Choose the Right Sales Period for Inventory Forecasting

One of the first settings you'll encounter in any inventory forecasting tool is the sales period — the window of historical sales data used to predict future demand. Get it right, and your forecasts will be reliable. Get it wrong, and you'll either overstock or run out. This is a core part of good demand planning.

The answer isn't the same for every store. It depends on how your products sell, how consistent your demand is, and what's happening in your business right now.

What Is a Sales Period?

The sales period is the number of days of sales history that your forecasting tool analyzes to calculate how fast products are selling. This sales rate then drives everything else — including when you hit your reorder point, recommended order quantities, and days of stock remaining.

In Sensible Forecasting, you have five options: 14 days, 30 days, 60 days, 90 days, and 90 days weighted average.

A shorter period means the forecast reacts quickly to recent changes. A longer period smooths out fluctuations and gives you a more stable baseline. Both approaches have trade-offs. If you're new to inventory metrics, our guide to the 5 numbers every Shopify store should know is a good place to start.

Short Sales Period (14-30 Days)

When to Use It

• Seasonal products — If you sell holiday decorations, summer gear, or anything with sharp seasonal swings, a 14 or 30-day window captures current demand without being dragged down by off-season months
• Volatile sales — Products that go viral, get featured by an influencer, or have unpredictable spikes benefit from a shorter lookback
• New products — If you launched a product two weeks ago, a 90-day lookback doesn't make sense because there's no 90-day history. A 14-day window uses what data exists
• Approaching a known peak — Heading into Black Friday? A shorter period captures the ramp-up in demand more accurately

The Risk

Short periods are sensitive to anomalies. If you had an unusually good (or bad) week, a 14-day forecast will overweight that week heavily. One big wholesale order can skew your entire forecast upward, leading you to overstock.

Long Sales Period (60-90 Days)

When to Use It

• Consistent, year-round sellers — Products with steady, predictable demand benefit from a longer lookback that averages out weekly noise
• Established products — If you've been selling the same catalog for years and demand is relatively stable, a 90-day window gives you the most reliable baseline
• Low-volume products — Products that sell a few units per week have noisy short-term data. A 90-day average is much more stable than a 14-day snapshot

The Risk

Long periods are slow to react. If demand suddenly doubles because of a marketing campaign or a competitor going out of stock, a 90-day average will underestimate the new reality for weeks. You might miss the window to restock in time.

The Sweet Spot: 90-Day Weighted Average

This is the default setting in Sensible Forecasting, and for good reason. The weighted average looks at 90 days of data but gives more importance to the most recent sales.

In practical terms, this means:

• If a product is trending upward (selling more recently than it was two months ago), the forecast adjusts upward
• If a product is slowing down, the forecast adjusts downward
• But a single unusual week doesn't throw off the entire forecast the way it would with a 14-day window

The weighted average gives you the stability of a long lookback with the responsiveness of a shorter one. For most Shopify stores with a mix of steady sellers and trending products, it's the best all-around choice.

Comparing Periods Side by Side

One of the most useful things you can do is compare sales rates across periods for the same product. In Sensible Forecasting's product detail view, you can see the sales rate for every period at once.

For example, you might see that a product sold 4 units in the last 90 days, but 2 of those were in the last 30 days alone. That tells you demand is accelerating — the 90-day average underestimates current demand, and you should lean toward the shorter period (or trust the weighted average to adjust).

Conversely, if a product sold 10 units in 90 days but only 1 in the last 30, demand is declining. The 90-day average overestimates current demand, and the shorter period is more realistic.

You Don't Have to Pick Just One

While the sales period is a global setting that affects the main Products table, the product detail view always shows you forecasts across all periods. Use the global setting for your most common scenario, and dive into the detail view when you need a more nuanced picture for specific products.

A practical approach: set your global period to 90-day weighted average, and before placing any large order, open the detail view to check if the recommendation makes sense given the product's recent trend.

When to Change Your Sales Period

Don't set it once and forget it. Whether you're using spreadsheets or a dedicated forecasting app, revisit your sales period setting when:

• Seasons change — Shorten the period heading into a peak season, lengthen it during stable periods
• You run a major sale — A big promotion can inflate your sales rate. Consider using the "exclude sales" setting in Sensible Forecasting to strip promotional spikes from your baseline
• Your catalog changes significantly — Added a bunch of new products? They won't have enough history for a 90-day lookback
• Something external changes — A competitor closes, a viral TikTok features your product, a new market opens up. These events shift demand in ways that only recent data will capture