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.

Forecast Accuracy Calculator (MAPE & WAPE) — Measure Error, Bias & MAE

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.

Sum of actuals for all periods

Sum of forecasts for all periods

Number of periods / data points

count

MAPE (avg absolute % error across points)

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

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

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).

How to Use the Forecast Accuracy (MAPE / WAPE) Calculator

Enter your total actuals, total forecasts, and number of periods, then add your MAPE% to estimate dollar error, bias, and an accuracy band.

Enter total actuals
- Fill in Sum of actuals for all periods (e.g., total revenue shipped/recognized across the periods you’re evaluating).
Enter total forecasts
- Fill in Sum of forecasts for all periods (the total you predicted for the same periods).
Set the number of periods / data points
- Enter Number of periods / data points (e.g., 12 for 12 months). This is used to estimate error per period.
Input your MAPE %
- - Enter MAPE (avg absolute % error across points) if you already calculated it from your period-level data. The calculator uses it to estimate dollar error:
formula (the formula in plain text, if is required)
- MAPE = average(|Actual − Forecast| / Actual) × 100
- Estimated total absolute error ≈ (MAPE% / 100) × Total Actuals
- Estimated MAE per period ≈ Estimated total absolute error / # periods
Read the outputs (accuracy, WAPE, MAE, bias)
- - MAPE: your entered accuracy metric.
- Approximate WAPE: a total-weighted error estimate based on totals.
- Estimated MAE per period: the “average $ miss” per period.
- Net bias (% of total actuals):
formula (the formula in plain text, if is required)
- Bias % = (Total Forecasts − Total Actuals) / Total Actuals × 100
- Forecast accuracy band: a quick qualitative label to communicate results.

Frequently Asked Questions

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

Methodology & Sources

Bibliography

Chris G. Caplice (2003). Demand Forecasting — MIT (Engineering Systems Division / MIT OpenCourseWare via DSpace) https://dspace.mit.edu/bitstream/handle/1721.1/39816/ESD-260JFall2003/NR/rdonlyres/Engineering-Systems-Division/ESD-260JFall2003/E09127E4-7613-4502-A67F-6C2BFF31BBC5/0/l02_3demandfcast.pdf
Accessed 2025-12-22
Rob J. Hyndman; Anne B. Koehler (2006). Another look at measures of forecast accuracy — International Journal of Forecasting (Elsevier) https://doi.org/10.1016/j.ijforecast.2006.03.001
Accessed 2025-12-22
Rob J. Hyndman (2025). WAPE: Weighted Absolute Percentage Error — Hyndsight (robjhyndman.com) https://robjhyndman.com/hyndsight/wape.html
Accessed 2025-12-22

Forecast Accuracy (MAPE / WAPE) Calculator

Results

What It Means?

What is Forecast Accuracy (MAPE / WAPE)?

Formula

Example

How to Use the Forecast Accuracy (MAPE / WAPE) Calculator

Enter total actuals

Enter total forecasts

Set the number of periods / data points

Input your MAPE %

Read the outputs (accuracy, WAPE, MAE, bias)

Frequently Asked Questions

Methodology & Sources

Bibliography

Other Calculators in Operations & Workforce Metrics

Total Cost of Quality Calculator

OSHA Total Recordable Incident Rate (TRIR) Calculator

Bradford Factor Calculator

Attrition Rate Calculator

Performance (Speed Loss) Calculator

Overall Equipment Effectiveness (OEE) Calculator