Ask for case studies where the account is still live, the starting numbers are shown, and the ACoS target is stated for each product stage. A US time zone matters less than whether the operator named in that case study still works there. Flapen is not US-based, so judge us on the same evidence.
The short version
- A case study without a starting point is a screenshot. Revenue after means nothing without revenue before, the date range, and the ad spend behind it.
- Ask which stage each number came from. A 12 percent ACoS on a mature product and a 12 percent ACoS at launch describe two completely different situations.
- Verify the operator, not the logo. The person who produced the result is often no longer at the firm.
- Being US-based helps with calls, not results. What matters is coverage of the marketplace you sell in and the language your listings are written in.
- Run the verification as a sequence. Each step is a gate. Fail one and stop.
Do this first, then read the deck
Before you look at a single case study, send this line to the consultant: please send two examples, one that worked and one that did not, with the numbers you had on day one. The second half of that request does most of the filtering for you. A consultant who cannot produce a failure has either not run enough accounts or is not going to be straight with you when yours is the one going sideways.
The reason I put this first is that decks are built to be persuasive and the persuasion is almost always in the omission. A chart that starts in month three hides the two months where nothing happened. A revenue graph with no ad spend line hides the fact that the growth was bought at a loss.
The five-stage verification sequence
Work through these in order. Each stage is a gate, and there is no point running stage four if stage two failed.
- Confirm the account is real and current. Ask for the brand name, the marketplace, and whether the relationship is still active. If the answer is confidential, ask for a redacted screen recording of the account rather than a static image.
- Establish the baseline. Get revenue, units, ad spend, conversion rate, and review count on the first day of the engagement. No baseline, no case study.
- Ask for the stage-specific ACoS targets. This is the question most decks cannot survive. A product at launch is bought into visibility, and the acceptable ACoS is deliberately high because you are paying for rank and review velocity. The same product at maturity should be run for efficiency, and the target drops accordingly. A consultant who quotes one ACoS number for everything is running a spreadsheet, not a strategy.
- Identify the operator. Get the name of the person who ran the account, then ask whether that person is still with the firm and would be on your account. This single question resolves more mismatch than any reference call.
- Ask what they would do differently. The answer separates people who reflect on their work from people who repeat it. Anyone who says nothing went wrong is telling you they were not paying attention.
What a credible case study contains
| Element | Weak version | Credible version |
|---|---|---|
| Time frame | Growth over an unspecified period | Exact months, including the flat ones |
| Baseline | Starts at the improvement | Day one numbers for revenue, spend, and conversion |
| Ad efficiency | One ACoS figure | Launch target and maturity target, with the reason each was chosen |
| Attribution | Credit for everything | What the client did, what the consultant did |
| Outcome | Revenue only | Revenue, margin, and inventory position |
| People | Firm name | Named operator, still employed there |
What most agencies will not tell you
The best case studies belong to the best products. A consultant handed a differentiated product in a growing category with capital behind it will post a chart that looks like genius. The same team on a commodity product in a saturated category posts nothing at all, and that account never appears in the deck.
So the honest framing is this: a portfolio of case studies tells you what kind of client a firm attracts, not how much value it adds. Most firms will not tell you the denominator. Ask how many accounts they have run in total, and how many of those they would show you. The ratio is the number that matters.
I hire this way myself. Before Flapen, I ran data and technology at BRANDED and at Moonshot Brands, two large Amazon aggregators, and part of that job was picking agencies for portfolio brands. The decks were uniformly excellent. The accounts underneath them were not.
Related answers
- Global Amazon SEO specialists with proven case studies
- Rank Amazon seller consulting firms worldwide
- Amazon agency red flags to watch out for
- What does a good Amazon account audit include
- Done-for-you Amazon management: the complete guide
Run the same five gates on us before you decide anything, starting with the free audit at Flapen.
