Free sales forecast calculator
Enter your historical sales data — this sales forecast calculator fits a linear trend using least-squares regression and projects the next period, updated live, as you type.
On this page10 sections
Does not include taxes, depreciation, or opportunity cost. Real-world break-even analysis should include all overhead costs.
Results are estimates. Consult a professional.
How the sales forecast calculator works
This calculator offers two forecasting methods: linear regression (least squares) and simple growth-rate projection. Linear regression finds the best-fit straight line through your historical sales data and extends it forward — useful when your trend has been consistent. The growth-rate method multiplies your most recent period by (1 + growth rate) — useful for extrapolating recent momentum. The calculator returns the period-7 (or next-period) forecast plus a 95% confidence interval for the regression method.
Neither method accounts for seasonality, sudden market disruption, or macro-economic shocks — they project the past trend forward. The confidence interval from linear regression tells you the statistical precision of the model given your data's spread around the trend line, but it is not a guarantee about the future. Use the interval as a planning range, not a floor-and-ceiling commitment.
SCORE — Sales forecasting guide for small businesses.Worked example: 6 months of sales data, forecast period 7
A retailer has six months of sales data: $80k, $90k, $95k, $105k, $115k, $120k (periods 1–6). They want a linear-regression forecast for period 7 and a 95% confidence range for budget planning.
6-period data, regression statistics, and period-7 forecast
The table below shows each period's actual sales, the regression-fitted value, and the residual (how far actual was from fitted). Small residuals confirm the trend is highly linear; large or patterned residuals suggest non-linearity or seasonality that the model misses.
| Period | Actual sales | Fitted value | Residual |
|---|---|---|---|
| 1 | $80,000 | $80,470 | −$470 |
| 2 | $90,000 | $88,610 | +$1,390 |
| 3 | $95,000 | $96,750 | −$1,750 |
| 4 | $105,000 | $104,890 | +$110 |
| 5 | $115,000 | $113,030 | +$1,970 |
| 6 | $120,000 | $121,170 | −$1,170 |
| 7 (forecast) | $129,310 | — | — |
Slope = 8.14, Intercept = 72.33. 95% CI for period 7: [$126,135 – $132,485]. Source: least-squares regression; SCORE forecasting guide.
Tips for building an accurate sales forecast
A forecasting model is only as good as the data and assumptions behind it. These five practices separate reliable forecasts from misleading ones.
- Use at least 6–12 periods of data. Fewer than six data points make regression highly sensitive to any single period. More periods improve statistical reliability and help the model distinguish trend from noise.
- Identify and adjust for seasonality. Linear regression will systematically over- or under-forecast if your business has seasonal patterns. Either use year-over-year growth rates or apply seasonal indices before running regression.
- Check the residuals for patterns. Random residuals (no up-down pattern) confirm linear regression is appropriate. If residuals curve upward then downward, the relationship is non-linear — consider a growth-curve or exponential model instead.
- Build scenarios, not a single number. Create a base case (regression forecast), an upside (top of the confidence interval plus upside assumptions), and a downside (bottom of interval minus a risk factor). Decision-makers need a range to plan cash flow and hiring.
- Refresh the model monthly. As each new period closes, add the actual to your data and re-run the regression. The model naturally improves as you accumulate more periods and correct for prior forecast errors.
Accuracy and limitations
This calculator implements the ordinary least-squares (OLS) linear regression and a simple compound growth-rate method. Both are widely used, mathematically sound, and appropriate for trending data over short to medium horizons. The 95% confidence interval is a statistical statement about the model's fit to historical data — not a prediction interval that accounts for all sources of future uncertainty. It will understate actual forecast uncertainty because it does not capture structural breaks, competitive disruption, new product launches, or macroeconomic shocks.
Long-horizon forecasts (beyond 3–4 periods out) using these methods degrade rapidly in reliability. For complex businesses with multiple products, regions, or strong seasonality, dedicated forecasting software — or a statistician-built model with seasonal decomposition and external predictors — will significantly outperform this calculator. Use results as a planning starting point, not a definitive projection.
Sales forecasting terms defined
About this sales forecast calculator
This calculator runs entirely in your browser — nothing you enter is sent to any server.
Browse more in our business calculators, or explore the complete library on the free calculators page.