Amazon changed the rules again, and my whole multi-platform inventory allocation plan for peak season had to be redone. In May 2025, FBA capacity math switched from six months of projected sales to five. Sellers woke up to 40% less space overnight, and some lost up to 75%. Public reports described sellers whose capacity dropped from 450 cubic feet to 130. Black Friday hadn't even started and their stock couldn't get through the door.

Here's the part that stings: Amazon has measured FBA capacity in cubic feet, not units, since March 2023. Bulky products burn through your allowance faster. Individual seller accounts are capped at just 15 cubic feet. Professional accounts get an allowance based on sales history and IPI score, and a score below 400 triggers restrictions. Want more space? Sure — go bid for it in Capacity Manager. You pay the reservation fee up front, and there's no refund if the stock doesn't sell.

So flip the question. Stop asking "how much can I cram into Amazon?" Start asking "how much does each platform deserve?"

Picture this: a home-storage seller in Los Angeles put 80% of his stock into FBA before Black Friday 2024. The minimal-split inbound deadline of October 20 passed with two truckloads still on the road, and Amazon simply refused them. Those trucks sat at a 3PL warehouse for three weeks while his Amazon listing stocked out for four days during Black Friday. His ads kept running. ACoS jumped from 18% to 31%, and his ranking fell two pages. The 20% he'd kept back for his own store, meanwhile, shipped same-day from his own warehouse and caught every order. He changed his rule after that season: never put all the eggs in one basket. (Scenario illustration: figures are illustrative.)

Multi-Platform Inventory Allocation Starts With Three Piles

My method is crude, but it works. Split peak-season stock into three piles: Amazon FBA, Walmart WFS, and your own warehouse, which feeds your independent store and covers emergency reallocations. Don't guess the split. Measure it with three rulers: each platform's share of your sales over the past 90 days, each platform's inbound lead time, and each platform's peak-season storage cost.

Sales share is the baseline. If your last 90 days were 55% Amazon, 25% Walmart, and 20% your own store, start there. Lead time is the correction. Amazon inbound means queues and appointments. For Black Friday 2025, Amazon's published inbound deadlines were October 20 for minimal-split shipments and October 30 for optimized splits — check the latest dates against current platform policy before you plan. Miss the date and your freight doesn't get in, no matter when it left your dock. Walmart and your own warehouse move faster, so their share can flex upward.

Storage cost is the third ruler, and the one most sellers ignore. Amazon's standard-size storage runs $0.78 per cubic foot per month from January through September, then jumps to $2.40 from October through December — three times the off-peak rate. Walmart WFS charges $0.75 over the same period, rising to $2.25 only for stock held more than 30 days in October through December. The same pallet sitting one extra month at Amazon can cost two to three times what it costs at Walmart. Confirm all rates against the platforms' official pages, since they change.

Peak Season Safety-Stock Math for Multi-Platform Allocation

Once the split is set, safety stock per pile has to be calculated, not felt. My formula is three lines: safety stock = daily sales × replenishment lead time in days × volatility factor. Daily sales means your 30-day average. Lead time means the real inbound time per platform — 14 days minimum for FBA in peak season. The volatility factor depends on your product: 1.2 for steady sellers, 1.5 for seasonal ones.

Here's a worked example with a hypothetical SKU selling 100 units a day:

Platform Daily sales (hypothetical) Lead time (hypothetical) Safety stock (hypothetical) 60-day peak stock (hypothetical)
Amazon FBA 55 units 14 days ~920 units ~3,300 units
Walmart WFS 25 units 10 days ~300 units ~1,500 units
Own warehouse 20 units 3 days ~70 units ~1,200 units

See the problem? FBA safety stock is the most expensive because its lead time is the longest. Here's the counterintuitive part: in peak season, don't stuff everything into Amazon. FBA peak storage costs three times the off-peak rate, so every extra day your stock sits there is money burned. The smarter play is keeping the bulk in your own warehouse or a 3PL and topping up FBA in waves as sales dictate. Your own warehouse ships flexibly and reallocates the same day. That's your real safety valve.

What a Stockout Costs on Each Platform

A stockout isn't just "a few slow days." The cost is different on every platform. On Amazon it hurts the most: your listing rank drops while your ads keep spending, and climbing back costs roughly double the ad spend. On Walmart the hit is softer, but a new seller's hard-won ranking weight drops just the same. On your own store, a stockout costs you repeat customers — someone who finally got into the habit of ordering direct hits one "out of stock" and goes straight back to Amazon next time.

So my triggers are fixed numbers, not gut feelings. FBA cover below 14 days while your own warehouse has stock: reallocate immediately. Walmart cover below 10 days: raise the replenishment order that same day. Own-store cover below 7 days: pause ads and send traffic to the platforms that still have stock. Write the triggers on a sheet, stick it on the warehouse wall, and whoever is on shift follows it.

Pitfalls I've Already Stepped In

Don't cut inbound deadlines close. Amazon's deadline is the day freight arrives at their dock, not the day it leaves yours. Three days stuck in traffic and the date is gone. My rule: freight ships at least 14 days before the deadline.

Don't bet every bestseller on one platform. I watch this happen every year: a seller puts a hot SKU entirely in FBA, capacity gets cut, the freight can't get in, and Walmart and the independent store have nothing. A whole Black Friday week with no stock to sell.

And don't wait for a stockout to reallocate. Reallocating on the day you run out means another week in FBA's inbound queue. Multi-platform inventory allocation for peak season is not a spreadsheet exercise. It's a weekly habit. My routine: 20 minutes every Monday morning, checking sellable days on all three platforms, and the moment a trigger trips, I move stock. Surviving peak season without a stockout isn't luck. It's discipline. This one I've tested myself, and it works.

By Ah K, E-commerce Seller (5 yrs)