Field Guide Amazon and multi-channel operations
Amazon IPI, explained by an operator who manages it
The Inventory Performance Index decides how much you are allowed to send to Amazon's warehouses. The 2026 threshold sits at 400, it updates weekly on a rolling 90-day window, and the four inputs are fixable in a specific order.
The Inventory Performance Index is the score that decides how much inventory Amazon will let you send to its warehouses. In 2026 the threshold sits at 400, the score updates weekly on a rolling 90-day window, and it is built from four inputs you can fix in a specific order. Stay above 400 and you are unrestricted. Fall below and your storage gets capped, usually at the worst possible time.
I manage this across a portfolio of brands, so I will skip the theory and give you what actually moves it.
What the score is really measuring
IPI is Amazon’s way of asking one question: are you using the warehouse space we give you efficiently, or are you parking dead stock in it? Everything in the score is a proxy for that question. Once you see it that way, the four inputs stop feeling arbitrary.
What IPI does not measure
Just as useful is knowing what does not move it, because most of the panic around IPI is spent on the wrong things. Your star ratings and reviews do not feed it. Your advertising spend does not feed it. A sudden spike in sales does not rescue it overnight, and a single slow week does not sink it. IPI is a measure of inventory efficiency over 90 days, nothing else. Stop reading it as a general health score for your account and you stop reacting to it emotionally.
Fix them in this order
Not all four move the needle equally, and some are faster than others. This is the order I work them.
Clear stranded inventory first
Stranded inventory is stock that is sitting in a fulfillment center with no active listing to sell it. It is pure dead weight on the score and it is the fastest thing to fix, because it is usually a listing error: a suppressed listing, a pricing error, a lost buy box eligibility. Open the Stranded Inventory page, fix or relist each one, and you often recover points within a week.
Cut excess inventory
Excess inventory is stock you are holding far more of than your sell-through justifies. It drags the score and it costs you storage fees on top. Create outbound removals or run a sensible markdown on the worst offenders. Do not fire sale everything; identify the specific ASINs flagged as excess and act on those.
Protect your in-stock rate
The in-stock rate measures how consistently your active, popular listings stay available. Running out of your best sellers hurts twice: you lose the sales and you lose IPI points. This is where restock discipline and a real reorder cadence earn their keep.
Improve sell-through over time
Sell-through is the slow one, the ratio of what you sell to what you store. It improves as a consequence of the first three done well, plus better demand forecasting. Do not expect to move it in a week; expect to move it over a quarter.
Manage the four inputs. The score is just the scoreboard.
Where each input lives in Seller Central
You do not have to guess at any of this. Amazon shows you the exact ASINs behind each input, which is what makes IPI a to-do list rather than a mystery.
- Stranded inventory has its own page under the Inventory menu, listing every unit with no sellable listing and the reason for each.
- Excess inventory appears in the Inventory Performance dashboard and the Manage Excess Inventory tool, which flags the specific ASINs and recommends an action for each.
- In-stock rate and sell-through both live on the Inventory Performance dashboard, with the historical trend so you can see which direction you are moving.
Work from those lists, not from a general sense that the score is low. Every point of IPI traces back to a named ASIN you can do something about.
What a storage restriction actually does to you
It helps to know the stakes before you are near them. When your IPI drops below the threshold, Amazon applies storage volume limits to your account. That is not a fee. It is a cap on how many units you are allowed to have in FBA at all, across your whole catalog. The practical effect is brutal: you cannot send in the inventory you need, so your best sellers go out of stock, which lowers your in-stock rate, which lowers your IPI further. The restriction feeds the problem that caused it. Avoiding that loop is the entire reason to manage IPI before it is a crisis.
Why the timing matters
Because the score runs on a rolling 90-day window and updates weekly, two things are true at once. Your fixes do not show up overnight, so do not panic-check it daily. And a bad quarter keeps weighing on you for about three months after you have corrected it, so the time to act is before you are near the line, not after.
A weekly IPI routine that keeps you clear
- Check and clear any new stranded inventory
- Review the excess inventory flags and action the worst
- Confirm your top sellers are in stock and reordering on time
- Glance at the score trend, not the daily number
Plan IPI around your peak season
The rolling window is also why peak-season planning matters. If you intend to send a large fourth-quarter shipment into FBA, your IPI needs headroom months before, because a restriction in October undoes the whole plan. Work your score up through the summer, clear excess before you reorder heavily, and go into your busy window with room to spare. The brands that get caught are the ones that treated IPI as a thing to check in December.
The number to actually target
The threshold is 400, but managing to 400 is managing to the edge of a cliff. The first slow season or supplier delay pushes you over. Experienced operators target 550 to 600 or higher, because that headroom is what lets you sleep through a soft month without a storage restriction landing on top of it.
So treat 550 as your floor, not 400. Run the weekly routine, fix stranded stock the moment it appears, and keep excess inventory honest. Do that and IPI stops being a thing you react to and becomes a thing that quietly stays handled. It is one piece of the wider Amazon and multi-channel operations discipline, where the same habit applies everywhere: manage the inputs, and the scores take care of themselves.
If you are running this across more than one brand and it has become a weekly fire drill, that is exactly the kind of operations problem I build systems for.