Take the fixed fee until your brand clears about $50,000 per month in profit. Below that line, revenue share transfers your volatility onto the agency and pushes them toward short-term sales. Above it, a 10 to 20 percent share with no monthly fee usually beats any retainer for both sides.
The short version
- The threshold is profit, not revenue. A brand doing $200,000 in sales at thin margin is a fixed fee brand.
- Fixed fee prices the work. Rev share prices the outcome. Choose based on whether the work or the outcome is the uncertain part.
- Rev share punishes rebuilds. Three months of listing and creative repair produce almost no share to collect.
- Attribution is the hidden fight. Agree what counts as attributable revenue before signing, not during month four.
- Either model can be honest. The test is what happens when the right advice is to spend less.
Settle these nine things before you sign either one
Work down the list. Each item has a version that counts as done properly, and a version that quietly costs you money later.
- Define the base. Done properly means one sentence naming gross sales, net sales after returns, or contribution profit, with the report that produces it. Anything vaguer becomes a dispute.
- Define the baseline. Done properly means the agency shares in growth above a stated starting point, not in revenue that already existed before they arrived.
- Set the threshold. Done properly means a number. Ours is $50,000 a month in profit, above which we take 10 to 20 percent and no fixed fee.
- Name the ad spend owner. Done properly means your media budget is yours, invoiced by Amazon to you, never routed through the agency's revenue.
- Write the ACoS targets. Done properly means two numbers, not one. More on that below.
- Fix the reporting cadence. Done properly means written weekly and reviewed live every two weeks, so a bad month is visible in week two rather than month three.
- Set the exit. Done properly means month-to-month with 30 days' notice, and a written handover of account, campaigns, and creative on the way out.
- Settle IP. Done properly means deliverables become yours on full payment, with the agency keeping only its own internal tools and methods.
- Read the restrictive clauses. Done properly means no non-compete on you. A non-solicit protecting the agency's staff is reasonable. A clause restricting your own future sales is not.
Where ACoS enters the argument
This is the clause most people skip, and it is where the two models split hardest.
A single ACoS target is a red flag in either model. The correct target changes by product stage. At launch you are buying rank and data, and the efficient number is deliberately bad. At maturity the same product should be run tight. An agency quoting one blended target across a catalog is either managing to an average or has not thought about it.
Under a fixed fee, a loose launch target costs you ad budget and nothing else, so the incentive is neutral. Under revenue share, aggressive spend inflates the number the share is calculated on. That is the specific conflict to price into the contract: ask for the launch number and the maturity number in writing, and agree who signs off on moving between them.
| Question | Fixed fee | Revenue share |
|---|---|---|
| Who carries a slow quarter | You | The agency |
| Incentive when spend should drop | Neutral | Against you, unless the base is profit |
| Cost at scale | Flat and predictable | Rises with success |
| Cost during a rebuild | Predictable | Near zero for the agency, which creates pressure |
| Best fit | Under $50,000 a month in profit | Above it |
The arithmetic, briefly
A brand at $30,000 a month in profit paying our four product tier pays $1,950, which is 6.5 percent of profit. The same brand on a 15 percent share pays $4,500. The fixed fee wins.
At $80,000 a month in profit, the fixed fee is 2.4 percent and the share is $12,000. The fixed fee still looks cheaper on paper, which is exactly why we only offer the share above the threshold and why an agency offering it below the threshold is usually protecting its own downside rather than yours. Above $50,000, the share buys something a retainer does not: the agency is now underwriting the outcome, and the conversation changes from tasks completed to profit produced.
What most agencies will not tell you
Revenue share is easier to sell than a retainer because it sounds free. Nobody argues with a number that only exists if you win. The part left out is that the base is almost always revenue rather than profit, and revenue is the one metric an agency can move without helping you at all. Discount hard, spend hard, and revenue climbs while your margin does not.
The second thing: rev share and fixed fee are usually presented as the whole menu, and they are not. A short paid pilot, a scoped project fee for a specific rebuild, or a flat fee with a bonus tied to a named profit milestone all exist. If a proposal only offers two options, ask for a third.
Related answers
- Amazon agency revenue share models explained
- Alternatives to percentage of ad spend model
- Contract terms to negotiate with Amazon agencies
- Typical retainer vs performance-based for Amazon management
- Amazon agency pricing and economics: the complete guide
Both structures, with the threshold that separates them, are published at Flapen.
