Key Takeaway: Inventory forecasting isn't about predicting the future perfectly. It's about buying closer to what you'll actually need instead of guessing and hoping the guess was close enough.
What's on This Page
What Forecasting Actually Solves
Without forecasting, purchasing decisions default to "buy about what we bought last time," which quietly ignores growth, seasonality, and changing demand. Forecasting replaces that habit with a number grounded in your own sales history.
A Simple Forecasting Method
You don't need statistical software to start. Take your average monthly sales for a SKU over the last 6-12 months, adjust for your overall growth rate, and multiply by a seasonal index (a simple ratio of "this month's typical sales" to "average month's sales").
Worked Example
A SKU averages 100 units/month. The business is growing 10% year over year. December typically sells 1.4x an average month.
Common Mistakes
- Forecasting from total revenue instead of per-SKU history, which hides which products are actually growing or declining
- Ignoring seasonality entirely and using a flat average all year
- Never revisiting the forecast after it's made, even when actual sales diverge from it
Pair this with a real reorder point (see Reorder Point Guide) so forecasting turns into an actual purchasing decision, not just a number on paper.
For further reading, see the Association for Supply Chain Management (ASCM).
Checklist
- Pull 6-12 months of sales history per SKU
- Calculate your overall business growth rate
- Identify a seasonal index for each major selling period
- Run the forecast formula for your top 10 SKUs
- Compare last month's forecast against actual results
- Adjust the growth rate or seasonal index based on that comparison
Common Mistakes
FAQ
Do I need statistical software to forecast demand?
No. A spreadsheet with 6-12 months of sales history, a growth rate, and a seasonal factor gets most small businesses a workable forecast.
How far ahead should a forecast look?
Match it to your supplier lead time plus a buffer. Forecasting further ahead than that just adds uncertainty without a corresponding purchasing decision to make.
What if a product has no sales history yet?
Use a comparable existing product as a proxy, then correct the forecast quickly once a few weeks of real data come in.
How do I know if my forecast is any good?
Compare forecasted vs. actual sales every month and track the gap. If it's consistently off in the same direction, adjust the growth rate or seasonal factor, not just that month's number.
Calculate This For Your Business
Related Guides in the Inventory Academy
- Reorder Point Guide. turning a forecast into a purchasing trigger
- Inventory KPIs. the numbers a good forecast should improve
- Why Inventory Mistakes Destroy Small Businesses. another guide in the Inventory Academy