Bangkok and Phuket show why hotel occupancy alone does not tell the revenue story

Bangkok sold more rooms without gaining more revenue per room

Bangkok's average hotel occupancy reached 76.2% in the first half of 2026, up 1.1 percentage points year on year. Yet average daily rate (ADR) fell 2.1% to THB 4,013, while revenue per available room (RevPAR) declined 0.6% to THB 3,056.

The split shows why occupancy can be misleading on its own. More rooms were being filled, but operators were not able to preserve the same average pricing. The pressure was strongest in the upscale segment, where ADR fell 4.4% to THB 4,197. Luxury hotels performed differently, with ADR edging up 0.6% to THB 7,010.

Supply is another part of the equation. Knight Frank counted 105,038 operating hotel rooms in Bangkok at the end of H1 and was tracking roughly 17,590 additional rooms in the development pipeline. Most of the dated pipeline is positioned in the luxury and upscale categories, raising the competitive bar for properties that do not have a clear product, location or service advantage.

Phuket traded some occupancy for stronger pricing

Phuket moved in the opposite direction. Average occupancy fell by 3.2 percentage points, from 80.0% to 76.8%, but ADR rose 5.3% to THB 7,117. RevPAR still increased 1.1% to THB 5,465.

That is a rate-led result: a smaller share of available rooms was occupied, yet the higher average room price was enough to keep room revenue growing. It does not mean rates can rise indefinitely. Knight Frank also recorded softer pricing in Phuket's luxury segment, while a large share of future supply is concentrated at the upper end of the market.

A further 2,498 rooms were scheduled to open on the island during the second half of 2026. Some projects may move beyond their planned opening dates, so the pipeline is best read as expected competitive pressure rather than guaranteed delivery on a fixed timetable.

What a private rental investor can — and cannot — take from the data

ADR is the average price of a sold hotel room. RevPAR measures room revenue across all available hotel rooms, whether occupied or not. Neither figure is the same as an owner's net rental yield.

A hotel carries staffing, distribution, marketing, food and beverage and other operating costs that do not map directly onto a privately owned condo or villa. Private rentals also face their own management fees, utilities, maintenance, seasonality and legal constraints. The H1 figures are therefore most useful as market context: they show how much pricing power and competition differ between destinations, but a property decision still needs evidence from comparable residential rentals and the actual cost structure of the asset.