Calculator Inputs
Enter historical units sold and planning assumptions. The form uses three columns on large screens, two on medium screens, and one on mobile.
Example Data Table
This sample shows a simple historical units pattern you can paste into the calculator.
| Period | Units Sold | Traffic Change % | Conversion Change % | Promo Active |
|---|---|---|---|---|
| Month 1 | 420 | 1.5 | 0.8 | No |
| Month 2 | 445 | 1.9 | 1.0 | No |
| Month 3 | 460 | 2.1 | 1.2 | No |
| Month 4 | 490 | 2.3 | 1.4 | Yes |
| Month 5 | 515 | 2.8 | 1.4 | No |
| Month 6 | 545 | 3.0 | 1.5 | No |
Formula Used
1. Seasonal index: Seasonal Index = Period Average ÷ Overall Average
2. Deseasonalized demand: Deseasonalized Units = Actual Units ÷ Seasonal Index
3. Linear trend: Trend Base = a + bt, where a is the intercept and b is the slope from least squares regression.
4. Exponential smoothing: Smoothed Value = α × Current Demand + (1 − α) × Previous Smoothed Value
5. Blended baseline: Blended Base = Trend Weight × Trend Forecast + (1 − Trend Weight) × Smoothed Forecast
6. Final forecast: Forecast Units = Blended Base × Seasonal Index × Growth × Traffic × Conversion × Price × Promotion × Returns × Stockout
7. Scenario range: Lower = Forecast × (1 − Spread), Upper = Forecast × (1 + Spread)
The calculator blends historical trend, smoothing behavior, and ecommerce planning assumptions into one units forecast workflow.
How to Use This Calculator
- Paste historical units sold in order from oldest to newest.
- Set the number of future periods you want to forecast.
- Enter season length, such as 12 for monthly yearly cycles.
- Adjust smoothing alpha and trend weight to fit your demand pattern.
- Add ecommerce assumptions like traffic, conversion, price, promotions, returns, and stockouts.
- Click the calculate button to show results above the form.
- Review the chart, forecast table, diagnostics, and scenario range.
- Use the CSV and PDF buttons to export the forecast output.
Frequently Asked Questions
1. What does this calculator forecast?
It estimates future ecommerce units sold using historical sales, trend direction, seasonality, smoothing, and business adjustments like promotions, returns, stockouts, traffic, and conversion changes.
2. When should I use seasonality?
Use seasonality when sales repeat in predictable cycles, such as monthly holiday peaks or weekend lifts. You usually need at least two full seasonal cycles for better seasonal estimates.
3. What does smoothing alpha control?
Alpha controls how strongly recent data influences the smoothed baseline. Higher alpha reacts faster to recent changes, while lower alpha produces a steadier forecast.
4. Why include traffic and conversion changes?
Units sold often depend on both visitors and purchase rate. Modeling those separately helps connect marketing plans and site performance changes to inventory expectations.
5. How does price elasticity affect units?
Price elasticity estimates how sensitive demand is to price movement. A higher elasticity means unit sales change more sharply when you raise or lower price.
6. What do lower and upper case values mean?
They provide a simple planning range around the central forecast. This helps teams prepare conservative and optimistic inventory positions rather than relying on one number.
7. What is MAPE in the results?
MAPE is mean absolute percentage error. It compares fitted values with actual history and gives a percentage view of how closely the model matched past sales.
8. Can I use this for weekly or daily forecasts?
Yes. Keep the history interval and season length consistent. For example, use weekly history with a season length of 52 or daily history with a season length matching your cycle.