Cap it, define it, and pay it on profit contribution rather than a ratio. A workable bonus is a fixed percentage of the incremental profit above an agreed baseline, measured over a quarter, capped in dollars, and payable only if the account is also healthy overall.
The short version
- Decide the maximum payout first, then solve for the percentage. Most sellers do this backwards and get surprised.
- Pay on the increment, never the total. You should not pay a bonus on profit the account already made without help.
- A quarter is the shortest honest window. Monthly advertising results on Amazon are mostly noise.
- The cap protects both sides. It keeps a lucky quarter from souring the relationship in month four.
- Tie the whole structure to an outcome you would hold any agency to, such as brand profitability inside the first year.
Start from the maximum, not the percentage
Write down the largest number you would be comfortable paying in an unusually good quarter. That figure is the cap. Everything else in the structure is arithmetic that fits inside it.
The reason to start there is that percentages hide magnitude. Fifteen percent of incremental contribution sounds modest until a seasonal quarter doubles your volume, and then a bonus that was designed as an incentive arrives as an unbudgeted invoice. The relationship rarely survives that, and the fault lies with the design rather than with either party.
Then work backwards. Estimate a realistic increment, divide the cap by it, and you have a percentage range that stays affordable at the top end and still means something at the target.
The arithmetic, with numbers
Here is an illustrative example. The figures are invented to show the shape of the calculation, not drawn from any client account.
| Line | Quarter figure |
|---|---|
| Baseline contribution after advertising spend, agreed at day zero | $120,000 |
| Actual contribution after advertising spend for the quarter | $156,000 |
| Incremental contribution | $36,000 |
| Bonus share, 15 percent of the increment | $5,400 |
| Cap agreed in advance | $9,000 |
| Base management fee for the quarter | $4,500 |
| Total paid to the agency | $9,900 |
| Increment retained by the seller | $30,600 |
Three things fall out of this table. The seller keeps 85 percent of the improvement, which is the test of whether a bonus is fair. The cap is about twice the target payout, so a great quarter is expensive but not alarming. And the base fee remains the larger share until the agency outperforms, which is what keeps the structure honest during a rebuild.
Now run the same table for a bad quarter. Contribution falls below baseline, the increment is negative, the bonus is zero, and the base fee still gets paid. That is the real risk split. Anyone describing this arrangement as shared risk is overselling it, and you should price it accordingly.
The clauses that make the arithmetic hold
- Baseline signed at day zero, from raw reports both parties exported, before any campaign or listing change.
- One metric only, stated with its formula. Contribution after advertising spend is the usual choice because you cannot buy it with budget.
- Cap stated in currency per quarter and per year, so an exceptional season cannot compound.
- Health conditions that void the bonus: rising return rate, degraded account health, or an inventory failure the agency caused.
- Exclusion windows for stockouts, agreed in advance so nobody negotiates a definition mid-quarter.
- A protected recommendation clause. Advising you to reduce spend or stop a product must never reduce what the agency earns.
Clause six carries most of the value. A bonus creates a standing financial reason to keep a weak product advertising, and the only reliable fix is to write the protection into the document.
Tie it to an outcome worth paying for
Bonuses fail when they reward a ratio rather than a result. A better anchor is the outcome you would use to judge any agency at all. The benchmark we hold ourselves to is that the majority of brands we take on are profitable within their first year, and that is a fair thing to build a bonus around: pay for the brand reaching profitability and staying there, rather than for an efficiency number that can be produced by shrinking the account.
If a candidate resists an outcome-linked structure, ask why. Sometimes the reason is legitimate, since profitability depends on landed cost and pricing decisions they do not control. That conversation is itself informative, because it forces both sides to name what the agency actually owns.
What most agencies will not tell you
Bonus structures are often proposed because they close deals, not because they change behavior. The base fee usually covers cost, so the bonus is upside on top of a business that was already viable, and the incentive effect on a busy operator carrying many accounts is smaller than the pitch implies.
The other unsaid thing is administrative. Somebody has to compute the number every quarter, agree it, and defend it. That work lands on you, because the agency computing its own bonus is not a control. Budget an hour a quarter for verification, or keep the flat fee and spend the hour on your product instead.
We stay on a flat monthly fee for exactly that reason, so weigh this page knowing my position. The structure I have described is what I would sign if I were buying, and I have declined to sell it, which tells you how much administrative weight it carries.
Related answers
- How does Amazon ads performance pricing work
- What’s a fair revenue split for Amazon sellers
- Amazon brand management pricing breakdown
- KPIs an Amazon agency should report weekly
- Hiring an Amazon agency: the complete guide
If you would rather skip the quarterly arithmetic entirely, our flat tiers are published at Flapen.
