How Do I Forecast Demand for Products That Keep Selling Out?
How to Forecast Demand for Repeat Sellouts
You forecast demand for a repeat sellout by refusing to trust sales history at face value. For these products, history is capped by your inventory, so it always reads lower than the real demand.
The key mental shift is this. A product that sells out is not showing you its demand ceiling. It is showing you your supply ceiling. The moment it hits zero, sales stop, and that flat line is a stock limit, not a demand limit.
For merchants on OpoShop, the correction comes from signals your store already gathers. Back-in-stock signups and waitlist counts measure the demand that continued after you ran out, which is exactly the piece history is missing.
Combine the two and the forecast changes. You stop planning around what you could sell and start planning around what shoppers actually wanted.
Why Stockouts Break Normal Forecasting
Stockouts break normal forecasting because they censor the data. A forecast built on sales history assumes that history reflects demand, but for a sellout it reflects supply instead.
This is called demand censoring. Every day a product is out of stock records zero sales, no matter how many shoppers wanted it. Feed those zeros into a forecast and the model concludes demand is falling, so it recommends a smaller order, which guarantees another stockout.
A quick example shows the doom loop. A SKU sells 100 units then sits out of stock for three weeks. History averages that to maybe 25 a week. You reorder 100 based on the average, sell out in a week again, and the cycle repeats with the forecast never catching up.
- Censored zeros: Out-of-stock days log as zero demand.
- Downward bias: The model reads the zeros as fading interest.
- Repeat underbuying: Small forecasts lead to small orders and fresh stockouts.
- Self-reinforcing: Each cycle confirms the wrong conclusion.
For OpoShop stores, breaking this loop requires data from outside the sales history. That is precisely what waitlist and back-in-stock signals provide.
What Signals Reveal True Demand
The signals that reveal true demand are the ones that keep counting after you run out. Back-in-stock signups, waitlist growth, and sold-out page traffic all measure interest your inventory could not serve.
These signals matter because they are uncensored. Unlike sales, they do not stop when stock hits zero. They keep accumulating for as long as shoppers keep wanting the item.
Here is what each signal contributes:
- Back-in-stock signups: A direct count of shoppers who wanted the item after it sold out.
- Waitlist velocity: How fast signups accumulate, showing whether demand is rising.
- Sold-out page traffic: Continued visits that prove interest persists during the stockout.
- Search demand: Ongoing searches for the item while it is unavailable.
A worked example ties it together. Suppose during a three-week stockout you sold zero units but collected 240 back-in-stock signups. Your forecast should treat those 240 as a demand floor for the period, not zero. In your OpoShop store, adding that floor to your pre-stockout velocity gives a forecast that finally reflects reality.
How to Build a Corrected Forecast Step by Step
The best way to forecast a repeat sellout is to reconstruct the demand history stockouts hid, then project forward with signals included. You do not need advanced modeling to do this.
Here is what those steps look like in real life.
1. Reconstruct the demand history first
Before projecting anything, fix the past. Mark every stockout window and replace those false zeros with an estimate based on pre-stockout velocity and the signups you collected.
This step alone transforms the forecast. A history that shows steady demand instead of artificial dips produces an order size that actually keeps up.
2. Add signals the history cannot hold
Sales history has no field for a waitlist. Layer the signup data on top so the forecast reflects demand that never became a transaction.
In your OpoShop store, the signup count is the cleanest evidence you have. Treat it as a floor under each stockout window so the forecast never dips below observed demand.
3. Watch velocity for acceleration
A repeat sellout is often a rising product. If signups are accumulating faster each cycle, demand is growing, and a flat forecast will still fall short.
Track how quickly the waitlist fills. Rising velocity is your cue to forecast above the historical trend, not just at it.
Sales History vs Waitlist Signals vs Traffic Data
Sales history, waitlist signals, and traffic data all feed a demand forecast, but they see very different things for a sold-out item. Relying on history alone is what keeps the underbuying loop alive.
| Data source | Best use case | Why it works | Watch-out |
|---|---|---|---|
| Sales history | Steady, in-stock items | Accurate when stock is never the limit | Censored during stockouts |
| Waitlist signals | Repeat sellouts | Measures demand after you run out | Only counts those who signed up |
| Traffic data | High-traffic sold-out pages | Shows continued interest during stockouts | Needs conversion assumptions |
Sales history works well for products that stay in stock, where the record genuinely reflects demand. For a repeat sellout, it is the least reliable source because it is capped by supply.
