Sample Readout
What a first campaign measurement readout looks like. The business and every number on this page are made up.
- Status
- A sample. The business and every number are made up to show the format. Real results are in the case studies.
- Business
- A home-services company with website form leads, a CRM, and Google and Meta ads.
- Question
- Which campaigns bring in paying customers, and where should next quarter's budget go?
- Period
- One quarter of form leads that came from an ad.
This is the kind of report a first campaign measurement project ends with. The business is invented and so are the numbers, so you can see the format without seeing anyone's data. Real results, and how each one was measured, are in the case studies.
The short version
- Google search brought in three of every four customers the ads produced, though about half of those searched for the business by name and may have come anyway.
- Meta ads brought the cheapest leads but the most expensive customers: each Meta customer cost about twice as much as a Google one.
- The platforms claimed 242 sales between them. The CRM shows 160.
- Next step: move about $4,000 a quarter from Meta new-audience ads to Google services search, and test the cut in two service areas before making it permanent.
Which campaigns brought in paying customers
Every number here counts closed sales in the CRM, not leads or the platforms' own reports.
| Campaign | Ad spend | Leads | Closed sales | Revenue, first 90 days | Ad spend per $1 of revenue |
|---|---|---|---|---|---|
| Google search, brand name | $6,000 | 310 | 62 | $93,000 | $0.06 |
| Google search, services | $18,000 | 420 | 58 | $98,600 | $0.18 |
| Meta, new audiences | $12,000 | 610 | 21 | $27,300 | $0.44 |
| Meta, retargeting | $4,000 | 140 | 19 | $26,600 | $0.15 |
| Total | $40,000 | 1,480 | 160 | $245,500 | $0.16 |
Meta new-audience ads brought in the most leads and the fewest sales. About 3 of every 100 of those leads became a customer, against about 14 of every 100 from Google services search.
What each new customer cost, by channel
| Channel | Ad spend | Cost per lead | Customers | Cost per customer |
|---|---|---|---|---|
| $24,000 | $33 | 120 | $200 | |
| Meta | $16,000 | $21 | 40 | $400 |
A Meta lead cost about a third less than a Google lead, but a Meta customer cost twice as much. Judged on leads alone, Meta looks like the better buy.
Why the platforms claim more sales than you made
Google Ads claimed 151 sales and Meta claimed 91, a total of 242. The CRM traces 160 closed sales to an ad. The difference has two parts:
- 41 customers clicked both a Google ad and a Meta ad before filling in the form, and both platforms counted them.
- 41 of Meta's claims are people who saw an ad without clicking it. There is no lead to check those against.
This report counts each sale once, for the last ad clicked before the form.
Where prospects drop off
| Step | Leads | Share of all ad leads |
|---|---|---|
| Filled in a form after an ad | 1,480 | 100% |
| Reached by phone or email | 1,184 | 80% |
| Got an estimate | 592 | 40% |
| Closed | 160 | 11% |
The biggest loss is between first contact and the estimate. It is worst for Meta new-audience leads: only 92 of 610 got an estimate (15%), against 500 of 870 (57%) for every other campaign.
What to change next, and how we'll check it worked
- Move about $4,000 a quarter from Meta new-audience ads to Google services search. At this quarter's cost per customer, $4,000 buys about 13 customers in services search and about 7 from Meta new audiences. Search volume may not stretch that far, so watch the cost per customer as the budget rises.
- Keep Meta retargeting at its current budget. It closes about as often as services search, at a lower cost per customer.
- Find out why leads stall before the estimate. For two weeks, the sales team records a reason whenever a lead goes cold.
The check. Before cutting Meta new-audience ads everywhere, switch them off in two of the five service areas for eight weeks. Compare closed sales from every source in those areas with the other three, against the same weeks last year. We agree on the decision rule before the test starts, so the result can't be bent to fit a story.
What the data can't show yet
- Phone calls. About a fifth of inquiries arrive by phone with no campaign tag, so campaigns that make people call look weaker than they are. Call-tracking numbers would fix this.
- Whether brand search customers would have come anyway. Brand search looks cheapest, but these people already searched for the business by name. A later area test for brand search is the way to find out.
- Revenue after 90 days. Repeat jobs and referrals aren't counted yet, so customer lifetime value waits for a later readout.
- Meta's view-through claims. There is no lead to check them against.
- Small numbers. Meta retargeting's 19 sales could come out anywhere from about 11 to 30 next quarter by chance alone, so its cost per customer is a rough figure.
How this was measured
- Each website form saves the Google or Meta click ID and the campaign tag in hidden fields. The CRM keeps them on the lead, so a closed sale joins back to the ad that brought the lead in.
- Each sale counts once, for the last ad clicked before the form. The platforms' own claims are shown beside it, not added in.
- Revenue is the first invoice within 90 days of the lead.
- The period is one quarter, and it covers every form lead that came from an ad.
A real readout uses your definitions, agreed at the start: what counts as a lead, a sale, and revenue. I walk through it with the people who will act on it.