Phuket Hotel RevPAR Fell 8.7% in First Half

Phuket hotels took 6.04 million guests between January and June 2026, a fall of 2.72% year on year, according to Cushman & Wakefield figures reported on 28 July 2026 by The Star and VnExpress International. Average occupancy came in at 80.0%, down from 84.1% in the first half of 2025. Average daily rate fell 4% to 6,820 baht, and revenue per available room, which combines the two, fell 8.7% to 5,456 baht from 5,975 baht.

The consultancy attributed the softening to international arrivals falling away from April onward, travel confidence dented by unrest in the Middle East, and higher airfares weighing on long-haul demand from Europe and the United States.

Why RevPAR Is the Number That Matters

Occupancy and rate can each be managed at the expense of the other. An operator can fill a hotel by discounting, or protect rate by accepting empty rooms, and either move looks respectable when reported on its own. RevPAR multiplies the two, so it is the figure that cannot be gamed by trading one against the other.

An 8.7% RevPAR decline is materially worse than either input suggests in isolation, and that is exactly the point. A private owner letting a condominium faces the same arithmetic with less pricing sophistication and a smaller inventory to average across.

What This Means for a Rental Underwrite

Hotel data is not a direct read on condominium income. Hotels have front desks, dynamic pricing teams and corporate distribution that a single owner does not, and they sit in a different competitive set. What hotel data does give you is the demand backdrop, and a backdrop where rate and occupancy both slipped is one in which optimistic private projections deserve scrutiny.

Our standing editorial position on Phuket yields has not moved: underwrite a managed condominium at 4 to 7% net after management fees, common area charges, vacancy and tax, against brochure gross figures of 8 to 12% that should be treated as ceilings rather than forecasts. A half year in which island ADR fell 4% is a reminder of why the gap between the two numbers exists.

Getting shown a yield projection you cannot check?

We rebuild the model on the specific building: real occupancy, the operator's actual fee stack, common charges and vacancy. Sometimes the answer is that the deal does not work.

Get an independent underwrite WhatsApp

Supply Is the Other Half of the Story

Demand softened into a market that is still adding rooms, and the new rooms are concentrated at the top. Knight Frank Thailand figures reported in August put Phuket at 47,195 hotel rooms at the end of the first half, roughly 58% of them luxury or upscale, with about 89% of rooms scheduled to open between 2026 and 2028 falling into those same categories.

For an investor, that is a competitive-set warning rather than a market verdict. A one-bedroom condominium marketed on resort-style amenity will spend the next three years competing against professionally operated luxury rooms that did not exist when the project was launched. Buildings with a genuine differentiator, whether that is location, layout or a management team that actually performs, will hold rate. Generic stock will discount.

Reading This Against the Sales Data

The same summer produced a very different headline from the sales side, with CBRE Thailand reporting Phuket luxury residential sales up more than 45% in the first half. Both can be true at once, because they measure different things: one is what tenants and guests paid for a night, the other is what buyers paid for an asset.

When those two diverge, the honest conclusion is that capital is arriving faster than income is growing. That is a normal phase in a resort market and it is not a crisis, but it is a phase in which the discipline that protects a buyer is underwriting on today’s operating numbers rather than on the sales momentum.

Frequently Asked Questions

What is RevPAR and why does it matter to a condo buyer?

Revenue per available room multiplies occupancy by average daily rate, so it captures the two levers together and cannot be flattered by trading one against the other. For a condominium owner the equivalent figure is annual rental revenue divided by the number of nights the unit was available, and it is the number that should drive a purchase decision rather than a headline occupancy or nightly rate quoted in isolation.

Does an 80% hotel occupancy figure mean my condo will be 80% occupied?

No. Hotels have front desks, dynamic pricing teams and corporate distribution channels that a single owner does not, and they compete in a different set. A well-managed Phuket condominium on short-stay letting typically underwrites below the island hotel average, which is why our editorial guidance models 4 to 7% net rather than applying hotel occupancy directly to a unit.

Why did Phuket hotel performance soften in the first half of 2026?

Cushman & Wakefield pointed to international arrivals falling away from April onward, travel confidence affected by unrest in the Middle East, and higher airfares reducing long-haul demand from Europe and the United States. Thailand's national arrivals were also running below the prior year through the period.

Is Phuket oversupplied with hotel rooms?

Knight Frank Thailand put the island at 47,195 hotel rooms at the end of the first half of 2026, with about 58% of existing stock and roughly 89% of the 2026 to 2028 pipeline in the luxury and upscale categories. That is concentration at the top of the market rather than blanket oversupply, and it means new luxury inventory is competing hardest for exactly the guest that resort-style condominiums target.

Should softer hotel numbers stop me buying in Phuket?

It should change what you pay and what you assume, not necessarily whether you buy. A half year of falling rate and occupancy is a reason to underwrite conservatively, to discount any projection that assumes last year's performance, and to negotiate on price. Buyers whose model still works on today's numbers are in a stronger position than buyers who needed the market to improve.

Read Also: