The Monthly Inventory Review That Keeps Amazon FBA Storage Fees Down

Check what day of the month it is before reading further. In the first week you still have every move available: a removal or liquidation order filed now pulls those units out of this month's aged inventory assessment, a markdown has weeks to work, and a restock can land before the next snapshot. Read the same numbers in the last week and you are writing a post mortem on a bill that is already set. A large share of what sellers pay in FBA storage comes down to that difference.

Amazon runs four meters on your inventory at the same time, and they point in different directions. One charges you for holding too much. One charges you for holding too little. One ignores how well a product sells and counts only birthdays. Any single rule of thumb about storage will be wrong for part of your catalog in any given month, which is why sellers who keep this line flat run it as a fixed monthly job instead of reacting to an invoice that looks off.

The four meters running on your inventory

Know what is being counted before deciding what to move.

• Base monthly storage fee. Charged on the daily average volume your packaged, ready-to-ship units occupy, not the bare product. The rate varies by size tier and time of year and steps up sharply for peak. Seller Central lists the number for each tier, and it is worth reading before you commit a quarter of buying.
• Storage utilization surcharge. A second layer on the base fee, driven by your utilization ratio: volume kept stored divided by volume actually shipped over a rolling window. It is worked out per size tier, so a clean standard-size number says nothing about your oversize problem, and it only lands on sellers who meet all of Amazon's criteria at once. The ratio sits on the FBA Dashboard.
• Aged inventory surcharge. The old long-term storage fee under a newer name, stepping up through age brackets and ignoring sell-through entirely.
• Low-inventory-level fee. The meter that runs the other way. When a product's historical days of supply falls under Amazon's threshold on both the short and the long window, a surcharge rides along with the fulfillment fee on every unit shipped. Thinning out FBA stock is not free.

The second and the fourth pull in opposite directions, and most expensive storage decisions come from optimizing one while forgetting the other exists.

How Amazon counts a unit's age

Age is the meter sellers misread most. Amazon ages inventory first in, first out across the whole network: a shipped unit is deducted from your oldest stock, no matter which physical box an associate actually picked from. A slow trickle of sales does not keep a pile young. It shaves the front edge off it while the back keeps getting older.

Brackets step up with age, and in the oldest ones Amazon bills whichever is greater, a charge per cubic foot or a charge per unit. That greater-of rule is what catches people out. A light, cheap, flat product used to be safe to leave sitting because its cubic volume was trivial, and the per-unit floor took that shelter away. In 2026 Amazon added another bracket above the old top one and cut removal and disposal fees for light standard-size aged items, saying plainly that the point was to get old stock moving. Both edits push the same way.

First in, first out also changes what a restock means. Age belongs to the SKU, not to the shipment, so sending fresh units into a SKU that already carries an old tail averages nothing down. You get a young front and an old back, and the snapshot bills the back. A SKU with a real aged tail wants a clear-out and a clean restart, not a top-up.

Where in the month to run it

1. The aged inventory surcharge is assessed from a snapshot taken on a fixed day each month, and whatever is still there gets billed at the bracket it has reached.
2. A removal, disposal, or liquidation order submitted before the cutoff spares those units that month even if the pallet physically leaves much later. The order counts, not the truck.
3. Base rates rise for the peak period, which makes an early autumn review worth several times the same review run in spring.

Both the snapshot date and the order cutoff move around, so look them up in Seller Central once, put them in your calendar as repeating events, and set the review a few days ahead of the earlier one. A review that lands after the snapshot is a report, not a decision.

Four screens, and nothing else open

• Monthly Storage Fees report. Base fee and surcharge split per SKU. Sellers guess wrong here, because one bulky slow mover eats more volume than a shelf of fast little ones.
• FBA Inventory, filtered by age. The forward view is the useful half: act on units about to cross a bracket, not on units that already crossed.
• FBA Dashboard, storage utilization ratio. One number per size tier, and the threshold is a cliff rather than a slope.
• Capacity Monitor. On-hand plus inbound against your limit, by storage type.

Twice a year, check the dimensions Amazon holds for your top SKUs as well. Storage is billed on measured volume, so a wrong measurement makes every line above it wrong.

Which signal sends a SKU where

Every SKU lands in one of five states each month, and each state has one right move.

