Judge the structure before you shortlist the firms. "Performance-based" covers three very
different models with different incentives, and one of them, percentage of ad spend, is
performance-linked in name only. Establish which model you are being offered before comparing
anyone.
The short version
- Three models hide behind one label. They behave very differently.
- Percentage of ad spend is not performance pricing. It tracks your budget.
- Revenue share aligns better but only works at scale.
- Reduced base plus bonus is the most workable hybrid.
- Ask what the baseline is and who measured it.
The three models called performance-based
I run Flapen with 50 operators managing about 70 brands. We charge a flat fee below
$50,000 a month in profit and revenue share above it, so I have operated on both sides of this.
| Model | What it tracks | Actual incentive |
|---|---|---|
| Percentage of ad spend | Your budget | Spend more |
| Percentage of revenue | Your sales | Grow sales, including unprofitably |
| Reduced base plus bonus | The agreed metric | Whatever metric you chose |
Only the third is performance pricing, and only if the metric is right.
Percentage of ad spend
The most common and the least defensible. Your goal is the lowest cost of customer acquisition
you can achieve. Their revenue rises with your budget. Those point in opposite directions.
The conflict is invisible in a good quarter and decisive in a bad one, because when a product
is failing the correct advice is usually to cut spend and fix conversion first.
Percentage of revenue
Better aligned, with two conditions. It needs an efficiency floor, otherwise the agency can
hit its number by pushing spend with your budget. And it punishes foundational work: if your
listing needs rebuilding before revenue can move, a revenue-share agency spends months earning
little for the work that matters most.
We offer it above $50,000 a month in profit, at 10 to 20 percent, because below that the
volatility is unfair to whichever side is unlucky.
Reduced base plus bonus
The workable hybrid. A lower fixed fee covers operating cost and a bonus rewards a defined
outcome.
The whole design decision is the metric. Contribution margin after ad spend and cost of
customer acquisition at held volume both work and neither can be gamed. Revenue can be bought
with your own budget. ACoS alone improves when you stop spending.
What to ask any performance-priced candidate
- Which of the three models is this, precisely?
- What is the baseline, and who measured it?
- What happens in a bad quarter?
- What happens when the right call is to reduce spend?
- Do kill criteria override the bonus?
Question two catches the most common weakness. A bonus paid against a baseline the agency set
for itself is not a performance structure, it is a target they chose.
Question five is the one nobody asks. Every performance structure creates pressure to keep a
failing product alive, because a dead product earns nothing. Write in that stopping a product
must not cost the agency its upside.
Why a flat fee is often better
Worth saying plainly, since I charge one.
A flat fee is the only structure where nobody in the room earns more by spending more of your
money. Ours runs $800 a month for one product up to $2,400 for five, and it does not move when
I recommend cutting spend or killing a product.
Every performance structure buys a share of risk and pays for it with a share of neutrality.
The advice you most need is the advice that reduces the agency's earnings, and each layer of
performance pay makes that advice more expensive for them to give.
What most agencies will not tell you
The label is doing a lot of work. "Performance-based" sounds like shared risk and is most
commonly implemented as percentage of ad spend, which shares nothing and tracks the volume of
your money moved.
Ask the direct test question: what happens to your income if I halve my ad spend. A flat fee
produces no change. Percentage of spend produces a proportional fall, which tells you exactly
where the alignment sits.
The other thing: performance pricing is easiest to sell to sellers who have been burned by a
retainer that delivered little. That is an understandable reaction and it frequently trades one
problem for a subtler one.
Related answers
- Alternatives to pay-for-performance Amazon management
- Fair Amazon agency pricing models
- What to ask about fee structure vs ad spend
- How to structure bonuses tied to Amazon revenue
- Hiring an Amazon agency: the complete guide
Halve your ad spend and our invoice does not move. Pricing is published at Flapen.
