Performance Marketing
Your monthly ad report has five rows. That's a receipt, not a report.
I’ve seen a fair number of monthly reports that agencies send their clients. Nearly all of them look the same: impressions, clicks, CPC, spend, maybe conversions. Five rows, a table, sometimes a chart. And a closing line along the lines of “good results, let’s keep going”.
A report like that isn’t wrong. The numbers are real. But it doesn’t answer the one question you’re paying for the report to answer: what do I do next month, and why.
The test
Before any list of metrics, one simple test. Read the report and ask yourself:
If I had 30% more budget next month, would this report tell me where to put it?
If it doesn’t, the report is a receipt. It confirms the money was spent. That’s all.
This matters, because the answer isn’t “more columns”. A report with forty metrics and no conclusion is just as useless, only more tiring to read. Every breakdown that goes into a report has to lead to a decision. If it doesn’t, it has no business being there.
What’s usually missing, and the decision it would unlock
The age and gender breakdown. In Ads Manager it’s two clicks away: Breakdown → Age, Gender. If the 25–34 segment brings you customers at half the cost of the 45–54 segment, you have an immediate decision: move the budget. If you don’t know, you pay the same for both. I’ve seen campaigns where half the budget went to an age group that never bought anything.
The placement breakdown. Facebook Feed, Reels, Stories, Audience Network. These are completely different environments with completely different costs. Audience Network in particular produces cheap clicks that mean nothing — people tapping a banner in a game by accident. If the report doesn’t separate placements, you can’t tell whether you’re paying for attention or for accidental taps.
For video: retention, not impressions. A clip with 50,000 impressions sounds good. If 90% of people drop off in the first three seconds, you paid for 50,000 people who saw nothing. Meta gives you the curve: how many made it to 25%, 50%, 75%, 95%, plus the average watch time. The decision it unlocks: if you lose people at second 3, the problem is the hook, not the budget. If you lose them at second 20, the problem is the length. Without the curve, the only possible reaction is “let’s put more money in”.
Frequency. How many times the same person has seen the ad. Above 3–4, on a small audience, it means you’re paying to knock on the same door for the fifth time. Costs go up, results go down, and it looks as if “the ad stopped working”. What actually happened is the audience ran out.
What changed since last month, and what happened next. This is the one most often missing, and the most important. A report without history is a snapshot. You want the film: we changed the copy on the 12th, cost per lead dropped from 40 to 28 RON, so we’re keeping the new version.
Cost per real result, not per click. A click isn’t a customer. Ask for the cost per lead, per order, or per booking — whatever a result means in your business. If the answer is “we can’t measure that”, that’s the first problem to fix, before any new campaign.
A trap I found in my own numbers
On the same subject: Meta reports two different figures that are both called “cost per click”.
CPC (cost per link click) — what it costs you to get one person onto your website.
CPC (all) — what any interaction costs you: a like, a comment, a tap on “see more”, a tap on the photo.
On one of my campaigns: 1,097 link clicks, but 1,635 clicks in total. Same money, two figures. 0.68 RON versus 0.46.
My own site showed 0.46. For four months. Nobody had lied — I’d taken the column next to the right one. I corrected it a few days ago.
If your report just says “CPC”, it’s worth asking which of the two. And asking for landing page views — how many people actually got to see the page. On my campaign, 855 of the 1,097 clicks made it. The rest closed the page while it was loading.
What to ask for, concretely
A message, not a meeting:
For this month’s report I’d also like: the age and placement breakdowns, video retention if we ran any clips, frequency, and what you changed in the account compared to last month. I’m less interested in the totals and more in what we decide for next month.
A serious provider will send these, because they already have them — they’re two clicks away in Ads Manager. One who explains why they aren’t relevant is telling you something about how closely they’ve looked at your account.
And when the answer comes in, read it with the test from the start: does it tell me where to put the budget next month?
If you’ve reached the point of wondering whether your current provider is the right one, I’ve written separately about how to choose a marketing agency.
// KEEP READING
More from this category
How to calculate ROAS correctly (and why platforms lie to you)
A pragmatic guide to calculating real ROAS in 2026: blended view, server-side tracking, GA4 vs Meta vs Google, and what to do when the numbers don't match.
What's a starting Facebook Ads budget in 2026?
Concrete recommendations per vertical: DTC e-commerce, B2B lead-gen, local services, medical. Why minimum budget went up in 2026 and how to avoid the learning-phase trap.
Facebook Ads or Google Ads for your small business: how to choose right
A practical guide for entrepreneurs: when Google Ads delivers, when Facebook Ads does, and how to choose based on your industry and budget — with real examples.
Put what you read into practice.
Book a free audit and let's apply the principles from this article on your business together.
Book audit →