How to calculate it
Divide the property’s occupancy by the comparison group’s occupancy and multiply by 100. Both figures must cover identical dates and use comparable rules for sold and available room nights. Use two percentages or two decimal fractions consistently, rather than mixing the formats.
The group’s aggregate occupancy comes from its total sold nights divided by its total available nights. It is generally not a simple average of each hotel’s percentage because properties differ in size. Check the group membership and whether the subject property is included. If benchmark occupancy is zero, or either occupancy measure is undefined, MPI cannot be calculated meaningfully.
A hypothetical example
The property sells 225 of its 300 available room nights, giving 75% occupancy. The comparison group sells 1,800 of 3,000 available nights, giving 60%. MPI is 125. The property’s occupancy is 25% higher in relative terms, while the difference between the two occupancy percentages is 15 percentage points.
How to read the base of 100
Above 100 indicates higher occupancy than this particular group; below 100 indicates lower occupancy. It is a relative measure. MPI can rise while the property’s own occupancy falls if the group loses proportionally more demand. Read the index alongside both underlying occupancy percentages before describing the business as growing.
When it is useful
MPI helps examine whether an occupancy change is specific to your property or shared with a wider demand environment. It requires reliable aggregate data and a relevant comparison group with appropriate location, product and customer characteristics. Keep the group consistent over time, or clearly identify changes that affect the comparison.
It does not measure rate or profit
A high MPI can coexist with lower rates. It does not prove higher profit or mean that the property holds an equivalent percentage of all market bookings. Add ARI and RGI to examine rate and room-revenue performance. An unsuitable comparison group can make an impressive-looking index misleading.