Just in case inventory: when holding buffers beats running lean

Just in case inventory (JIC) is the deliberate decision to hold more stock than current demand requires, so that surprises — supplier delays, demand spikes, freight disruptions — never reach your customers. It's the honest counterpoint to just-in-time ordering: where JIT treats inventory as waste to be minimized, JIC treats dark bestsellers as the waste to be minimized. Neither is right everywhere. The useful question is which errors cost you more on each of your supplier lanes, and this article gives you a fresh worked comparison that answers it with arithmetic instead of philosophy.

What JIC actually means (and doesn't)

JIC is not "keep everything fully stocked." It's a stocking policy sized to a stated tolerance for disruption: you quantify the shocks you're willing to absorb — a supplier running N days late, a demand week at 1.5× forecast — and you hold enough buffer to survive them. Everything beyond that tolerance is dead-weight capital, and JIC done properly refuses to buy it.

Three things JIC is often confused with:

• Par levels. A fixed "never drop below 20 units" rule is a habit, not a policy, until the 20 is derived from your lead time and demand variability. JIC is the derivation; sizing the buffer is the math.
• Hoarding. Buying a year of stock "just in case" of price increases is a different decision — a bet on input costs — and it usually costs more in carrying cost than it saves.
• The opposite of JIT. In practice they share machinery: JIC lanes still use reorder points and PO discipline. The difference is what the reorder point covers — only expected demand (JIT-flavored) or expected demand plus a sized buffer (JIC-flavored).

The lane that decides it: a worked comparison

Take a fresh example. You sell fishing lures and camping stoves from a Shopify store. Two supplier lanes:

• Lures: local tackle manufacturer, 9-day lead time, historically ±2 days. Steady demand on the core range, 3.2 units/day on your bestseller.
• Camp stoves: overseas factory, 42-day quoted lead time, and your last three orders landed at 42, 51, and 58 days (POs INV-2027-014, INV-2027-021, INV-2027-029 — measure your own PO history rather than trusting quotes, per lead-time discipline). Demand 2.1 units/day, spikier — campfire season triples it for six weeks.

The variable that decides whether JIC or lean wins on each lane is the same one: how bad is the realistic worst replenishment delay? So run the sensitivity. Assume a shock where demand runs hot (1.5×) during the same window the shipment slips. Unit cost $2.10 for lures, $19 for stoves; retail $9.99 and $64. Carrying cost assumption: treat stock as costing roughly 15–30% of its value per year to hold (capital, space, obsolescence — treat this as an assumption for your own business, not a law).

Lane 1 — lures, 9 ± 2 day lane. Your p80 lead time is ~11 days. A lean order covers expected demand over 11 days: 3.2 × 11 = 35 units. A JIC order adds a buffer for the worst plausible shock — a 5-day slip plus 1.5× demand for two weeks — which is about 30 units. The buffer costs 30 × $2.10 = $63 of inventory value, maybe ~$13/year in carrying cost at 21%.

Lane 2 — stoves, 42–58 day lane. Your measured spread already spans 16 days. Expected coverage over a p80 of ~55 days: 2.1 × 55 = 116 units. A lean policy that ordered against the 42-day quote would arrive, in a bad cycle, up to 16 days short: 2.1 × 16 = ~34 units of unmet demand, and if a season start overlaps the slip, 1.5× demand turns that into 50 units — **$3,200 of retail** you can't serve, in your best six weeks of the year. A JIC buffer sized to the 16-day spread plus hot demand is roughly 55 units: 55 × $19 = $1,045 of extra inventory value, ~$220/year carrying cost.

The cost-of-error table

Put the two errors side by side. "Lean error" = you held too little and got hit. "JIC error" = you held buffer and demand stayed flat, so you paid carrying cost for nothing.

The pattern generalizes: the longer and more variable the lane, the cheaper the buffer looks relative to the failure it prevents. On short, tight lanes the logic reverses — buffer mostly insures against delays that don't happen. That asymmetry is why experienced operators run different replenishment methods on different lanes instead of one store-wide philosophy.

Two refinements the table hides:

• The lean error isn't one-time. Every shopper who bought a stove elsewhere is a data point in their purchase history — the classic out-of-stock finding from the 2002 study commissioned by the Grocery Manufacturers of America showed 31% of shoppers bought elsewhere and only 9% bought nothing (dated, accessed October 2026). Some of that demand doesn't come back next cycle. Lean errors are partially permanent; JIC errors fully refund.
• The JIC error shrinks with rotation. The buffer on a product that sells year-round isn't parked — it turns with the cycle. $1,045 of stove buffer is not $1,045 spent; it's float that moves. Dead-weight buffer only occurs on products whose demand you misread, which is why buffers belong on proven sellers, not hopeful ones.

