Ready to take your Amazon Vendor data ops to the next level?
Request access today and let InsightLeap help grow your business!
November 10, 2026 by InsightLeap
You're planning production for the next few months, and the POs Amazon sends keep landing above or below the number you planned against. Amazon publishes its own weekly demand forecast for every ASIN you sell it, and it sits in Retail Analytics, so you can plan against Amazon's view of demand instead of reacting to each PO as it arrives.
This guide covers where to find the forecast in Vendor Central, what the Mean, P70, P80, and P90 figures tell you, what the report leaves out, and how to use it for production planning, PO conversations with your vendor manager, and Annual Vendor Negotiation prep.
The forecast report is Amazon's estimate of future weekly customer demand for each of your ASINs, built from historical demand, profitability, and vendor lead times. Because it's Amazon's own number, it's the closest you get to seeing how Amazon expects your catalog to sell.
The report runs weekly figures for a stretch of weeks ahead. Published guides disagree on how far out it goes, with figures ranging from 24 to 48 weeks, so check the horizon on your own download before you build a plan that depends on the far end.
Profitability and lead time both feed the forecast, so read an ASIN's numbers alongside its Net PPM and your vendor lead time for it.
In Vendor Central, Amazon is the retailer, so this report is Amazon's forecast of its own sales to shoppers. It measures sales out rather than the units Amazon plans to order from you, which is why it's useful for planning and also why it won't match your POs one for one. The Seller Central to Vendor Central guide maps the other reports that moved.
The Retail Analytics glossary entry and the Retail Analytics how-to cover the dashboard itself, and the export guide covers downloads.
The report gives four forecasts per ASIN per week:
The three P figures climb as the probability rises. P70 is the lowest of the three, and demand is expected to run above it roughly three weeks in ten, while P90 is the high end, with demand expected to exceed it about one week in ten. The gap between P70 and P90 shows how uncertain Amazon is about the ASIN, so on a narrow gap the band you plan to barely changes the number, but on a wide one the wrong choice leaves you with excess stock or missed orders.
| Forecast | What it says | When to plan to it |
|---|---|---|
| Mean | Amazon's best estimate | Tracking the trend for an ASIN week over week |
| P70 | Demand exceeds it about 3 weeks in 10 | Slow movers, items that are expensive to hold, items near end of life |
| P80 | Demand exceeds it about 2 weeks in 10 | A middle setting for steady core items |
| P90 | Demand exceeds it about 1 week in 10 | Items where running out costs you more than carrying extra stock, and items with long lead times |
Two limits matter before you plan against it.
Because the forecast is sales out, the number that matters to you as a vendor is how much of that demand Amazon still needs to buy from you. Here's a rough way to estimate it per ASIN:
Take a hypothetical ASIN with a four-week lead time. Its P80 adds up to 1,200 units over those four weeks, Amazon holds 500 units on hand, and 300 more are on open POs, which leaves about 400 units of demand you'd expect future POs to cover. If the next POs come in well below that, the gap is worth raising with your vendor manager. Treat the result as a planning estimate, and track weeks of coverage for the same ASIN as a second read on whether Amazon is running lean.
If Amazon is a large share of your distribution, its forecast belongs in your manufacturing forecast. Set a band per ASIN using the table above, pull the report every week, and compare this week's figures with last week's. When a forecast has been sliding for several weeks, talk to your production planner while there's still time to change what you build.
Vendors use the forecast to request manual PO adjustments from their vendor manager. When a PO arrives well below what the forecast and your overlay suggest, bring the specifics: the ASIN, the weeks, the band you planned to, Amazon's on-hand units, and the open POs. Bring your promotion calendar too, since the forecast can't see it. For handling the POs themselves, see how to accept purchase orders.
When Amazon consistently orders below its own forecast, that can signal a profitability concern, and it may mean your trade funding needs a fresh look. Keep a running record per ASIN of the forecast against what was actually ordered and read it alongside Net PPM, so you go into AVN with the evidence in hand. The AVN preparation guide covers the negotiation itself.
The forecast report gives you Amazon's sales-out view, and the overlay with your POs and shipments is yours to build. Doing it by hand means saving each week's download and joining it to your PO and sales data by ASIN, every week, and seeing how the forecast has moved over the months means keeping every one of those files. InsightLeap keeps Amazon's forecast history per product and sets it next to what actually shipped, with on-hand and open PO units alongside, so the trend and the overlay are ready when you sit down to plan.
Pull the forecast weekly, choose a band for each ASIN, overlay it with on-hand and open PO units, and take the gaps to your vendor manager while the next PO can still change.
Request access today and let InsightLeap help grow your business!