Inventory how-to

What to do when your forecast is wrong

Every forecast is wrong by some amount, so the goal is to catch it early and correct the order before you overbuy or run out.

Short answer: Every forecast is wrong by some amount, so the goal is to catch it early and correct the order before you overbuy or run out. Check actual sales against the forecast weekly, adjust the inputs that caused the gap, and widen safety days for SKUs that keep surprising you.

Why forecasts miss

A forecast is a guess based on recent sales, usually a 28-day window plus seasonality once there is enough history. It misses when demand changes faster than the window can pick up: a product goes viral, a competitor goes out of stock, a influencer posts, or the weather shifts early. It also misses when the sales history itself is bad, for example a stockout was counted as low demand instead of no stock.

Find out why before you change anything

Before adjusting numbers, figure out if the miss was a demand shift or a data problem. Pull up the SKU's recent sales and see if the spike or drop lines up with a marketing push, a price change, a competitor, or a stockout. If it was a one-off event like a single big wholesale order, treat it as an outlier, not a new trend.

Adjust the right lever

If demand has genuinely shifted, shorten the sales window you rely on mentally, or increase safety days so the reorder point reacts faster next time. If a stockout dragged your average down, exclude those out-of-stock days when you judge the trend by hand. If lead times from the supplier have grown or shrunk, update that first since it changes the reorder point and order quantity directly, independent of the forecast.

Build in a manual override

Treat the suggested quantity as a starting point, not a final answer, for any SKU you know is affected by a promotion, seasonality not yet in the history, or a one-time push. Increase or decrease the quantity by hand and note why, so next time you can check if your override was closer than the forecast.

Step by step

  1. Compare actual sales to forecast weekly for your top SKUs by revenue.
  2. Check whether a miss was caused by a real demand shift, a stockout, or a one-off order.
  3. If demand shifted, increase safety days for that SKU so future orders react faster.
  4. If a stockout skewed the data, mentally exclude those days when judging the real sales rate.
  5. Confirm the supplier's lead time is still accurate, since a wrong lead time causes the same symptoms as a wrong forecast.
  6. Override the suggested quantity by hand for any SKU tied to a known promotion or seasonal event not yet in the history.
  7. Revisit the override after the next sales cycle to see if it was closer than the system's number.

Worked example

Say a candle SKU normally sells 5 units a day, and with a 10-day lead time and 4-day safety stock, the reorder point is 5 x (10+4) = 70 units. A influencer features the candle and sales jump to 15 units a day for two weeks, but the 28-day window only shows an average of 9 units a day by the time you notice. You manually override the order quantity upward based on the 15-unit daily rate and lower the window's apparent average by excluding the pre-spike days when sales were closer to 5. After two weeks the spike fades and you remove the override, letting the rolling window settle back to a true 5-6 units a day.

How SteadyShelf helps

SteadyShelf shows you the forecast and the order-by date so you can see a miss early and adjust safety days or override the quantity before the order ships. It does not detect why a forecast is wrong or auto-correct for promotions and viral events, so spotting the cause and deciding on a manual override is still on you.

$29/month flat, unlimited products and suppliers, 30-day free trial. Run the numbers now with the free reorder point calculator.

SteadyShelf is launching on the Shopify App Store. Join early access to get the install link first.

Related guides

All inventory how-to guides ยท Inventory glossary

FAQ

How much should I raise safety days after a bad miss?

Start small, for example add 2 to 3 days, and watch the next few cycles. Raising it too much just ties up cash in extra stock, so adjust in small steps and only for the SKUs that actually keep surprising you.

Should I lower safety days if the forecast overshoots and I end up overstocked?

Yes, if the overstock is a pattern and not a one-off. Check a few cycles first, since a single high forecast after a slow month can be normal noise rather than a reason to change your settings.

Let SteadyShelf do this math for every SKU

Order-by dates per supplier, seasonal forecasts and draft POs for your Shopify store. $29/month after a 30-day free trial.

Get early access