The Amazon FBA sell through rate compares units shipped over the past 90 days with average inventory held in Amazon fulfillment centers during that period. Sales can rise while the ratio falls if average inventory rises faster. This guide isolates that numerator-and-denominator problem; for the broader inventory score, see the Amazon IPI guide.
Amazon’s US Inventory performance help page, checked November 12, 2026, supplies the formula used below. The examples are hypothetical calculations, not results from a real seller account.
Amazon FBA Sell Through Rate Formula and Units
FBA sell-through rate = units shipped in the past 90 days ÷ average units on hand in Amazon fulfillment centers over the same period.
That is the definition on Amazon’s US Inventory performance page, checked November 12, 2026. The page also describes its related “Units sold (past 90 days)” metric as total FBA units sold and shipped during that period. The numerator is units, not revenue, order count, page views, or units sent inbound to Amazon.
Suppose 900 units shipped during the relevant period and average inventory was 300 units:
900 ÷ 300 = 3.0
The result is a ratio of 3.0. It compares a flow of shipped units with the average stock supporting that flow. It does not mean that 3% of an inbound shipment sold, and it is not constrained to a maximum of 1.0.
Generic retail sell-through calculations often compare sales with units received or available for a particular purchase batch and express the result as a percentage. That answers a different question. Before comparing a benchmark with an FBA dashboard value, check both the denominator and the reporting period.
Amazon’s public FBA inventory blog uses slightly different wording: units sold and delivered over 90 days divided by average sellable units in fulfillment centers (checked November 12, 2026). For dashboard diagnosis, start with the dedicated Inventory performance help definition and the fields shown in your account. Do not silently substitute a delivered-sales export for a shipped-units field simply because both sources discuss sell-through.
Audit the Denominator Before Recalculating
Average inventory is not today’s inventory. A current stock balance is a point-in-time observation; the denominator describes inventory over the same trailing period as the numerator. Replacing the average with today’s balance changes the question even if the division itself is correct.
Suppose the numerator is 1,200 units. With an average inventory of 400 units, the ratio is 3.0. If today’s balance happens to be 500 and you substitute it, the answer becomes 2.4. That difference comes entirely from choosing a different denominator—not from an error in Amazon’s division.
Use this audit sequence before building a spreadsheet:
- Match the marketplace. Compare the same account and marketplace in each source. A combined sales export and a single-market inventory balance are not a matched pair.
- Match the period. Use the trailing 90-day window associated with the metric, not a calendar quarter, this month’s sales, or a convenient export date range.
- Match the fulfillment channel. The help page describes FBA shipped units and inventory at Amazon fulfillment centers. Keep merchant-fulfilled activity separate when selecting fields.
- Match the inventory definition. Check whether your export describes sellable, fulfillable, total on-hand, reserved, or another state. Record the field names rather than assuming they are interchangeable.
- Match the averaging method. Use an average supplied with a defined method, or document how your own reconstruction was made. Do not label an approximation an exact dashboard reconciliation.
The public US help page checked for this article defines the period average but does not specify a snapshot schedule or averaging algorithm. It therefore does not establish which inventory snapshots to use for an exact reconstruction. Keep any spreadsheet averaging method explicit rather than presenting it as Amazon’s documented algorithm.
If your reconstruction differs from the displayed value, preserve both readings and the export details. First locate the mismatch in scope, timing, or field definition. Repeatedly changing the denominator until the answers agree conceals the problem instead of resolving it.
Why Sales Can Rise While Sell-Through Falls
A ratio can deteriorate even when its numerator improves. The decisive comparison is how much average inventory changed relative to shipped units.
| Hypothetical observation | Units shipped in its trailing window | Average inventory in that window | Calculated ratio |
|---|---|---|---|
| Earlier observation | 900 | 300 | 3.0 |
| Later observation | 1,200 | 600 | 2.0 |
| Different later scenario | 1,800 | 450 | 4.0 |
In the second row, more units shipped, but the business carried a much larger average stock position. The ratio fell because inventory expanded faster than shipment volume. That is consistent with a major replenishment arriving before its expected demand, or with stock accumulating in slower products while faster products continue selling.
In the third row, shipments grew enough relative to average inventory that the ratio improved. The useful distinction is not simply “sales up” versus “sales down.” It is sales relative to the inventory held to support them.
The table’s observations are separate trailing windows, not three isolated days of inventory. They also do not establish why the change occurred. Add the operational events before choosing a response: a planned seasonal receipt has a different meaning from an accidental duplicate reorder.
A simple diagnostic is to write the two fractions side by side. If the newer numerator is larger but the newer fraction is smaller, inspect the denominator before blaming traffic, conversion, or the calculation. Conversely, if average inventory stayed steady while shipped units fell, investigate the sales side first.
Read Changes as a Rolling Window
The help page describes a rolling 90-day sell-through measure. As the observation date advances, older activity leaves the window and newer activity enters. A changed ratio can therefore reflect both something that just happened and something that stopped being included.