What you read	What it says	What you do	What that costs
Days of supply, per FNSKU	Both windows under the threshold	Restock deeper	Stored volume, capacity, cash
Days of supply and unit age	Nothing flagged	Leave it alone	The urge to tidy it anyway
Utilization ratio by size tier	Ratio near or over the threshold	Hold part of the buy upstream, replenish in waves	Second warehouse bill, transit time, placement per shipment
Age filter, forward view	Units cross a bracket before the snapshot	Mark down, or remove before the cutoff	Margin now, or a removal fee and weeks of processing
Age filter, deep brackets	No sell-through plan at all	Liquidate or dispose	Most of the value, but the meter stops

Fast movers running thin

Read historical days of supply on your best sellers, on both windows, per FNSKU. The low-inventory-level fee now works at that level rather than at the parent ASIN, so a weak size or color no longer hides behind a strong one, and a listing that looks healthy in aggregate can hide one variation adding a surcharge to every unit it ships.

Both windows under the threshold, or one under with the other sliding toward it, means restock, and restock deeper than the storage fee makes you comfortable with. This is the one group where the right answer is more inventory inside FBA. The price is stored volume, the input the utilization ratio feeds on. On a genuinely fast SKU the deeper restock lifts shipped volume too and the ratio holds. On a SKU you only believe is fast, that is how sellers walk into the surcharge, so split unproven quantities and send the second half later.

An exception sits at the bottom of this group. Products below Amazon's velocity floor are exempt from the low-inventory-level fee, and the tradeoff is slower delivery promises that no fee report will show you.

Well covered, and young

Days of supply comfortably above the fee threshold, the ratio for that size tier below the surcharge line, the age filter coming back empty. Nothing is flashing, so do nothing and close the tab.

What this group costs is discipline, because it is the one most likely to get optimized for no reason. The classic version happens in early autumn: a seller trims a healthy SKU to free room for something new, then pays the low-inventory-level fee on it through the busiest weeks of the year.

More stock than you can ship

Read the utilization ratio per size tier alongside the SKUs the Monthly Storage Fees report names as your biggest volume consumers. A ratio near or over the threshold, or a surcharge line next to the base fee, means these units are still young and the aged surcharge has not arrived yet. It will.

The move is to stop sending the whole purchase order into FBA. Keep the cover you can genuinely ship inside Amazon, hold the rest upstream at a warehouse that can also prep and label it, and replenish in waves. This is where third party storage for Amazon sellers earns its place: the units still exist, still labeled and ready, they are simply not on the meter.

Price three things before committing. A second warehouse bill, smaller per cubic foot than peak FBA storage but not zero. Transit time from the buffer, which has to be shorter than your cover or you have relocated a stockout rather than prevented one. And inbound placement, paid per shipment, so many small waves cost more than a few larger ones. Underneath all three sits a fourth: Amazon waives the utilization surcharge and the low-inventory-level fee for products it auto-replenishes from its own upstream warehouses, and an independent buffer carries no such waiver. Run that comparison with our side of it taken from our published rates rather than assumed.

Units walking toward an age bracket

Set the age filter to show what will cross into the next surcharge bracket before the next snapshot. The signal is a SKU holding more units than its sell-through will clear by that date. Do the arithmetic on the pile, not on the listing.

Work the options in order. Mark it down and push sell-through, including through Amazon Outlet, which requires a real discount to qualify. If that discount sells the units below what you can live with, file a removal and bring them to a warehouse where they can be re-inspected, relabeled, and either sent back later or sold elsewhere. If they are damaged or worth less than the freight, dispose. Each of the three is filed differently and comes back on a different timeline, so check the filing steps for removals, disposals, and liquidations before you promise a date to anyone downstream.

What belongs in this review is the arithmetic. A markdown gives up margin now instead of surcharge later, usually the cheaper trade if you catch it a bracket early. When the two look close, compare them as running totals rather than prices: take the monthly surcharge the pile will accrue in the bracket it is entering, multiply by the months current sell-through needs to clear it, and set that against the removal fee, the freight, and the cost of getting the units back into sellable condition, which you can price from our per-unit list. The surcharge side keeps growing while you postpone. The removal side does not.

For specific catalogs, a couple of details move the answer. Since Amazon stopped prepping and labeling units inside its own buildings at the start of 2026, anything you pull out has to be finished by someone else before it can go back, which belongs in the removal column as a real line (what the labeling change means for resellers). And dated goods need their own pass, since a bracket edge and a shelf-life edge rarely fall in the same week.

