I will not publish client numbers I cannot let you verify, and neither should anyone else. What a real revenue share case study must contain: a recorded baseline, a fixed window, the profit definition, what was excluded, and what failed. Without those five, it is a screenshot.
The short version
- A case study is a sample of one, chosen by the seller. That is not an accusation, it is the definition.
- Revenue share rewards volume. A study reporting revenue growth without margin has reported the wrong number.
- The baseline decides the headline. Measuring from the bottom of a trough doubles any result for free.
- Ask what they killed. A portfolio with no discontinued products has no decision discipline.
- Structure matters more than anecdote. Threshold, definition, and exit terms predict your outcome. Somebody else's chart does not.
The mistake this question usually contains
The search for revenue share case studies is almost always a search for reassurance before signing a performance deal. It is the wrong instrument for the job. Every case study you will find was selected by the party being evaluated, from a population you cannot see, over a window they chose after the results were known.
I could write one right now that would be entirely accurate and completely useless to you. So could anybody. The useful version of this question is not "show me a winner" but "show me the structure that made the win possible, and show me the losses from the same period".
That reframing is the whole point of this page.
What a revenue share case study has to contain
| Element | Why it decides everything | What a weak study does instead |
|---|---|---|
| Recorded baseline | The denominator of every percentage claimed | Starts the chart at the lowest month |
| Fixed measurement window | Prevents results being harvested after the fact | Picks the best quarter retrospectively |
| Profit definition | Revenue growth can destroy margin | Reports gross sales only |
| Excluded factors | Seasonality, a competitor stockout, a price change | Attributes everything to the agency |
| What failed | Shows the decision discipline actually existed | Presents an unbroken run of wins |
The last row is the one to insist on. Any operator running a portfolio has products that did not work. A firm presenting only successes is either new, or filtering, and both of those change how much weight the study deserves.
Failure modes of revenue share deals, ranked by what they cost
1. No kill criteria, which is the expensive one
Early on I poured money into a failing product for three months, convinced that better advertising would turn it around. It did not. The rating was drifting down, the return rate was telling me something I did not want to hear, and I kept funding it because stopping felt like admitting the product selection had been wrong.
That experience is where our scale, fix, and kill criteria came from. Rating trend, return rate, conversion rate, and customer acquisition cost trajectory, assessed over a defined window, with a decision at the end of it rather than a feeling. Under a revenue share arrangement this matters even more than under a flat fee, because the party advising you on whether to continue is paid on continuation. Write the kill criteria into the agreement before the first month, and name who has authority to call it.
2. Measuring the share against revenue instead of profit
Discounting increases revenue. Aggressive advertising increases revenue. Both can leave you with less money than you started with, and both pay the agency more. Define the share against gross profit after COGS, Amazon fees, freight, and ad spend, or accept that you have created an incentive to sell your product cheaply.
3. A threshold set too low
Revenue share below a certain scale is unfair to whoever is on the wrong side of a volatile month. We only work this way above $50,000 a month in profit, at 10 to 20 percent with no fixed fee, because below that a single stockout or a seasonal dip swings the economics violently. If a firm offers revenue share on a small account, ask what happens to them in a bad month, and then ask what that pressure will do to their advice.
4. Attribution left undefined
Who gets credit for a sale that came from an email list you built, a wholesale relationship you already had, or a category-wide demand spike? Under a flat fee nobody cares. Under revenue share it becomes the argument that ends the relationship. Write the attribution rule down while everyone is still friendly.
5. No exit path
Performance deals tend to acquire long terms, because the agency is carrying risk early and wants time to recover it. That is a reasonable ask and it is also how brands get trapped. Our engagements stay month to month with 30 days' notice regardless of the fee structure, and on exit the client keeps their Seller Central account, campaigns, creative, and a written handover.
What a published case study will not tell you
The denominator. How many accounts the firm worked on in the same period, and what happened to the ones not shown. A study of the winners from a large sample and a study of the only client are visually identical and mean opposite things.
It will also not tell you the starting condition. A brand with a broken main image and a good product has enormous available upside, and fixing it produces a chart that looks like genius. The same team on a well-run account produces a modest chart while doing harder work. Without the starting condition, growth percentages rank the state of the account before anyone arrived.
What I can tell you about our own book is the aggregate rather than the anecdote: the majority of brands we take on are profitable within their first year. That number includes everything, and it is the shape of claim you should ask any firm to make.
Related answers
- Amazon agency revenue share models explained
- Fixed fee vs rev share for Amazon agencies
- Compare hybrid fee plus rev share models
- Questions to ask before signing performance deal
- Amazon agency pricing and economics: the complete guide
Ask us for the aggregate rather than the anecdote at Flapen.
