I co-founded Garage Collective after spending seven years on the client side, including three years reviewing agency reports that I now realize were designed to obscure more than they revealed. When your agency sends you a 40-page deck full of charts, it feels like transparency. It is usually the opposite. Here are the seven red flags I have learned to spot — and what a genuinely honest report looks like.
Red Flag 1: Cherry-Picked Date Ranges
The most common trick in agency reporting is comparing the best recent period to the worst past period. "Conversions up 45% vs last quarter!" — but last quarter included Diwali, when the client paused campaigns for two weeks. Honest reporting uses consistent date ranges: month-over-month for the same month last year, or trailing 4-week averages that smooth out anomalies.
Another variant: reporting on a 7-day window that happens to include a flash sale. If your agency's reporting period mysteriously shifts from month to month, ask why. At GC, we lock in reporting periods during onboarding and never deviate without flagging the reason in the report itself.
Red Flag 2: Impressions and Reach as Headline Metrics
If the first page of your report leads with impressions and reach, your agency is padding. These are input metrics, not output metrics. They are relevant as context — understanding why conversions dipped because reach dropped is useful. But leading with them signals that the agency does not have a better story to tell.
A proper report leads with business metrics: revenue or leads generated, cost per acquisition, ROAS or cost per lead, and how these compare to target. Input metrics like impressions, reach, and CTR appear in the supporting sections to explain the drivers behind the business results.
Red Flag 3: No Attribution Discussion
If your report shows conversions without specifying the attribution model, it is incomplete at best and misleading at worst. A campaign can show 500 conversions on last-click attribution, 320 on time-decay, and 220 on incrementality testing. Same campaign, same spend — wildly different results depending on how you count.
Your agency should state the attribution model on every report, explain its limitations, and ideally run periodic incrementality tests (geo-based holdout tests or conversion lift studies) to validate that the reported numbers are real. If they have never mentioned the word "incrementality" to you, that is a red flag the size of a billboard.
Red Flag 4: Hiding Underperformers in Averages
"Your Meta campaigns averaged a 3.2x ROAS this month." Great — but that average includes a retargeting campaign at 8x and three prospecting campaigns at 1.1x. Averages hide the distribution. A good report shows performance by campaign, by audience segment, and by creative — with explicit callouts on what is underperforming and what the plan is to fix it.
We have a rule at GC: every monthly report must include a "What did not work" section that is at least as long as the "What worked" section. This is uncomfortable for agencies because it feels like admitting failure. But marketing is an experimentation discipline — if nothing failed, you were not testing aggressively enough.
What a Good Report Actually Looks Like
Our reporting framework at Garage Collective has four sections. First, an Executive Summary: 3-5 sentences covering business results vs targets, the single biggest win, and the single biggest concern. Second, Business Metrics: revenue/leads, CPA, ROAS, LTV — compared to target, last month, and same month last year. Third, Channel Deep Dives: per-channel breakdown with campaign-level data, creative performance, audience insights, and search term analysis. Fourth, the Action Plan: specific changes we are making next month based on this month's data, with expected impact.
The entire report is 8-12 pages, takes 15 minutes to read, and every chart answers a specific question. If your agency report takes an hour to get through and you still do not know whether marketing is working, the report is not serving you — it is serving the agency's retention strategy. Request a copy of our reporting template by reaching out to us directly.
Key Takeaways
- Watch for cherry-picked date ranges — insist on consistent month-over-month comparisons with YoY context.
- If impressions and reach lead the report, the agency is hiding weak performance metrics behind vanity numbers.
- Every report must specify the attribution model and include periodic incrementality tests to validate conversions.
- Demand campaign-level performance breakdowns — averages hide underperformers that are burning your budget.
- A good report has a mandatory "What did not work" section and a specific action plan for next month.