How do you read an agency report?
Read an agency report from the top down: revenue, orders, profit and new customers from your store backend first, then total ad spend and MER, and only then channel ROAS and click data. Check what the ad platforms claim against your store numbers, and ask whether the report leads you to a decision.
Monthly reports are often built the other way round. Page one shows impressions up, click-through rate up and "record engagement"; revenue appears on page eight, and even then it's the number the platform counted for itself. Whether or not that order is deliberate, the effect is the same: the real question, what the ad money did for the business, slides into the background.
Which metrics should a monthly report cover?
It helps to think in three layers. The top layer is the business result, the middle layer is advertising efficiency, and the bottom layer is for diagnosis.
| Layer | Metric | Where it comes from | Question it answers |
|---|---|---|---|
| Business result | Total revenue, orders, gross profit | Store backend (Shopify, ikas, others) | Did the business grow? |
| Business result | New customers and repeat purchase rate | Store backend or CRM | Is growth coming from new or existing customers? |
| Efficiency | MER (total revenue / total ad spend) | Store data plus ad invoices | What did total spend return? |
| Efficiency | Customer acquisition cost (CAC) | Ad spend / new customers | What did one new customer cost? |
| Efficiency | Channel ROAS and CPA | Google Ads, Meta, TikTok dashboards | Which channel and campaign is more efficient? |
| Diagnosis | CTR, CPC, CPM, conversion rate, frequency | Ad platforms and GA4 | Is the problem the creative, the targeting or the site? |
If the top layer looks bad, good news in the bottom layer doesn't mean much. And if revenue beat target, a slightly lower CTR is no reason to panic.
Why should MER sit at the top of the report?
Because MER doesn't depend on any platform's attribution model. You divide total revenue by total ad spend, and that's it. Example: you spent 200,000 TL across Google and Meta in a month and the store made 1,000,000 TL. MER is 5. The platforms may claim 1,400,000 TL in sales between them; the gap comes from the same order being counted by more than one channel. We cover MER and ROAS in more depth in our MER guide.
Why doesn't platform ROAS match GA4?
Each platform claims sales by its own rules. Meta counts click-through and view-through conversions by default, Google Ads uses its own attribution model, and GA4 assigns visits using its own channel grouping. According to Google's help page, the "Conversions" column in Google Ads includes only primary conversion actions and can include modeled conversions, while "All conversions" also covers secondary actions and view-through conversions. Comparing two months without knowing which column the report uses can mislead you.
GA4 has two separate reports as well. The traffic acquisition report shows where sessions from new and returning users came from; the user acquisition report shows only how new users first found you. If your agency says "new users from Meta are up", it's fair to ask which report they mean.
You won't make these gaps disappear. What you can ask for is platform data, GA4 data and store data side by side in the report, with the gap explained the same way every month.
What are vanity metrics, and why do they mislead?
Impressions, reach, clicks, follower growth and engagement rate aren't good news on their own. They are steps on the way to a result. A pattern we see often: a campaign opened to a broad audience brings cheap clicks, CTR rises and the report fills with green arrows, but the visitors don't intend to buy, so orders stay flat.
When you see a vanity metric, ask one question: "How did this increase show up in revenue or profit?" If the answer isn't clear, the metric belongs in the context section, not the headline.
What sections does a good agency report include?
Formats vary between agencies, but reports that work tend to follow this structure:
- Executive summary: the month's business result, the comparison with target and the three or four most important developments. One page.
- Target versus actual: revenue, MER, CAC and orders, each with target, actual and previous period.
- Channel performance: spend, platform ROAS and GA4 data side by side for Google, Meta, TikTok and anything else.
- Changes and tests: what changed this month, which tests ran and what happened, failed tests included.
- Problems and risks: tracking errors, stock issues, disapproved ads, site-driven drops.
- Next month's plan: budget split, new tests and the decisions the agency needs from you.
A live dashboard behind the report is also a good sign. Our guide on building an ecommerce report in Looker Studio shows how to set one up.
What should you ask your agency when reviewing the report?
A report meeting should be a working session, not a presentation. These questions usually open the most useful conversations:
- How much of this month's revenue change came from ads, and how much from seasonality or promotions?
- Did the gap between platform ROAS and MER widen or narrow compared with last month? Why?
- Can we see sales to new customers separately from returning ones?
- What happened to the tests we planned last month?
- If we raised budget by twenty percent, where would it go and what would you expect?
The last question is especially telling. It shows quickly how well the agency knows the account and whether it thinks in terms of marginal return.
Which signs in a report should worry you?
Some signs aren't a problem on their own but call for a serious conversation when they appear together. Targets that change quietly every month, for example ROAS last month and "reach" this month. Limited access to raw data; if you don't have admin access to your own ad accounts, that is a separate issue, and when you work with an ecommerce agency the accounts should ideally be opened in your company's name. And the most common one: a weak month padded out with bottom-layer metrics.
Note the opposite too. An agency that tells you plainly "this month was weak, here is why, here is what we're changing" is usually one you can trust over the long run.
If you'd like a second pair of eyes on your reports and a clear view of which metrics matter for your store, you can request a free growth audit from the Performetic team.
Key takeaways
- Read an agency report from the business result down: revenue, orders, profit, new customers.
- MER is independent of platform attribution and belongs at the top of the report.
- Platform ROAS, GA4 and store data give different numbers; the gap should be explained the same way each month.
- Impressions, clicks and engagement are context, not proof of success.
- A good report says what happened, why it happened and what changes next month.