Demand Forecaster Calculator
Enter 6–12 months of historical sales data and get 30, 60, and 90-day demand forecasts with confidence intervals and trend direction.
Demand Forecaster
Enter monthly sales data (units sold) — most recent 6–12 months
| Month | Units Sold |
|---|
Forecasts are statistical projections based on historical patterns. Actual demand will vary. Do not use as sole basis for major purchasing decisions. Review with your operations and commercial teams.
How to Use the Demand Forecaster Calculator
The Demand Forecaster Calculator uses your historical monthly sales data to project future demand over 30, 60, and 90 days. It combines weighted moving averages with linear trend detection and optional seasonality adjustment to generate statistically grounded forecasts — without requiring a data science background.
Step 1: Enter Your Historical Sales Data
Enter your monthly sales figures in units sold, starting from the oldest month and working forward to the most recent. You can enter 6, 9, or 12 months of data — more data generally means better accuracy, especially for detecting seasonality.
Use units consistently (don't mix revenue and units). If you only have revenue data, convert it to units by dividing by your average selling price. Leave months blank if you have fewer than the selected horizon — the calculator will use available data only.
Step 2: Choose Your Seasonality Setting
Seasonality adjustment analyzes month-over-month patterns in your historical data and applies multipliers to the forecast. Turn it On if your business has clear seasonal patterns (e.g., higher demand in Q4, summer slowdown). Leave it Off for products with relatively flat demand throughout the year, or if you have less than 12 months of data.
With fewer than 9 months of data, seasonality adjustment is less reliable and may introduce noise rather than signal.
Step 3: Generate and Interpret the Forecast
After clicking Generate Forecast, you'll see three outputs:
- 30-day forecast — Most reliable; used for immediate procurement decisions
- 60-day forecast — Medium-term planning; good for lead time planning
- 90-day forecast — Longest horizon; use for capacity and budget planning with wide confidence intervals
Each forecast includes a 95% confidence interval — the range within which actual demand is statistically likely to fall. Plan your safety stock around the upper bound of the confidence interval, not the point estimate, to avoid stockouts.
Understanding the Trend Indicator
The trend indicator shows the direction and magnitude of demand change month-over-month:
- 📈 Growing — Demand is increasing. Ensure stock levels and reorder quantities grow proportionally.
- 📉 Declining — Demand is falling. Reduce order quantities and monitor for dead stock risk.
- ➡️ Stable — Demand is relatively flat. Standard reorder policies apply.
How the Forecast is Calculated
The forecaster uses a weighted moving average that gives progressively more weight to recent months, combined with a linear regression trend component:
Trend = linear regression slope × months ahead
Forecast = Weighted Average + Trend Adjustment
Confidence intervals are calculated from the standard deviation of historical demand residuals (actual vs. fitted values), expanded proportionally for longer horizons:
When to Use This Tool
- Planning purchase orders for the next quarter
- Sizing safety stock for the upcoming season
- Preparing inventory budget forecasts
- Identifying demand trends before your next supplier meeting
- Evaluating whether current stock aligns with projected demand
Limitations to Know
This tool uses statistical projection from historical data. It cannot account for:
- New product launches or discontinuations
- Major promotional campaigns or pricing changes
- Supply chain disruptions or competitor moves
- Macroeconomic shifts
For products with highly irregular demand or significant external drivers, use this forecast as a baseline and adjust manually based on market intelligence.
Frequently Asked Questions
A demand forecaster calculator uses historical sales data to project future demand over a defined horizon (30, 60, or 90 days). It applies statistical methods like weighted moving averages and linear trend detection to identify patterns and generate forecasts with confidence intervals showing the expected range of outcomes.
90-day forecasts typically have 15–25% mean absolute percentage error (MAPE) for consumer products without major external disruptions. Accuracy improves with more historical data (12+ months), stable demand patterns, and seasonality adjustments. Confidence intervals widen for longer horizons to reflect increasing uncertainty.
At minimum, you need 6 months of historical sales data (units sold per month). For better accuracy, 12 months captures a full seasonal cycle. Enter your monthly sales figures from oldest to newest. The tool calculates the trend direction, weighted moving average, and confidence intervals automatically.
Seasonality adjustment applies seasonal index multipliers based on historical month-over-month patterns in your data. If your data shows that demand is consistently 30% higher in December, the tool amplifies the December forecast accordingly. Without seasonality adjustment, the forecast uses a flat trend projection.
Confidence intervals show the range within which actual demand is likely to fall. A 95% confidence interval means that in 95% of cases, actual demand will be within the shown range. Wider intervals indicate more uncertainty — common for longer horizons or highly variable demand. Narrower intervals indicate more predictable demand.
A weighted moving average gives more weight to recent months, making it responsive to recent demand shifts. Linear regression fits a trend line to all historical data, which is better for products with a clear growth or decline trend. This tool uses a weighted moving average with a trend adjustment, giving you the benefits of both approaches.