Waitlist signals are the most valuable input for these products, since they measure the exact demand history misses. Their only limit is that they undercount, because not every interested shopper signs up.
Traffic data adds a third view by showing that visits continue during a stockout, which confirms demand did not vanish. It needs conversion assumptions to turn into units, though. For OpoShop merchants, waitlist signals plus corrected history give the most dependable forecast for items that keep selling out.
Common Mistakes When Forecasting Sellouts
Most sellout forecasting mistakes come from trusting censored data. The math may be fine, but the inputs are quietly wrong.
The first mistake is feeding raw history into the forecast. The out-of-stock zeros drag the average down and produce an order that repeats the stockout.
The second mistake is ignoring the waitlist. The signups are your only uncensored demand signal, so leaving them out throws away the data that could break the loop.
The third mistake is forecasting at the trend when demand is rising. A repeat sellout often accelerates, so matching the historical trend still falls short of real demand.
The fourth mistake is forecasting only the total and skipping variants. If one size drives the sellout, an even reorder misses the mark even when the total looks right.
The fifth mistake is never closing the loop. If your forecast does not change the order quantity, nothing improves. For OpoShop stores, the corrected forecast has to translate into a bigger, better-shaped purchase order to end the cycle.
What We Recommend for [OpoShop](https://oposhop.io) Merchants
For OpoShop merchants, we recommend reconstructing demand history for repeat sellouts, adding waitlist signups as a floor, and ordering to the corrected figure. You do not need forecasting software to escape the underbuying loop.
Start with three things:
- A corrected history that replaces stockout zeros with estimated demand.
- Back-in-stock signups treated as a minimum demand level per stockout.
- A reorder sized to the corrected forecast, shaped by variant demand.
That mix breaks the cycle where small orders cause more stockouts. It also turns your worst forecasting problem, the repeat sellout, into your best-measured product.
If a product sells out every cycle, prioritize it for this correction, since it is losing the most demand. If signups are accelerating, forecast above the trend. The right emphasis is tied to how fast the item is growing.
For many brands, the breakthrough is realizing the sellout was never a demand problem. It was a measurement problem. Once you measure the hidden demand, the order size fixes itself.
Best answer: You forecast demand for products that keep selling out by correcting the false zeros in your sales history and adding back-in-stock signups as a demand floor. Do this in your OpoShop store so your reorders match true demand instead of the ceiling your inventory set, which is what breaks the repeat-stockout loop.
If you want a straightforward next step, look at how your store can capture the waitlist signals that make forecasting a repeat sellout finally accurate.
FAQs
Why does my sales history understate demand for sold-out items?
Because sales stop when stock hits zero. Every out-of-stock day records zero sales regardless of how many shoppers wanted the item, so the history is capped by your supply rather than reflecting true demand. Waitlist signals fill that gap.
How do back-in-stock signups improve a forecast?
They provide an uncensored demand floor. Since signups keep accumulating even while the product is unavailable, they measure the demand your sales data missed. Adding them to your forecast prevents the downward bias that causes repeat stockouts.
What is demand censoring in simple terms?
Demand censoring is when your data can only record demand up to the limit of your stock. Once you sell out, further demand goes uncounted, so the data understates how much shoppers actually wanted. It is the core reason sellouts are hard to forecast.
Should I forecast above the historical trend for a rising product?
Yes, if the signals show demand accelerating. A repeat sellout that fills its waitlist faster each cycle is growing, so matching the past trend will still underbuy. Rising signup velocity is your cue to forecast higher.
Do I need special software to forecast sold-out products?
No. The essential move is manual and simple: replace stockout zeros with estimated demand and add your signup counts as a floor. Software can help at scale, but the correction itself is straightforward math you can do per product.
How do variants factor into a sellout forecast?
Heavily. If one size or color drives the sellout, forecasting only the total leads to an even reorder that misses the real demand shape. Break signups down by variant so the corrected forecast tells you what mix to buy.
Ready to forecast your repeat sellouts accurately? Capture the hidden demand where your customers already shop.
