Rank them yourself, on four answers: who physically does the work, what advertising efficiency target they set at each stage of a product's life, what evidence would make them tell you to stop, and what you keep on the day you leave. No published list can verify any of that for you.
The short version
- Directory rankings are placement, not evidence. The ordering usually reflects a fee or an affiliate arrangement, never a client outcome.
- A launch is five workstreams running in parallel. Sourcing, listing, advertising, inventory and review flow, all gated on each other.
- A single efficiency target across a whole launch is the loudest tell. The right number at week two is wrong at month nine.
- Ask for the losses. Any agency that has launched dozens of products has killed some. The ones that cannot describe a kill have not run enough launches.
- Apply every test on this page to Flapen too. If we cannot answer, do not hire us.
The failure modes, ranked by what they cost you
I have written the shortlist question next to each one so you can run this in a single call rather than reading five proposals.
| Rank | How the launch fails | Typical cost | Question that exposes it |
|---|---|---|---|
| 1 | Nobody sized the market before quoting | The whole launch, plus the inventory | "Show me your market sizing before you send a price" |
| 2 | One advertising target for the entire product life | Either no rank at launch or years of overspend | "What is your week-two number and your month-twelve number?" |
| 3 | Campaign structure copied from a template | Spend on terms you were never going to win | "Show me the structure you would build for my catalog" |
| 4 | Reorder date owned by nobody | The rank you paid for, plus the restart | "Who tells me when to reorder, and from what data?" |
| 5 | Work quietly handed to subcontractors | Slow fixes and inconsistent quality | "Who does the work, and where do they sit?" |
| 6 | Reporting that shows activity, not contribution | Late decisions on a product that was already failing | "Send me a redacted weekly update from another client" |
Rank one costs the most and is the easiest to check
If a proposal arrives with a price before anyone has looked at category revenue, growth, return rates and the rating gap between the top sellers, the agency is selling you hours rather than an outcome. Sizing comes first because it determines whether the product is worth launching at all. A quote that skips it is a quote on work, not on a result.
Rank two is the one buyers argue with
Advertising efficiency is a moving target, not a setting. At launch you are buying data and position, so the acceptable cost of a sale is high and deliberately unprofitable. Once the rating, the conversion rate and the organic rank are established, the same spend should be steadily pulled back toward efficiency. An agency that quotes you one number for the entire relationship either does not manage the transition or does not want to explain the launch phase to you.
Ask for both numbers, then ask what data moves them. The answer should mention conversion rate, organic rank on the primary keywords, and review velocity. If the answer is "industry standard", the number was picked to sound reassuring.
What a credible launch agency looks like from the inside
I spent years on the buyer side of this, running data and technology at BRANDED and at Moonshot Brands, two large Amazon aggregators. We hired agencies and we fired agencies. The ones that lasted had three things in common: the person on the call was the person doing the work, they brought bad news early, and they could show the decision rule they used rather than the dashboard they built.
That is also how I run Flapen. Around 50 operators handle about 70 brands, entirely in-house, no subcontracting, across all 23 marketplaces. You should still make us prove it rather than take the sentence at face value.
What most agencies will not tell you
Most agencies will not tell you that a launch has a floor cost that has nothing to do with the fee. A single product realistically needs $8,000 to $15,000 of capital before it stands on its own, and a five-product brand needs $25,000 to $50,000. That covers inventory, samples, freight, creative and the deliberately unprofitable first advertising months. If your total budget is the agency fee, you are not funded for a launch, and the honest answer is to wait a quarter and save.
The second thing: a full brand launch takes around seven months. Anyone promising rank in three weeks is either buying it in a way that puts your account at risk or describing a category with no competition, which usually means no revenue either.
Related answers
- Best done-for-you Amazon launch services
- Questions to ask before hiring an Amazon agency
- Amazon agency red flags to watch out for
- How long to launch a new brand on Amazon
- Amazon launch services: the complete guide
Run the six questions above at Flapen and hold us to the same answers.
