How to set a safety stock level on Shopify

It is the number you refuse to sell past. Safety stock is stock you physically hold and deliberately keep back, so that the things you cannot predict land on the buffer instead of on a customer: a delivery that slips a week, a product that suddenly sells five times its usual rate, two shoppers checking out the last unit within the same few seconds. That last one is overselling in its purest form, and it is the failure a floor exists to absorb.

It costs you something, and it is worth being honest about that. Twenty units held back on a $40 product is $800 of stock you cannot turn into money today. What you buy with it is the expensive problem going away. A sale you never took costs you that margin once. A sale you took and could not ship costs you the refund, the apology thread, and often the customer.

Multiply what it sells in a day by the number of days a reorder takes to reach your shelf. That is the whole starting formula, and it works because the buffer only has to keep you selling while the replacement is in transit. A product shifting four a day with a five-day restock lands on twenty.

Both figures come out of your own admin. For daily sales, open Analytics, then Reports, then Sales by product, set the range to the last 30 days and divide the units sold by 30. For restock days, count from the day you send the order to your supplier to the day the boxes are open and counted in, which is nearly always longer than the number the supplier quotes.

Then adjust for how much it hurts to get it wrong. Push the number up on anything where the supplier misses dates, where demand spikes without warning, or where the product is the reason people came to your store in the first place. Pull it down on anything you can reorder in two days from someone local.

Run it end to end on one product. A candle selling 62 units in 30 days is a shade over two a day, and a supplier who is reliable at ten working days is really fourteen calendar days, so two times fourteen gives 28. Against a shelf of 40 that feels steep, so 15 is a sane place to start and raise later. Starting low and raising is the right direction to be wrong in, because a number set too high parks stock you could have sold, while a number set too low quietly does nothing at all.

If you would rather not do this product by product, Easy Stock Buffer works it out for you on Pro. It reads how much each product has sold over a window you pick, 30 days by default, multiplies the daily rate by how many days of cover you want, 3 by default, and rounds up. Set cover to 14 and you get the lead-time answer above.

Products with fewer than three sales in the window still appear, flagged as a thin read rather than a confident one, because a suggestion built on two sales is a guess with a decimal point on it.

One caveat before you trust the suggestions on day one. They are built from stock movements the app has watched since you installed it, so a fresh install has nothing to go on and says so. Give it a fortnight of trading and the numbers are worth taking.

No. Your Shopify inventory count does not move. Easy Stock Buffer never writes to your quantities, so a product holding 100 still reads 100 in your admin, in a CSV export, in a stock count and to whoever syncs your warehouse. The only thing your number changes is how far down the storefront is allowed to sell.

That is the difference between a buffer and the two workarounds most people try first. Editing the count down to 80 to fake a reserve hides the stock from you as well, and the next sync from a fulfilment partner writes 100 straight back over the top of it. Shopify's own Safety stock state genuinely moves the units, so they show as Unavailable and your available count drops, which is fine as a one-off split of a shelf and useless as a rule, because nothing recalculates it as your sales rate changes.

Inside the app you see both figures next to each other. The Thresholds screen has a column for the number you set and a column for what the product currently holds, plus a status telling you whether it is sitting on the line right now. Pro adds a Reports screen that counts how many products are protected, how many are at their number today, and how many low-stock emails went out over the last 30 days.

The most specific one. A number typed onto the variant itself beats a bulk rule covering its collection, tag or product type, and that in turn beats your shop-wide default. A product matched by none of the three is simply not held back.

This is what makes a shop-wide default safe to switch on. Put 3 under the whole catalogue, add bulk rules where a category needs more, then override the ten products that need their own answer. You are never fighting a rule you set last month, because the narrower setting always wins the argument.

The shop-wide default is on the free plan. Bulk rules by collection, tag or product type are Pro, as is importing numbers from a spreadsheet, which takes up to 5,000 rows in one go.

They keep their cart and they cannot finish the order. Stock can fall below your number while somebody is still browsing, so the cart is deliberately left alone: on the cart page and in the cart drawer they get a message telling them the item is being held back, and they can edit the basket rather than being thrown out of it. The order stops at checkout.

If part of what they asked for is still above your number, they are told how many they can have instead of being refused outright. A cart asking for four when two sit above the line gets "Only 2 of Blue Hoodie can be ordered right now". Both that wording and the fully-held-back version are yours to rewrite under Settings, then Customer messages, in your own voice and in every language you sell in.

The Add to cart button stops at your number as well, which catches most shoppers before they have built a basket at all. That stop and the cart message both live in your theme, so both need the app switched on in your theme editor, and both are included on the free plan.

Yes, and that is exactly why the stop happens at checkout rather than only in your theme. Express buttons skip your product page and your cart, so anything hidden with theme code is advice a shopper can walk straight past. This check sits inside Shopify's own checkout, so it holds whichever button they pressed, including Shop Pay, PayPal, the Shop app and a saved checkout link someone bookmarked.

One step is manual and it is the one people miss. The rule has to be switched on once in your Shopify admin under Settings, then Checkout, and until it is, nothing stops. The app says so on its home screen while it is still off, and turning on checkout blocking walks through the clicks.

On slow movers it mostly parks money. A product selling one unit every six weeks does not need three held back, because a low-stock email will reach you long before anyone races you for the last one. Leave the number off those and let the alert do the work.

On anything made to order, printed on demand or sold as a preorder, a number blocks the exact sale you are trying to take. Leave those alone, along with any product where you have deliberately switched on Continue selling when out of stock, since the two settings want opposite things. That deliberate case has its own page: how to set up backorders on Shopify.

Variants with Track quantity switched off are skipped automatically, and this is worth knowing before you wonder why. Shopify reports a quantity of zero for an untracked variant, so a number on one would block it permanently instead of protecting anything. The app leaves them out by default and the switch controlling that sits under Settings, then Enforcement.

It is also the wrong tool when what you want is a cap on how much of something one customer can buy. That is a different job with a different setting, and order limits against safety stock sets the two side by side, because they get mistaken for each other constantly.

There is also a version of this where you need no app at all. If your counts only ever go wrong because a handful of variants still have Continue selling when out of stock switched on from a launch, fix those and stop there. A buffer earns its keep on the gap between what Shopify thinks you hold and what is physically on the shelf. No gap, no reason to pay for one.

Unblock the sale first and diagnose afterwards. Pro has a pause that switches every number off for one, six or twenty-four hours in a single click, which is what you want when a delivery has physically landed and nobody has counted it in yet. It expires on its own, so there is no setting left quietly switched off next month.

On the free plan the equivalent is opening that product and lowering its number, or clearing it. Either way the next checkout uses the new figure immediately, and putting the old one back is the same two clicks.

Then find the cause, because it is nearly always one of three. Your shelf and your count disagree, so recount and set the on-hand figure from what is physically there. Or the number is too high for the way that product sells now, which the suggestion screen will tell you in seconds. Or the stock is sitting at a location that does not serve your website, which needs a per-location number on Pro and is covered in how to prevent overselling on Shopify.

Alerts are the early-warning half of all this, and they are what stop the day happening in the first place. Every product that reaches its number can email you, ten a month on the free plan and unlimited on Pro, where you can take one daily or weekly summary instead if per-product emails turn into noise.