Forecast Accuracy (MAPE / WAPE) Calculator

What is Forecast Accuracy (MAPE / WAPE)? Forecast accuracy measures how close your forecasts are to actuals, using error percentages that work across revenue, demand, cost, or v...

Forecast Accuracy (MAPE / WAPE) Calculator

Measure forecast accuracy using Mean Absolute Percentage Error (MAPE) and Weighted Absolute Percentage Error (WAPE). Enter total actuals and forecasts for a set of periods to see error metrics, a quick interpretation, and example FP&A scenarios.

$

Total actual value across all periods you are evaluating (e.g., sum of actual monthly revenue for the last 12 months). Use the same currency for actuals and forecasts. If you track units instead of currency, treat this as a generic volume.

$

Total forecast value across the same periods (e.g., sum of the original monthly revenue forecast for the last 12 months). Must cover the same periods and metric as the actuals.

count

How many forecast vs actual points you are summarizing (e.g., 12 months, 4 quarters, or 52 weeks). Used to convert MAPE into an approximate Mean Absolute Error (MAE) per period. Must be a positive whole number.

%

Mean of the absolute percentage error at the individual period level: Average(|Actual – Forecast| ÷ Actual × 100). This is your MAPE input. Summary mode: enter your already-calculated MAPE to estimate MAE and WAPE from totals — this calculator does not ingest per-period rows. If your data includes many zero or negative actuals, MAPE can be distorted or not defined. Standard MAPE formulas require non-zero actuals, so treat results as rough guidance in those cases.

Scenarios
Load common FP&A use cases to see how totals and average % error combine into MAPE, WAPE, and bias.
Stable revenue plan (good accuracy)High-growth, volatile businessNew product launch (noisy forecast)

Results

  • MAPE (Mean Absolute Percentage Error)— %
  • Approximate WAPE (Weighted Absolute Percentage Error)— %
  • Estimated MAE per period$
  • Estimated total absolute error$
  • Net bias (% of total actuals)— %
  • Forecast accuracy band—

Enter your inputs above to calculate the results.

What is Forecast Accuracy (MAPE / WAPE)?

Forecast accuracy measures how close your forecasts are to actuals, using error percentages that work across revenue, demand, cost, or volume. MAPE shows the average absolute % error per period; WAPE weights error by scale, so big-dollar periods or segments drive the result.

These metrics matter because forecast error turns into working capital drag, margin dilution (expedites, waste, discounting), and weaker capital allocation decisions.

Formula

MAPE = 100 / nsumt = 1n|At-Ft / At|,quad WAPE = (100sumt = 1n|At-Ft|) / (sumt = 1n|At|),quad Bias = 100(Σ F-Σ A) / sum A

Example

Assume 12 periods with total actuals of $1,200,000 and total forecasts of $1,150,000, and the average per-period MAPE computed from the underlying points is 6.5%.

Estimated total absolute error ≈ 6.5% × $1,200,000 = $78,000, so estimated MAE per period ≈ $78,000 ÷ 12 = $6,500.

Net bias ≈ ($1,150,000 − $1,200,000) ÷ $1,200,000 = −4.2%, meaning forecasts are systematically low (under-forecasting).

Frequently Asked Questions

Should I use MAPE or WAPE to judge forecast accuracy for my business?

Use WAPE when volumes vary a lot across periods/items (big months should matter more). Use MAPE when each period/item should count equally and actuals aren’t near zero.

Why does my MAPE look “terrible” even when the forecast seems close?

MAPE blows up when actuals are small or zero. One low-actual period can dominate the average % error. In those cases, rely more on WAPE (or other metrics like sMAPE/MASE) and segment out low-volume periods.

What does “Net bias” mean, and how do I read the sign?

It shows systematic over/under-forecasting: Bias % = (Total Forecasts − Total Actuals) / Total Actuals. Positive = overall over-forecast; negative = overall under-forecast.

How do I translate a MAPE % into “$ error” I can explain to stakeholders?

Convert it to an estimated absolute error in dollars: Estimated total absolute error ≈ MAPE% × Total Actuals, then Estimated MAE per period ≈ (MAPE% × Total Actuals) / # periods.

Sources & Methodology