What is MAD/MAPE?

In Benchmarking Revenue view, you can enable MAD/MAPE to be displayed as extra columns so you can quickly judge forecast quality in comparative values and percentages across KPI’s like RevPAR, TrevPAR, Occupancy and ARR.

MAD = Mean absolute deviation

How MAD is calculated (in simple terms)

• For each day or period, calculate the difference between forecast and actual.
• Ignore plus/minus and use the absolute size of the error (how big the miss is).
• MAD is the average of these absolute errors across all days/periods.

Calculation examples to be used to cross-check calculations in Excel:

1. Formulas per day
   • For each day calculate the absolute deviation (no negatives):
     =ABS(Actual-Forecast)     eg.  =ABS(Y5-V5)
2. MAD over the period, e.g January month
   • At the bottom to get MAD for all days, 1 Jan – 31 Jan:
     =AVERAGE(Z5:Z35)

* See screenshots below for examples

How to interpret MAD

• Lower MAD → forecasts are generally close to actuals; typical misses are small in real terms.
• Higher MAD → larger typical forecast errors; the forecast often misses the mark by a bigger amount.

This makes MAD a practical indicator for assessing and improving forecast quality in views like Benchmarking, especially when you want to understand the average error in actual monetary or KPI units rather than in percentages.

MAPE = Mean Absolute Percentage Error

How MAPE is calculated (in simple terms)

• For each day or period, calculate the difference between forecast and actual.
• Convert that difference into a percentage of the actual value.
• Ignore plus/minus and use the absolute size of the error.
• MAPE is the average of these percentage errors across all days/periods.

Calculation examples to be used to cross-check calculations in Excel:

Change the format of the full column to percentage.

1. For each date:
• For each day calculate the absolute error (no negatives):
  =ABS((Actual-Forecast)/Actual)     eg.  =ABS((Y5-V5)/Y5)
• In stead of changing the format to percentage, you can also use this formula:
  =ABS((Actual-Forecast)/Actual)*100

1. MAPE over the period, e.g January month
   • At the bottom to get MAD for all days,. 1 Jan – 31 Jan:
     =AVERAGE(Z5:Z35)

How to interpret MAPE

• Low MAPE (e.g. around 0–10%) → forecasts are generally accurate; actuals are close to forecast.
• Higher MAPE (e.g. above 10–15%) → larger typical forecast errors; forecast quality needs attention.

This makes MAPE a practical indicator for assessing and improving forecast quality.

Image:  RevPAR MAD & MAPE in Benchmarking

Image: Typical example in Excel with formulas to calculate MAD/MAPE for a month