caused every credited sale.
Include the costs and value that matter
Keep a media-only view for buying decisions and a broader view for business economics. Creative, incentives, agency work, qualification, and sales handling may also affect what a customer costs to acquire. State which items are included rather than comparing a fully loaded figure with someone else's media-only CPA.
On the value side, look at collected revenue, relevant variable costs, cancellations, and refunds over a stated period. For a subscription, distinguish the first payment from renewals. For a lead, a booked appointment and a completed sale answer different questions. Keep projected future value separate from what has already happened.
Explain the difference before assigning blame
A low acceptance rate can reflect several problems: the audience does not fit, the offer attracts the wrong expectation, the form omits a qualification question, or the handoff reaches an overloaded team. Treat the rate as a starting point for investigation.
- Compare customers with the same time to convert and consistent event definitions.
- Review qualification and rejection reasons with the team handling the response.
- Check whether the ad, landing page, and follow-up describe the same offer.
- Look for missing feedback, duplicate records, or changes to the product and reporting.
Keep the underlying counts visible. Three sales from one small group and five from another leave substantial uncertainty. A dramatic percentage difference should not conceal how little evidence the team has.
End the review with a decision
Agree on what would justify continuing, changing, or expanding a campaign. That could mean improving qualification, fixing the handoff, waiting for a cohort to mature, or making a bounded budget increase. A lower reported CPA alone should not decide the next investment.
Attribution and customer quality are useful evidence. They still leave the question of how much business would have happened without the campaign. Where a suitable experiment can be supported, investigate that question separately. Keep the daily report practical while remaining clear about what it can prove.
