Reorder point after a marketing push: recalibrate before scaled ads break it

Your reorder points didn't stop working. The velocity they were built on stopped being true — the same day you raised the budgets. A reorder point is just history multiplied by your lead time, and a spend step-change rewrites history overnight while the threshold sits still. The result is a specific, predictable failure: the threshold fires late, the lead time eats the difference, and your bestsellers go dark in the middle of your most expensive traffic. Here's the arithmetic of exactly how it breaks, and the recalibration routine that keeps thresholds honest when spend moves.

Why static thresholds break when spend scales

The reorder point formula (velocity × lead time + safety stock — the full formula walkthrough covers how it's built) has exactly two moving parts. Lead time lives with your supplier and doesn't care about your ad account. Velocity is the part you changed. When you scale spend, three things happen to it:

• Level: demand during the lead time rises with the new traffic, so the "demand during the wait" the threshold was sized to cover is understated.
• Volatility: scaled campaigns shift the demand distribution — creative winners spike it, auction costs reshape it — so safety stock sized on the old variability is also undersized.
• Composition: new-customer demand lands on different SKUs than your organic mix, so the velocity change is uneven across the catalog.

None of this is subtle at the SKU level. It's invisible at the dashboard level, which is why it keeps happening: the store looks fine, the ROAS looks fine, and the reorder point silently belongs to last quarter.

A worked scenario: the week you 2.5× your budget

Illustrative numbers, but the shape is common. You sell phone stands — say the matte-black one is your top seller. Your last 90 days of in-stock selling averaged 11 units/day. Your supplier delivers in 14 days. You carry 40 units of safety stock.

Reorder point as set: 11 × 14 + 40 = 194 units.

On November 3 you raise prospecting spend from $85/day to $210/day. By day 7, blended velocity on the SKU is running 17 units/day — a 55% lift, plausible when prospecting adds mostly new-customer volume to a hero SKU. The reorder point hasn't moved. Trace what happens:

• Lead-time demand at the new velocity: 17 × 14 = 238 units. Your threshold was sized for 154.
• Cover when the threshold fires: the moment stock hits 194, you have 194 ÷ 17 ≈ 11.4 days of cover against a 14-day lead. You go dark for roughly 2.6 days on an average cycle — before any supplier slip.
• The corrected threshold: 17 × 14 + 40 = 278 units, and even that safety stock is stale, sized on old variability. Bump it deliberately — say to 60, a number you justify with safety stock math rather than a hunch — and the honest reorder point is 17 × 14 + 60 = 298.

The gap between the threshold you had and the threshold you needed is 104 units — about six days of demand at the new velocity. That's the entire cost of scaling ads without recalibrating: not a forecasting failure, just one input left stale. And note the asymmetry: the campaign only had to outperform for about a week before the reorder point was already wrong, because the reorder point commits you to orders that arrive 14 days later, inside the sustained spike.

Recalibrate before the change, not after

The routine that prevents this is small enough to do the morning of a budget change:

1. Estimate the new velocity before budgets move. Last 14 days of in-stock sales gives a baseline; multiply by your expected lift and you have a planning number. It won't be exact — it needs to be in the right neighborhood, set before the traffic arrives.
2. Recompute the reorder point the same day. New velocity × unchanged lead time + buffer. Ten minutes in the sheet.
3. Size the buffer for the new variability, deliberately. Scaled spend usually deserves a bigger buffer early on, trimmed back once you have two or three weeks of new-regime data.
4. Check pipeline depth. If stock is already below the new threshold when you raise budgets, the recalibration includes an expedite decision — the PO you needed was the one that should have gone out yesterday.
5. Re-review at day 7 and day 14. If realized velocity differs from your estimate by more than about 20%, run the update again. Two passes usually converge.

The same logic runs in reverse when you cut spend — thresholds set during a push fire early on a smaller demand base, and you'll over-order into a cash squeeze. Recalibrate in both directions.

Not every spike deserves a new threshold

Recalibration costs attention, so spend it where the change is real and sustained. A one-day creative flash doesn't justify rewriting thresholds; a new always-on budget does. Useful distinctions:

• Step-change (sustained budget or CPA target change): recalibrate before it lands, per the routine above.
• Event bump (launch day, viral moment, one-off placement): hold the threshold, place or expedite one PO against the estimated bump, and let the threshold re-earn its number after the event.
• Seasonal ramps: these belong on a promo calendar so they're expected — the reorder point update is scheduled, not reactive.

If you find thresholds drifting every week, the underlying issue is usually that velocity was measured over too long a window for how fast your demand regime actually changes. That's a measurement design problem, and it's worth fixing once instead of patching forever — the velocity-correction guide covers what to measure and what to exclude.

What the threshold can't see

One honest limitation, since thresholds are our trade here: no inventory tool watches your ad account. InventoryX (in development), for example, flags reorder risk using the velocity it sees in your store data and the manual lead times you enter — it will tell you a SKU is about to run dry and what's at stake, but it won't know you tripled a budget on Tuesday. The recalibration step in this article is the human part of the loop: you change the spend, you update the numbers the thresholds run on. Tools can draft the PO; only you can close the gap between the ad account and the buying sheet.

Common questions

When should I update reorder points after increasing ad spend? Before the spend lands — estimate the new velocity from your planned lift and recompute the threshold the same day budgets change, then review against realized numbers at day 7 and day 14. Updating only after you observe the spike means your first reorder under the new regime is already committed against old demand.

How much does ad spend change demand per SKU? It depends on your campaign mix and how much of the traffic is new-customer volume on hero SKUs — there's no universal multiplier, and anyone quoting one is guessing. Measure it: compare daily units per SKU across the 14 days before and after the change, excluding any days the SKU was dark.

Do I need more safety stock when scaling ads? Usually yes, at least temporarily. Scaled spend raises both the level and the variability of demand, and safety stock is sized against variability. Recompute it deliberately — from recent daily demand swings — rather than padding the reorder point with vibes, then trim it once two or three weeks of new-regime data exist.

Should I pause ads on products with low stock? Often the right call — paying for clicks on units you can't sell burns budget and trains customers toward alternatives — but it's a per-SKU decision about spend waste, not a reorder point change. The ads and out-of-stock guide covers when to pause and when to hold; this article's job is making sure the thresholds behind the stock are honest.

What about planned promos — do they break thresholds too? A promo is a known future demand change, which makes it easier than an ad-scale surprise: schedule it, reserve stock against the expected bump, and update the threshold for the promo window. The failure mode is the same — an unchanged threshold meeting a changed demand curve — but with a promo you see it coming.

Next step

Take the one SKU your current campaign pushes hardest and run the five-step recalibration against its next budget change — new 14-day in-stock velocity, unchanged lead time, deliberate buffer, pipeline check. If the corrected reorder point differs from the standing one by more than about five days of cover, that's your proof the process earns its ten minutes. The reorder point formula walkthrough has the spreadsheet template to run it in.