Home › Blog › Raise your low stock thresholds before Q4, not during it

Most merchants set a low stock threshold once, usually on the day they install an alerting app, then never touch it again. That is fine for ten months of the year. It stops being fine somewhere around the middle of October.

A threshold is not a stock level you like. It is a deadline. It says: when this variant reaches this quantity, there is still just enough time to reorder and receive before the shelf empties. Both halves of that sentence depend on how fast the variant is selling and how long your supplier takes.

Q4 breaks both halves at once.

Why the summer number fails in November

Say a variant sells 2 units a day and your supplier reliably delivers in 14 days. 28 units is the point of no return, so a threshold of 35 gives you a week of slack. That is a sensible number in July.

In late November the same variant sells 8 units a day. That same 14 day supplier now needs 112 units of cover, not 28. Your alert at 35 units does still fire. It fires about four days before you run out, with 10 days of waiting still ahead of you. You get the email, you place the order, you sell out anyway.

The alert was not wrong. It answered the question it was asked. The question was stale.

Delivery times move against you at the same moment. Your supplier is handling everyone else's Q4 order too, carriers are jammed and a 14 day promise made in June is often a 21 day reality in November. So the amount you need to cover goes up and the time you need to cover it goes up as well.

The number to use instead

Work it the same way you worked the original, with two substitutions.

Use your peak daily sales rather than today's. If you have last year's numbers, use the busiest week you actually had, not the monthly average. A monthly average flattens out exactly the days that empty your shelves.

Use the worst delivery time you have realistically seen rather than the one you were quoted. If a supplier has ever taken 21 days, 21 days is your planning number.

Multiply the two, then add whatever buffer you were already carrying for the weeks that go wrong. That total is the point where reordering still works. Your threshold goes above it, not on it, because you want to hear about it before the deadline rather than on the day.

For the variant above that is 8 a day times 21 days, so 168 units of cover, with a threshold somewhere above that depending on how much slack you want. It is a far bigger number than 35. That gap is the entire point.

When to change it

Change it before the sales rate changes, not after.

That sounds obvious written down. In practice almost everyone does the opposite, because the thing that makes people review a threshold is usually a stockout and a stockout happens after sales have already moved. By then the reorder is already late.

Work backwards from your first heavy sales day and subtract your worst delivery time. That is the date your higher threshold needs to already be switched on, because an alert firing that day needs the full delivery time behind it to be any use. For most merchants selling into Black Friday, that puts the change in the second half of October rather than in November.

Two practical notes.

Do it per variant, not per store. One threshold across a whole catalog where some SKUs sell one a week and others sell 50 a day is a number that is wrong for almost everything in it. If your alerts only allow one store wide value, that limit costs you more in Q4 than in the whole rest of the year.

Write down what you changed each one to and why. In January you will want to put them back and the version of you reading that spreadsheet will not remember which numbers were seasonal and which were fixes you meant to keep.

Putting them back

Putting them back is less urgent, but it is not free. A threshold left at its Q4 level through a quiet February fires constantly on variants that are in no danger at all. An alert that fires when nothing is wrong is an alert your team stops opening. That is how a working setup turns into noise. Not because it broke, but because nobody turned it back down.

Set a calendar reminder for the second week of January. It takes ten minutes and it protects the thing that makes the alerts worth having, which is that people still read them.

Know before a SKU hits zero

Stockwell watches every variant at every location and emails your team when stock falls under the level you set. Install it from the Shopify App Store and the first alert is running in a couple of minutes.

Get it on the Shopify App StoreFree plan covers 50 SKUs with a daily summary. Paid plans add instant alerts and more recipients. No card needed to start.