Selling the same stock on your own store and a marketplace? Enter each product's sales per channel and how often your counts update. The calculator tells you how many units to hold back on each channel so the last unit never sells twice.
Your numbers. Your buffer. No signup — nothing leaves your browser; your inputs are saved on your device only.
The first row is an example with an invented product so you can see the shape of the answer. Overwrite it or remove it.
2. Channels that sell this stock
Tick separate stock pool for a channel that ships from stock the marketplace holds for you (Amazon FBA, Walmart WFS). Your other channels cannot oversell that stock, so the channel gets no buffer and adds nothing to the others' exposure.
3. Products: sales per channel
Per channel: the average units sold per day and the best single day in the last 90 days. The buffer is sized to the best day, because oversells cluster on busy days.
Header row: product, channel, avg_per_day, best_day (a marketplace_fulfilled column of yes/no is optional). Channel names: Own store, Amazon, Walmart, eBay, Etsy, TikTok Shop, Other. Importing replaces the rows above. The file is read by your browser and never uploaded.
Results
Everything on this page runs in your browser. Your sales numbers are never uploaded — inputs save to your own device, and a share link carries them only to the people you send it to.
Holding units back covers a gap between counts that each update on their own schedule. OmniOrders keeps stock counts in sync across your channels, with low-stock warnings colour-coded in the inventory list. The calculator above works with any update method, including that one.
Start Free TrialAn oversell happens in the gap between a sale on one channel and the moment every other channel shows the lower count. During that gap the other channels keep selling stock that is already gone. When the last unit sells twice, one of those orders gets cancelled, and the marketplace counts the cancellation against you: Amazon's published target for the cancellation rate on seller-fulfilled orders is under 2.5%, and breaching it can lead to deactivation of seller-fulfilled offers. Your own Account Health dashboard shows the current line.
A buffer holds back enough units on each channel to cover what the other channels could sell before the count catches up. Two things set its size: how fast the other channels sell on their busiest day, and how long the gap is. For each channel that sells from the shared pool, this page calculates:
exposure = (best-day units on every other shared-pool channel) ÷ 24 × gap in hours buffer = exposure, rounded up to a whole unit
Worked example (an invented product; the numbers are made up to show the contrast). The Trailhead 32oz bottle sells on the seller's own store, 4 a day on average with a best day of 10, and on Amazon, seller-fulfilled, 6 a day on average with a best day of 15.
Same product, same sales. The only thing that changed is how long the other channel keeps selling before it hears about the sale. Whether 25 units is a lot depends on your margins and how much stock you carry, which is why the tool shows the units held back rather than a verdict.
The model is deliberately simple so you can check every number by hand. Four things it leaves out:
A channel buffer is a number of units you hold back on a sales channel: you have them in stock, but that channel's listing shows fewer. The held-back units cover what your other channels could sell in the gap before this channel's count is updated, so the same last unit is not sold twice.
Because the buffer only has to cover the gap. If counts are updated by hand once a day, a sale on Amazon can stay invisible to your own store for up to 24 hours, and the store keeps selling stock that is already gone. If an app updates counts every 15 minutes, the same sale is invisible for at most a quarter of an hour. Same sales, a gap 96 times shorter, so a much smaller buffer. The calculator uses the gap you enter; it does not assume one.
Oversells cluster on busy days. A buffer sized to an ordinary day is too small exactly when it matters: during a promotion, a feature or a seasonal spike. Using each channel's best day in the last 90 days sizes the buffer for the day the last unit is most likely to sell twice. The average stays in your inputs so you can see how far above it your peak sits.
Stock sitting in a marketplace's own warehouse is a separate pool: only that marketplace sells from it, so your other channels cannot oversell it and it cannot oversell them. Tick "separate stock pool" for that channel and the calculator gives it no buffer and leaves it out of the other channels' exposure. If the same marketplace also sells seller-fulfilled units from your own warehouse, treat that as a shared-pool channel.
Sales you could have made. Every unit held back on a channel is a unit that channel cannot sell until your count updates, which is why the calculator shows the total units held back and not just the buffer. In this model the only way to shrink the buffer without accepting more oversell risk is a shorter gap between count updates.
No. The form, the math and the CSV export all run in your browser. Your inputs are saved to your own device (localStorage) so they are there when you come back, and clearing your browser storage removes them. The share-link button encodes your inputs into the link itself, so only the people you send it to see them. Nothing is uploaded to OmniOrders.
They answer different questions. Safety stock and a reorder point tell you when to order more from your supplier so you do not run out. A channel buffer decides how much of the stock you already have each channel is allowed to see, so two channels do not sell the same last unit. Most multichannel sellers need both; the free reorder point calculator on this site covers the first.
