Never out of stock: building the core list that keeps your bestsellers buyable
Never out of stock starts with a core list, not faster reordering. See the four-part test that keeps bestsellers buyable, with a worked example.
InventoryX is in development and not available today. ClearStock is the live Shopify app.
A Friday promo email goes out for your best-selling dog harness. By Saturday afternoon the medium is gone; by Sunday night the whole style is; the promo still has two days to run. Shoppers who hit the empty product page don't wait around — most buy somewhere else, and the ones who switch brands rarely come back when your restock finally lands. The fix isn't reordering faster next time. It's deciding in advance that a short list of products will never go out of stock — and organizing your buying week around defending that list.
The apparel trade calls this a never-out-of-stock program — "NOOS" on wholesale order sheets. In plain words: a core list. A handful of SKUs you commit to keeping buyable every week of the year, replenished on a schedule instead of on feel. Everything else in the catalog competes for cash and attention; the core list is exempt from that competition. This article is about building the list: which SKUs earn the commitment, what you buy for them, and the discipline that keeps the promise intact.
One boundary first. How you compute the actual trigger — reorder points, par levels, calendar top-ups — is a method choice, and the five replenishment methods compared side by side is its own article. The core list sits upstream of all of them. It decides which SKUs get a protected trigger; the methods decide when that trigger fires.
The four-part test for core-list candidates
Not every bestseller deserves the commitment. A core-list SKU ties up cash in every cycle, and the discipline only works if the list stays short. Four tests, all answerable from data you already have:
| Test | The question | A pass looks like |
|---|---|---|
| Demand continuity | Does it sell in every week of the last 12 months, not just in spikes? | No zero-sale weeks in the variant's sales history |
| Predictability | Can weekly volume be forecast from history without guessing? | Week-to-week swings stay inside roughly ±20% (a working rule, not an industry standard) |
| Supplier reliability | Does the supplier hit agreed dates cycle after cycle? | A measured lead time you'd bet a promo on — here's how to measure it from PO history |
| Economics | Does the margin carry the cost of always holding buffer? | The buffer's cash cost is small against the profit the SKU protects |
Three of the four tests are arithmetic. Continuity is a scan of the sales history for zero weeks. Reliability is your PO history, order date to stock-sellable date. Economics is unit margin against the buffer you'd hold. Predictability is the judgment call — and it's the test that disqualifies the most candidates, because plenty of products sell steadily right up until they don't.
What never earns a spot
A few categories almost always fail, no matter how good the sales look:
- Seasonal styles. A pumpkin-spice anything sells "continuously" for eight weeks a year. Continuity means all twelve months.
- Trend pieces. Demand you can't explain from history is demand you can't commit to. When the trend turns, the core list turns into dead stock with a schedule.
- Anything with less than six months of history. Provisional status at most. New products get watched, not committed.
- Closeout or one-time-buy lots. If you can't reorder it, it can't be never out of stock by definition.
The test isn't whether a product is popular. It's whether its demand is continuous and explainable — because the whole point of the list is that you buy it the same way in March as in November.
A worked example: 12 harness SKUs, 3 make the list
Ridgeline Pet Co. sells three harness styles in four sizes — 12 SKUs, one domestic supplier. Twelve months of variant sales:
| Style | Share of harness units | Zero-sale weeks in 12 months | Verdict |
|---|---|---|---|
| Everyday Harness | 71% | 0 (S, M, L); 2 (XS) | S, M, L qualify; XS provisional |
| Summit Print | 19% (86% of it Oct–Dec) | 31 | Fails continuity |
| Reflective Night | 10% | 7 | Fails — only 6 months of history |
Three SKUs make the list. The Summit Print sells more units in its eleven hot weeks than the Reflective Night sells all year, and it's still wrong for the list: you cannot promise continuous availability on a style that legitimately disappears every January.
What the buy math looks like for a core SKU
The Everyday Harness, size M. First input: velocity from in-stock days only — the M was dark 9 days last quarter, and a raw average across those weeks understates real demand (why velocity numbers lie during out-of-stock days).