Where JIC clearly wins for a small store

• Long or variable lanes. Anything over ~3 weeks quoted, or whose actual arrivals spread wider than the quote.
• Hero SKUs. The handful of products that carry the store. A buffer on a bestseller is priced against the whole store's revenue, not that SKU's margin.
• Seasonal peaks on long lanes. You can't chase mid-season on a 45-day lane; the only defense is having bought ahead — which is JIC by another name (anticipatory stock).
• Single-supplier products. Until you've built a second source, the buffer is the only thing standing between a factory hiccup and a dark category page.

Where lean (or near-lean) wins

• Local, fast, boring. Short lanes with tight variance; waiting a week costs little and substitute purchase is easy.
• Slow tail. If a SKU sells 0.4/day, a buffer sized for shocks sits for months. That's the tail where carrying cost quietly out-eats the risk it insures — and where the answer is often "less stock," not "more."
• Unproven products. JIC's precondition is known demand. Buffering a launch is guessing with extra steps; buy small, read the data, then decide.

Operating JIC without turning into a warehouse

The failure mode of JIC isn't the buffer — it's the drift. Buffers set once and forgotten grow year over year. Keep it disciplined:

1. Size each buffer from your own numbers: measured lead-time spread (p80, not quote) × daily demand × the shock window you choose to tolerate. The safety stock method is the formal version.
2. Re-derive per lane, quarterly. Lead times move; your PO history will tell you before your gut does.
3. Write the tolerance down. "This buffer survives a 16-day slip at 1.5× demand" is a policy. "Extra two weeks of stock" is a mood.
4. Watch the JIC error line. If a buffer goes unused for two full cycles on a non-hero SKU, shrink it — that's the discipline that keeps JIC from becoming hoarding.

One tooling note, honestly framed: most small-store tooling assumes a single fixed reorder point and doesn't distinguish lanes. We are building InventoryX as a read-only layer over Shopify that computes days of cover against your manually entered lead times and produces a weekly buying list — it won't size buffers for you, but it makes the lane reality (what you ordered, what arrived late, what ran dark) visible enough to set them by hand. In development as of October 2026, and it never touches your stock.

Common questions

What is just in case inventory? A stocking policy that deliberately holds buffer stock beyond expected demand, sized to absorb defined disruptions — supplier delays, demand spikes, freight problems. Unlike just-in-time ordering, which minimizes inventory on the assumption replenishment is fast and reliable, JIC assumes surprises will happen and pre-pays for them in inventory value instead of in missed sales.

What's the difference between just in case and just in time inventory? JIT orders close to need and holds almost no buffer, saving carrying cost but exposing you to every delay; JIC holds a sized buffer and absorbs delays at the cost of carrying it. JIT's economics win on short, reliable lanes; JIC's win on long, variable ones. Most small stores should run both, assigned per supplier lane — the full lean argument is covered in our just-in-time article.

How much buffer stock should I hold? Enough to cover your realistic worst replenishment window: measured p80 lead time minus quoted lead time, times daily demand, plus allowance for demand running hot during a slip. Derive it from your own PO history rather than a fixed "two extra weeks" habit, and review it quarterly — buffers set once and forgotten quietly become hoarding.

Is just in case inventory wasteful? Only when it's unsized. A buffer on a proven bestseller on a long lane is the cheapest insurance on the shelf — in our worked example, ~$220/year of carrying cost against ~$2,200–3,200 of retail that a single bad cycle would strand. The waste case is buffering slow movers or unproven products, where the buffer sits unsold and carrying cost out-eats the risk.

Does JIC mean I should buy in bulk? No. Bulk buying is a unit-cost decision; JIC is a disruption-tolerance decision, and they're sized differently. A buffer sized to a 16-day lead-time spread is usually a few weeks of demand — not a case quantity negotiated for a price break. Conflate the two and you end up holding a quarter's worth of stock to solve a pricing problem.

Next step

Sort your supplier lanes into the two rows of the table above: fast-and-tight versus long-and-spread. For every long lane, pull your last three PO arrival dates and compute your real spread before you place the next order — then size the buffer from that number with the safety stock method instead of the quote. On a 40+ day lane, that one correction usually pays for the entire buffer in a single avoided bad cycle.