Fulfillment by Amazon is the arrangement where Amazon stores your inventory, handles the shipping, and charges fees for it. That is the whole definition. The decision underneath it is larger: whether this product deserves a warehouse at all, which market size, return rate, and 200 units of real data answer.
The short version
- The letters name a service, not a strategy. Amazon holds the units, ships the orders, and bills fees for doing both.
- The decision sits one level above the letters. Whether a first product belongs in a warehouse is settled by its market, not by who tapes the box.
- A market under $2M a year does not earn a pallet. Below that line, what remains after acquiring customers never repays the capital.
- Under 8% is the return rate we hold a market to. Returns erode margin quietly, and no fulfillment arrangement repairs them.
- 90+ data points decide the call, never a review count. Market size, growth trajectory, return rate, and the rating gap move the answer.
FBA on Amazon in one line, and the decision it hides
Amazon stores the inventory, handles the shipping, and charges for the service. Everything else attached to the letters is a schedule of fees and conditions that changes by category and by date. Confirm the current one inside your own Seller Central account, because that is the only copy written against your products.
The definition takes a line. The search is common because sellers treat the letters as a business model, when they name a service line on a cost sheet.
A market too small to repay the work stays too small in any warehouse. A product customers send back keeps coming back whoever receives the parcel. So the first decision is not which arrangement to use, and it is settled long before a unit ships.
Three answers to one product, side by side
Every candidate product has three available answers, and one makes the warehouse question live. Read the row your product qualifies for today, not the one you prefer.
| The answer | What has to be true first | What it costs | What the letters decide |
|---|---|---|---|
| Do not order it | The market turns over under $2M a year, or returns run above 8% | A week of research | Nothing, since there is no unit to store |
| Order 200 units and validate | The market clears the floor, grows year over year, and can be reached profitably | $5,000 to $10,000 once, up to 4 products at a time | Where a small run sits, and how fast a defect reaches you |
| Commit real capital | Rating, conversion rate, and cost of customer acquisition are proven | Phase 2 money, sized by the share you can realistically win | Warehousing at volume, an operations question with money behind it |
Flapen figures as of September 2026.
One decision rule closes the table. The fulfillment question is worth asking on the middle row, and only once the top row is ruled out.
Most sellers searching the definition already stand on the middle row, having ordered stock before the top row was checked. The common advice says launch aggressively with full inventory, and Phase 1 says the opposite, so a wrong call costs one small run instead of a container.
What decides the call: 90+ data points, not a review count
The tools most sellers open report review count and a snapshot of today. We read 90+ data points behind every launch decision, and four carry most of the weight: market size, growth trajectory, return rate, and the rating gap. Step 2 of our system then sets the product bar at 0.2 stars above the niche average, read from the negative reviews of everything already selling.
| What a snapshot answers | What the decision needs | Where the reading comes from |
|---|---|---|
| How many reviews the sellers above you hold | Whether the market is growing year over year | Trajectory across seasons, never one screenshot |
| How many units a page is estimated to move | Whether returns hold under 8% and the rating gap leaves room | Return behavior and the negative reviews in the niche |
The rule under both rows: a number without a direction is a snapshot, and a snapshot has never decided a launch here.
We scored 193,753 niches at the 2026-08-26 capture and 4.8% of them passed. Most markets fail on size, growth, or returns long before a fulfillment arrangement is chosen.
The seller who writes to me at this point runs one to three products at $5K to $30K a month, and the line reads I'm spending money on ads but don't know if it's working. Advertising is rarely the layer that broke. Fifty operators here run about 70 brands by hand, and the accounts that stall cleared no market bar before the first purchase order.
What an agency will not tell you about FBA
Three sentences get sold at this stage, and each replaces a decision. Read them beside what is being decided.
| What the pitch says | What is being decided | What you put in writing instead |
|---|---|---|
| We will set up and manage your fulfillment | A task measured in hours, done once | Who owns the market call before stock is ordered |
| We will keep your costs efficient | A cost line already printed inside your own account | The four signals that would make them tell you to stop |
| The category is enormous | Whether the market clears $2M a year and still grows | The size, the trajectory, and the return rate they measured |
One rule ranks the three rows. Pay for the decision that comes before the pallet, never for the administration that follows it.
The four signals in the middle row are rating trend, return rate, conversion rate, and cost of customer acquisition trajectory. If none of them improves inside 60 to 90 days, the product gets killed. No emotion.
Hold us to all three rows. Management here is flat at $800 a month for one product with all 50+ services included, on 30 days' notice, and you keep the account, the campaigns, and the creative on exit. If the market reading says do not order the units, do not order them, and do not hire us either.
Related answers
- FBA creatives
- Amazon FBA consulting
- Best agencies for Amazon product launch strategy
- Done-for-you Amazon catalog management
- Done-for-you Amazon management: the complete guide
One free thing to do this week, if you run one to three products at $5K to $30K a month. Write one line per product carrying three numbers: what the market turned over last year, whether that figure grew, and your own return rate over the last 90 days.
Then hold each line against $2M a year and 8%. A product that misses both does not have a fulfillment problem.
To have those three numbers read against the same thresholds our operators use, ask for the free written audit and get prioritized fixes back within 48 hours at no cost from Flapen.