- True velocity: 6.0 units/day
- Measured supplier lead time: 21 days (domestic, consistent)
- Demand during the wait: 6.0 × 21 = 126 units
- Core buffer: one week of demand = 6.0 × 7 = 42 units
- Reorder point: 126 + 42 = 168 units
- Order quantity: supplier ships in cases of 40 → 3 cases = 120 units (120 ÷ 6.0 = 20 days of stock per cycle)
- Cover at the trigger: 168 ÷ 6.0 = 28 days — the order fires with nearly a month of stock on hand, which is what absorbing a slow delivery looks like
The formula behind those lines is ten minutes of arithmetic per SKU, and the reorder point formula with a spreadsheet calculator walks it end to end. What the core list adds is the rule that this SKU's order fires at 168 units every cycle — not when cash flow feels comfortable, not when you remember.
One Shopify setting belongs in this section: a variant left on "continue selling when out of stock" will happily sell past zero and defeat the entire program (Shopify Help Center, "Selling products when they're out of stock," accessed November 2026). Core-list SKUs should have it off, so the trigger and the storefront tell the truth.
Protecting the list: four rules
The list is easy to build and easy to quietly abandon. Four rules keep it real:
- Core reorders fire on schedule, even in a cash squeeze. That's the commitment. A core item going dark during a promo is the most expensive stock you can not have.
- Promo demand doesn't touch the core trigger. Forecast the promo's units on top of baseline velocity and order them separately. If the promo forecast quietly consumes the core buffer, you've turned your bestseller into a gamble.
- Any dark day on a core SKU is an incident, not bad luck. Run the 30-minute post-mortem for SKUs that ran out and fix the input that failed — lead time, trigger level, or discipline.
- Re-qualify quarterly. Last quarter's trend piece earns a spot only if it passes all four tests on fresh data. Supplier reliability deserves the closest look, because it can collapse in a single cycle.
The tooling ground truth
As of October 2026, the core Shopify admin has no built-in low-stock email alert — saved low-stock views and Shopify Flow recipes are the native options (Shopify Help Center). For a program built on "never dark," that means the watching is on you.
The obligatory tooling disclosure: InventoryX is intended to send a Monday Buying List that sorts your tracked SKUs by days of cover and flags which ones will run dry first — the free tier covers your 50 fastest-moving SKUs, and $9.99/mo Starter expands it — useful for defending a core list week after week without a spreadsheet session. What V1 won't do is choose the list for you: there's no ABC intelligence in the app, so the four-part test stays yours. We'd rather leave the judgment where it belongs.
Common questions
What does NOOS mean in retail? NOOS stands for "never out of stock" — a replenishment program where a core set of styles is kept continuously buyable and replenished on a fixed schedule. In practice it means a short protected list: those SKUs reorder on schedule regardless of cash timing, while the rest of the catalog buys more opportunistically.
How many SKUs should a never-out-of-stock program cover? Start smaller than feels comfortable — often 10 to 20 SKUs. The list earns its keep through discipline, and discipline gets harder with every item added. A working rule: any SKU on the list should pass all four qualification tests and move enough weekly volume that going dark actually costs money.
Is a core list the same as a reorder point? No. The reorder point is trigger math — the stock level that fires an order. The core list is the commitment upstream of it: which SKUs get a protected trigger, order on schedule, and are never allowed to go dark. Every core-list SKU needs a reorder point; most reorder points in your catalog sit outside any core list.
What happens if a core item's supplier fails? Treat it as an incident: count days of remaining cover, then bridge — expedite a partial shipment, add a second source for that SKU, or capture demand with a waitlist instead of taking orders you can't ship. Afterward, re-run the supplier reliability test honestly before the SKU returns to the list.
How often should I re-qualify the core list? Quarterly is the working rhythm, plus after any demand shock — a viral moment, a sustained ad campaign, a price change. Continuity and predictability drift slowly, but supplier reliability can change in one bad cycle, so check the qualification table against fresh data every quarter.
Next step
Run the four-part test on your single best category this week — twelve months of variant sales, your PO history, and the table above. Whatever passes, compute its reorder point tonight using the reorder point formula: the arithmetic takes ten minutes per SKU, and it's the difference between a commitment and a wish.