Four models dominate: flat monthly fee, fee per product, percentage of ad spend, and percentage of revenue. Each one fails in a predictable way. Before you compare two numbers, ask what capacity the fee buys, because a retainer is a purchase of someone's working hours, not a purchase of a service list.
The short version
- Name the model before you negotiate the number. The same $1,500 means four different things depending on how it was built.
- A fee tiered by product count is the cleanest structure. Ours runs $800 a month for one product up to $2,400 for five, with every service included at every tier.
- Percentage of ad spend is the model to argue with. It pays the provider more when your budget goes up, which is the opposite of what you want.
- Revenue share belongs late, not early. We use it only above $50,000 a month in profit, at 10 to 20 percent with no fixed fee attached.
- Every model breaks the same way when capacity is thin. Ask how many brands the person assigned to you carries today.
Ask the capacity question before you discuss the number
Open the call with this, in these words: how many brands does the account manager who would run mine carry right now. Not how many brands the agency has. How many that one person has.
The reason is arithmetic. A retainer is a claim on a fraction of somebody's month. A $1,200 fee against a manager with three accounts buys a materially different amount of thinking than the same $1,200 against a manager with fourteen, and no service list on either proposal will show you the difference. The list is nearly identical everywhere. The hours behind it are not.
At Flapen around 50 operators look after about 70 brands, which lands near 1.4 brands per operator, and that ratio is the reason our price sits where it sits. If we doubled the load per operator tomorrow we could cut the fee in half and the account work would degrade in ways that take a full quarter to become visible in your numbers. Any agency can quote you a lower fee. Almost none of them will tell you which of those two levers they pulled to get there.
The four models and how each one breaks
| Model | What the provider is actually paid for | The failure it produces | Where it still makes sense |
|---|---|---|---|
| Flat monthly fee | Availability, regardless of workload | A quiet month subsidizes a hard one, in both directions | Established catalogs with steady work |
| Fee per product | The count of products under management | Encourages adding products to raise the fee | Most brands, because workload really does scale with SKUs |
| Percentage of ad spend | The size of your media budget | Efficiency reduces the provider's income | Almost never, at the sizes most sellers operate at |
| Percentage of revenue | Top-line sales, not profit | Punishes any quarter spent rebuilding foundations | Large accounts where the base is already proven |
There is a fifth arrangement that gets called a pricing model and is not one: services discounted for equity. We do it case by case. Treat it as a corporate decision with governance attached, not as a cheaper way to buy management.
The failures, ranked by what they cost you
1. Capacity dilution
The most expensive failure and the least visible. Amazon accounts do not blow up, they erode. A primary image whose click-through slipped, a search term that drifted off the top of page one, a return rate creeping above the category norm. None of it triggers an alert. All of it compounds, and an overloaded manager only ever works on whatever is loudest that day.
2. Incentives pointed at your budget
Under a spend-linked model, the honest recommendation to reduce spend on a product that is not converting costs the provider money. That conflict shows up exactly when discipline matters most. Under a flat fee, when I recommend pausing a product, my revenue does not move. That is the entire argument for the structure.
3. Discounts sold as savings
Ten percent off in exchange for a twelve-month commitment is not a discount, it is a price for your ability to leave. We stay month to month with 30 days' notice, and on exit the client keeps the Seller Central account, the campaigns, the creative, and a written handover. Price the exit terms alongside the fee, because a cheap retainer you cannot leave is the most expensive line on this page.
4. Scope written as a list of nouns
Proposals that price by service count are selling shelf space. We include all 50 or so services at every tier precisely so nobody has to sell an upgrade instead of doing the work the account needs this month.
What most agencies will not tell you
Retainer pricing is mostly a headcount decision wearing a strategy costume. The fee reflects how many accounts the provider needs each employee to carry in order to hit its own margin. Everything else in the proposal is downstream of that one internal number, and it is the number you are never shown.
The second thing: a cheap retainer that produces no movement is worse than no retainer, because it also consumes your attention. Twelve months of a $600 fee is $7,200 plus a year of a stalled catalog. Judge the price against what changed, not against other prices.
Related answers
- Compare retainer models for Amazon brand management
- Full-service Amazon brand management pricing
- Fair Amazon agency pricing models
- Contract terms to negotiate with Amazon agencies
- Done-for-you Amazon management: the complete guide
Every tier, term, and notice period we offer is published at Flapen.