A promotion can leave the numerator. A strong earlier selling period eventually moves outside the trailing window. Recent sales may look normal while the 90-day total becomes weaker. Compare the activity leaving the window with the activity entering it before treating the change as a fresh demand collapse.
Replenishment can change the inventory side before its sales arrive. Stock is received ahead of demand. Depending on its timing and the period-average calculation, a receipt can raise average inventory while the associated customer shipments are still ahead. That can be a planned working-capital decision rather than an immediate pricing problem.
A removal changes stock, not customer demand. Removing inventory does not turn those units into sales. Its influence on a period average also differs from simply subtracting the removed quantity from today’s balance. Record the removal event and keep the shipped-unit numerator separate.
A stockout can make the ratio look efficient while reducing selling opportunity. A thinner inventory position is not automatically a better business outcome. Inspect availability and missed demand alongside the ratio, especially when planning the next order.
For each event, record its date, affected products, and expected side of the fraction. Then inspect actual values. This prevents the same dashboard movement from producing an automatic discount recommendation in one week and an automatic inventory cut in the next.
Use a Two-Window Diagnostic Worksheet
Keep a record for two observation dates. Those dates may have overlapping trailing windows; the purpose is to explain movement, not to pretend they are independent experiments.
| Worksheet field | Earlier observation | Later observation | What the comparison resolves |
|---|---|---|---|
| Marketplace and scope | Record account, market, product scope | Use the same scope | Whether like is being compared with like |
| Observation and export dates | Save both timestamps | Save both timestamps | Whether a data-timing mismatch is plausible |
| Window boundaries | Record the relevant 90-day period | Record the newer period | Which sales enter and leave |
| Shipped-unit numerator | Save value and source field | Save value and source field | Whether shipment volume changed |
| Average inventory denominator | Save value and definition | Save value and definition | Whether stock increased relative to sales |
| Reconstructed ratio | Divide matched values | Divide matched values | Whether your inputs explain the direction |
| Inventory events | Receipts, removals, availability issues | Newly relevant events | What may explain the denominator |
| Demand events | Promotions, price or listing changes | Newly relevant events | What may explain the numerator |
| Next action | Name an owner and action | Record follow-up outcome | Whether the diagnosis led to a useful decision |
When the account-level ratio moves, inspect the product mix. An aggregate ratio can hide strong products alongside a growing tail of slow stock. Rank the products requiring attention by their inventory position and commercial significance, then examine the underlying demand and replenishment history.
Avoid averaging individual product ratios to reconstruct an overall ratio. Ratios with different denominators do not combine correctly through a simple unweighted average. For any combined calculation you build, combine the appropriately matched unit inputs first, and describe the scope of that calculation. An internal product-level worksheet is a diagnostic aid, not proof that it duplicates every dashboard calculation.
Keep annotations short and specific: “additional receipt preceded seasonal demand” is more useful than “bad sell-through.” If the underlying data cannot support an explanation, make the next action a data check rather than changing price to fix a number you have not yet understood.
Choose an Action Without Chasing the Ratio
Use the diagnosis to choose the intervention:
- Inventory grew ahead of an intentional demand event: revisit the forecast and inbound schedule. Check that the expected demand still exists before canceling or repeating replenishment.
- Slow products accumulated while other products sold well: revise reorder quantities by product. A blanket reduction can starve the products that already have healthy demand.
- Shipments weakened while inventory stayed similar: inspect availability, listing status, price, traffic, and conversion. Choose a commercial response after identifying the constraint.
- The calculation cannot be reconciled: confirm scope, dates, and source fields before treating the discrepancy as an operating failure.
- Stock has become economically unattractive to hold: compare clearance and exit choices using the FBA liquidations guide rather than assuming an improved ratio justifies any discount.
For a repeatable replenishment process, the restock software guide provides a separate tool-selection framework. The worksheet here should establish what needs fixing before a software shortlist enters the decision.
What counts as a good rate depends on the operating context you are trying to support. The public US help page checked November 12, 2026 displays Excellent, Good, Fair, and Poor performance categories but does not provide numeric cutoffs in its text. Use your account’s current display rather than importing an undated threshold from another article.
The practical objective is a stock position that supports profitable demand without unnecessary accumulation. A higher ratio achieved only by running out of a profitable product is not a successful replenishment plan. Review the ratio with availability, margin, and the next expected receipt—not in isolation.
Frequently Asked Questions
Is an FBA sell-through rate of 3.0 the same as 300% of a shipment sold?
No. The FBA denominator is average inventory over the period, not a particular inbound shipment. A value of 3.0 means shipped units were three times that average stock figure. It does not identify what fraction of any specific purchase batch sold.
Can I calculate the rate when the denominator is zero?
Division by zero is undefined. Do not force the spreadsheet to display a meaningful ratio by replacing zero with an arbitrary small number. Check inventory scope, timing, and the metric displayed in the account before interpreting the result.
Does improving sell-through guarantee a particular IPI score?
This ratio alone does not calculate an IPI score. Diagnose the stock-and-sales relationship here, then use the linked IPI guide for the broader framework. A sell-through calculation is not a points-to-score conversion formula.