How to Choose an Amazon Agency in the USA

What the US market changes about the work

In smaller marketplaces an agency can create growth largely by doing competent things that nobody else is doing. In the US that is rarely available. Most viable categories already contain several well-funded operators who understand advertising, hold inventory properly and defend their listings. Growth comes from being better on specifics rather than from being present.

That shifts the emphasis in three ways. Advertising becomes a margin question rather than a volume question, because cost per click in competitive US categories will absorb any budget you hand it. Conversion rate work carries more weight, because a one point conversion improvement on US traffic volumes is worth more than most bid adjustments. And defence becomes a real workstream: unauthorised sellers, listing hijacking, review manipulation by competitors and variation abuse all occur more frequently at US scale.

The operational load is also heavier. Sales tax nexus across states, higher return volumes, more frequent FBA fee changes affecting margin, and a larger catalogue surface for things to go wrong on. An agency that is strong on advertising and thin on operations will look good for a quarter and then start losing money quietly in reimbursements and stranded inventory.

The four workstreams and how to tell they are actually staffed

Every agency will claim advertising, content, operations and strategy. The way to test it is to ask for the artefacts each workstream produces.

For advertising, ask to see a placement report and a search term report they have acted on, with the decisions annotated. Anyone managing spend seriously works from these weekly. For content, ask how they decide what to change on a listing and how they measure whether it worked. The good answer references search query performance and conversion rate by ASIN, not a rewrite that felt better.

For operations, ask for their case log or reimbursement summary. This is the clearest tell in the whole process. Teams that run operations properly have a filing rhythm, know their recovery rate and can tell you which categories of claim they pursue. Teams that do not will describe operations as something they help with.

For strategy, ask what they would stop doing in your account. A team with judgement will name something. A team selling hours will propose additions only.

Diligence that filters properly

Reference calls are worth more than case studies, and the question that matters is not whether the client was happy. It is what happened when something went wrong. Ask a reference about the worst month of the engagement and how the agency handled it.

On case studies, press on the baseline. A claim of two hundred percent growth means little without knowing the starting revenue, whether the category was growing, whether the brand launched new products in the period and how much additional ad spend went in. Reputable teams will give you the context. If the answer stays vague, treat the number as marketing.

Check whether the people in the pitch are the people on the account. This is the most common failure mode in US agency engagements: senior talent sells, junior talent delivers, and nobody told you. It is a reasonable thing for an agency to do, but you should know the shape of it and meet the actual operator before signing.

Finally, read the contract for three things: notice period, who owns the advertising data and creative assets when you leave, and whether there is an exclusivity clause that prevents you engaging a specialist elsewhere. Long lock-ins with short notice windows are the clause that causes the most regret.

Pricing models in the US market

The US market uses the same three models as everywhere else, with one addition that shows up more often at larger spend levels: a hybrid of a reduced retainer plus a performance component tied to an agreed metric.

Hybrid arrangements are the most interesting and the easiest to get wrong. The performance metric has to be something the agency genuinely controls and that you genuinely want. Tying a bonus to revenue growth rewards them for your seasonality. Tying it to total advertising cost of sales at a held or growing revenue level is closer to the thing you actually care about, which is profitable growth rather than growth.

Whatever the structure, get the scope boundary in writing. The disputes that end US agency engagements are almost never about the fee. They are about whether creative production, catalogue migration, a new marketplace launch or a suspension appeal was inside or outside the agreed scope.

Signals worth walking away from

Guaranteed rank or guaranteed revenue. Nobody controls the Amazon search algorithm or your competitors' budgets. A guarantee is either meaningless or it is being funded by tactics you would not approve of if they were described plainly.

Vagueness about who holds the account credentials and under what permissions. The correct arrangement is that you own the Seller Central account and grant scoped user permissions. Any structure where the agency owns or co-owns the account itself is a risk you do not need to take.

Reporting that only shows advertising cost of sales. It is the easiest metric to flatter by cutting spend on discovery campaigns, which improves the number while shrinking the business. Insist on total advertising cost of sales alongside it, plus units and contribution margin.

No mention of operations, compliance or account health in the entire pitch. It means either they do not do it or they have not thought about it, and both cost you later. In a US account at any real scale, operations is not a supporting function, it is where a meaningful share of the value sits.

Frequently Asked Questions