Hotels cut hours per occupied room every month of H1 2026, even as wages rose

Hotels got more done with fewer hours in every month of the first half of 2026. HotelData, the research arm built on Hotel Effectiveness labour data, published its half-year report on 17 September covering roughly 5,000 US hotels.

Full service properties cut hours per occupied room by 3.1% year on year. Select service cut 3.5%. Both did it while running busier.

The numbers

Occupancy across the sample reached 67.9% in H1 2026, up from 66.8% a year earlier. RevPAR rose 8.9% and TRevPAR 9.2%.

Headcount went the other way. Average headcount fell 2% at full service hotels and 1.5% at select service. Wages did not: hourly pay across housekeeping positions rose between 2.9% and 3.3%.

Housekeeping carried much of the productivity gain. Minutes per occupied room for select service room attendants fell 5%, from 24.01 to 22.82. Laundry attendants at select service properties fell 3.1%. At full service, room attendants fell 2.7% and laundry 2.2%.

Sarah McCay Tams, head of research and editorial at Actabl, summed it up: "Hotel operators moved the needle on their largest controllable cost while serving more rooms and paying more per hour."

The part in the quarterly split

The headline is good. The trend inside it is less comfortable.

Full service HPOR improved 3.7% in Q1 and 2.5% in Q2. Select service improved 5.1% in Q1 and 1.8% in Q2. In both segments the rate of improvement roughly halved between quarters, and in select service it fell by nearly two thirds.

That is what running out of easy wins looks like. The first round of productivity gains comes from scheduling better, matching labour to actual arrivals and cutting the hours nobody could justify. Those are available once. What comes after is harder and lands closer to the floor.

The report's own framing is careful about this. The stronger measure of success, it argues, is whether hotels can keep matching paid labour to demand while holding the service standards that create that demand in the first place.

What it means on the floor

A 5% cut in minutes per occupied room at select service is about 1.2 minutes less per room. Across a fifteen room section that is roughly eighteen minutes, which is real but survivable. Across a year of quarters each taking another slice, it stops being survivable, and the cost shows up as turnover rather than as a line in the labour report.

For operators reading these figures as a benchmark, two things are worth separating. Matching labour to demand more precisely is a scheduling and forecasting problem, and that is where the software helps. Cutting the standard time to clean a room is a different decision with a different consequence, and no system makes it painless.

The useful question to ask of your own numbers is which of the two produced your improvement. If HPOR came down because your forecast got better, that holds. If it came down because sections got bigger, the quarterly slowdown in this data is a preview of what happens next.