Mean Absolute Percentage Error (MAPE)

What Is MAPE?

MAPE stands for Mean Absolute Percentage Error. It measures the accuracy of your forecasts by comparing the difference between the actual and forecasted values as a percentage.

• Absolute Error: The difference between your forecast and the actual result.
• Percentage Error: This difference is expressed as a percentage of the actual result.
• Mean: The average of these percentage errors over all data points.

Calculating MAPE in Excel: Step-by-Step

1. Gather Your Data

Ensure you have your Actual Values and Forecast Values in two columns. For example:

2. Calculate the Percentage Error for Each Data Point

In a new column, calculate the absolute percentage error using this formula:

=ABS((Actual - Forecast) / Actual)

For example, if your Actual is in cell B2 and Forecast is in C2, then in D2 enter:

=ABS((B2 - C2) / B2)

3. Compute the Mean (Average) of These Errors

In another cell, use the AVERAGE function to find the MAPE:

=AVERAGE(D2:D4)

Format it to %

4. Interpret Your Result

A MAPE of 0% is perfect forecasting, while higher percentages indicate more deviation from actual values.

Interpreting MAPE Values: High vs. Low

When it comes to MAPE, the value itself tells you about the accuracy of your forecasts:

• Low MAPE (Good Forecast Accuracy):
  • A lower percentage means your forecast errors are smaller relative to the actual values. For example, a MAPE of 5% indicates that, on average, your forecast is only off by 5% from the real value.
  • In many industries, a MAPE below 10% is considered excellent, while values up to 20% may be acceptable depending on the volatility of the data.
• High MAPE (Poor Forecast Accuracy):
  • A high MAPE indicates larger discrepancies between your forecasts and actual outcomes. This suggests that your model may need adjustments or that the market conditions are too unpredictable.
  • What counts as “high” or “low” can vary by industry and application. In dynamic markets, even a higher MAPE might be typical.

Industry Benchmarks for MAPE

Understanding industry benchmarks can help you set realistic expectations and evaluate the performance of your forecasting model. The following are estimates of MAPE by industry:

• Retail:
  • In retail, where demand can be highly variable due to seasonality and promotions, a MAPE between 10% and 20% is common.
• Manufacturing:
  • Manufacturing often deals with more stable demand patterns. A MAPE below 10% is generally expected, though industries with custom orders or volatile demand might see slightly higher values.
• Consumer Goods:
  • For fast-moving consumer goods, benchmarks can range between 5% and 15%, reflecting the balance between steady demand and market fluctuations.
• Service Industries:
  • Service sectors, where external factors play a big role, might experience MAPEs from 15% to 25%.

Common Challenges with Using MAPE

While MAPE is a popular metric, there are several challenges to be aware of:

• Zero or Near-Zero Actual Values:
  • When actual values are zero (or very close to zero), the calculation can become undefined or misleading. Always check your data and consider alternative metrics if zeros are common.
• Large Outliers:
  • A few extreme differences between forecast and actual values can skew the average percentage error, making your overall MAPE seem worse than it might be in typical scenarios.
• Scale Differences Across Data Sets:
  • Comparing MAPE values across products or time periods with different scales can be misleading. What’s acceptable in one context may be poor in another.
• Overfitting Concerns:
  • Models might be overly tuned to past data (overfitting), resulting in a low MAPE historically but poor performance on new data. It’s important to validate your model on fresh data

🚿Shower Thoughts

• What Hidden Patterns Exist in your Data?
  Have you noticed recurring trends or seasonal effects in your data that might explain consistent forecasting errors?
• Is a Perfect MAPE Always Ideal?
  Could an extremely low MAPE indicate overfitting, where your model works well on past data but struggles with new scenarios?
• Are External Factors at Play?
  Consider how external events—like marketing promotions, economic shifts, or market trends—might be influencing your forecast errors. How might these factors be integrated into your model?
• Beyond MAPE:
  What other metrics could complement MAPE to give you a fuller picture of your forecasting accuracy? How might combining different metrics enhance your decision-making?
• Data Quality Check:
  Reflect on the quality of your input data. Could data anomalies, such as zeros or outliers, be skewing your results? What steps can you take to clean your data?

Ready to dive even deeper into supply chain insights? Sign up now for our newsletter and get clear, actionable articles delivered straight to your inbox—no spam, just valuable tips to boost your forecasting and supply chain strategies. Join our community today!