Stock that is not coming back

These are the SKUs with months of near-zero sales already sitting in the deeper brackets: no sell-through plan, no channel waiting for them, and a surcharge line that grows every month you keep thinking about it. Liquidate or dispose, before the snapshot rather than after. Once eligible inventory is submitted with a liquidation order, storage fees and the aged surcharge stop accruing against it.

This costs most of the value, and there is no version where it does not. Liquidators buy at a small fraction of what the product is worth, Amazon takes its own cut of the recovery, and the net can land near zero or go negative on cheap, heavy units. Disposal ends the meter without pretending to recover anything. Both beat another year of surcharges, and the only reason to take a removal instead is having somewhere real to sell the units.

What a buffer outside Amazon actually changes

The point of holding stock off Amazon is not that outside storage is cheap. It is that a buffer separates two decisions FBA forces you to make together: how much you buy, and how much you expose to the storage meter. Split them and purchase orders can follow your supplier's pricing while replenishments follow your sell-through. It lets you take a container-load price without carrying a container-load ratio, and that is where a buffer pays for itself long before it saves a dollar in storage fees.

• Capacity stops dictating your buying. Only on-hand plus inbound counts against the limit. Units in an outside warehouse do not, so a good quarter does not end with a blocked shipping queue.
• Rework becomes possible. Units inside FBA cannot be inspected, relabeled, bundled, or repackaged without a removal first. Units in a prep warehouse can. A listing change, a replaced barcode, a batch that needs a new poly bag: all of it happens before the meter starts, which is the ordinary case for FBA prep handled outside Amazon.
• Seasonal categories get a place to wait. Some products should not sit in Amazon's network during certain months at all, and they still have to be somewhere. Meltable goods are the clearest example, with their own seasonal restriction calendar.

Size the buffer to hold safety stock and overage. It is not there to let you run FBA on fumes.

Capacity limits are a buying decision

Every group above assumes you can ship what you decide to ship. The capacity limit decides that, and sellers usually discover it in the week they most need room.

• Limits are per storage type. Standard-size, oversize, apparel, footwear, and extra-large each carry their own, so emptying an oversize SKU does nothing for a standard-size restock.
• On-hand plus inbound is the number that counts. A shipment plan you created and never cancelled holds room against your limit while it sits open on your desk. Clearing that queue is part of the review.
• The published limit arrives late. It covers the month ahead, with estimates for the two after it, which is later than most freight has to be committed. Plan against the estimate, verify before booking.
• Extra room is auctioned in advance. You name a reservation fee per cubic foot, requests are granted from the highest offer down, and sales credits on the reserved space offset the fee. You cannot buy it for the month you are standing in, so if Q4 needs room, that decision belongs in a summer review, next to everything else on the Q4 prep timeline.
• The limit itself is earned. Inventory performance, sales history, Amazon's forecasts, and network capacity all feed it. Aged stock weighs on the same picture that sets your ceiling, so clearing it buys room twice: once by leaving, once by improving the input.

Inventory questions with a deadline

If I send fresh units into an old SKU, does the average age come down?

No. Amazon deducts sales from your oldest units first, so the old part of the pile keeps aging regardless of what you stack on top. Restocking a SKU with an aged tail gives you two populations inside one line item, and the tail is what shows up in the surcharge. Clear the tail first, then restock into a clean SKU.

Is it cheaper to discount aged units or to remove them?

A markdown gives up margin once and ends the storage clock through real sales. A removal costs a per-unit fee plus freight and processing time, and it earns its keep only if the units have a future somewhere else. If neither is true, the real comparison is liquidation against disposal.

Will a third-party warehouse lower my storage utilization surcharge?

It lowers the input the surcharge is calculated from, which is your stored volume inside Amazon's network relative to what you ship out of it. It is not a waiver. Amazon grants one only for products auto-replenished from its own upstream warehouses. A buffer works by moving the ratio, not by exemption.

How much inventory should I send to FBA at one time?

Enough cover to stay above the low-inventory-level threshold on both windows, plus your lead time from the buffer, plus room for a spike. Below that you pay a fee on every unit shipped. Far above it you pay the utilization surcharge and eventually the aged one. Send fewer, larger waves, since inbound placement is charged per shipment. There is no universal number, but those thresholds define the range you get to work in.

What if a SKU lands in a different group every month?

That is a forecasting problem wearing an inventory costume. Fix the replenishment trigger, hold the swing stock outside Amazon while you do it, and the group assignment settles down on its own.