Holiday hotel prices 2026: another site listed Thanksgiving week 5% or more below Booking.com at 97% of the hotels we compared

1. Thanksgiving week, November 25 to 29, 2026: how often and by how much

   Another site listed the stay 5% or more below Booking.com at 74 of the 76 hotels compared (97%; 95% interval 91% to 99%). The median gap there was 22% (95% interval 18% to 25%), about $40 a night. Any amount lower, $1 a night or more: 74 of 76 (97%), a median 22%, about $40 a night. Booking.com's price was the lowest or tied at 2 of the 76 (no other site at least $1 a night below). The other 24 of the 100 hotels could not be compared: no price on Google Hotels for the dates (3), Booking.com not listed (7), Booking.com listed only for another number of guests (14). Priced November 3, 2026, 22 days before check-in.

2. Christmas week, December 23 to 27, 2026: how often and by how much

   Another site listed the stay 5% or more below Booking.com at 74 of the 79 hotels compared (94%; 95% interval 86% to 97%). The median gap there was 20% (95% interval 19% to 24%), about $32 a night. Any amount lower, $1 a night or more: 77 of 79 (97%), a median 20%, about $32 a night. Booking.com's price was the lowest or tied at 2 of the 79 (no other site at least $1 a night below). The other 21 of the 100 hotels could not be compared: no price on Google Hotels for the dates (1), Booking.com not listed (1), Booking.com listed only for another number of guests (16), Booking.com listed only without a tax split (3). Priced November 4, 2026, 49 days before check-in.

3. Which sites listed lower

   For Thanksgiving week, the lowest of the lower listings was most often on Traveluro (18), then Super.com (17), then Hotelscombined.com and KAYAK.com (15 each); the hotel's own website was the lowest at 4 of the 74. For Christmas week, the lowest of the lower listings was most often on Super.com (24), then KAYAK.com (16), then Hotelscombined.com and momondo.com (15 each); the hotel's own website was the lowest at 4 of the 74. A tie counts for each site, so the counts can add up to more than the hotels. The sites are named as Google Hotels lists them.

4. Rooms and cancellation terms

   For Thanksgiving week, the lowest-priced other listing named a room at 3 of the 74 hotels and showed free cancellation at 6; Booking.com's listing at those hotels named a room at 59 and showed free cancellation at 38. For Christmas week, the lowest-priced other listing named a room at 5 of the 74 hotels and showed free cancellation at 7; Booking.com's listing at those hotels named a room at 60 and showed free cancellation at 31. Naming a room is not proof it is the same room, and a lower price can come with stricter terms, so check the room and the cancellation terms before you book.

5. Taxes, fees and what was left out

   Every price here is a nightly rate averaged over the 4 nights, with taxes and fees as each site reports them; a resort or destination fee may be extra. Google Hotels' own description of the hotel names a resort, destination or similar fee at 5 of the 76 hotels compared for Thanksgiving week and 5 of the 79 for Christmas week; at 5 and 5 of them another site was 5% or more below Booking.com, a gap that may come from a lower price leaving that fee out. Listings that gave no tax split were left out, because their price may be before taxes; counting them as listed, the share would be 97% for Thanksgiving week and 93% for Christmas week. Sign-up prices (a lower price for joining the hotel's or a site's program) were left out too; counting them would have added no hotel to either week.

The same comparison on earlier days

Context, not the headline: the same 100 hotels and the same stays, from our earlier holiday snapshots, with the same rules (no new search). Each row says how many hotels could be compared, at how many another site was 5% or more below Booking.com, and the median gap there. Same hotels, both weeks: 4 had a listing 5% or more below Booking.com only for Thanksgiving week and 0 only for Christmas week, of the 72 compared for both; the difference is within what chance alone could produce in a sample this size (exact McNemar test, p = 0.125).

What this shows, and what it does not

It shows what sites listed on Google Hotels for 100 hotels, 4 in each of 25 large US travel markets, on the day we priced each week. It is not every hotel, not every site, and not a forecast: prices for these stays will move before the holidays, and your hotel and dates can come out differently. It does not say Booking.com charges more in general. A lower listing may be for a different room or come with different terms, and prices include taxes and fees as each site reports them; a resort or destination fee may be extra. The hotels were chosen by a rule fixed before any search (Google's order, and a price on October 12, 2026), never by Booking.com or by any price; hotels sold out on Google Hotels that day are not among them. Four hotels a city are too few for city figures, so the page gives none.

Looking at a Black Friday or Cyber Monday hotel sale? This study priced holiday stays, not sale prices: is a Black Friday hotel sale price lower than other sites?

Holiday hotel prices and Booking.com: common questions

Method and data

Written down before any search (pre-registered November 2, 2026), with the collection code of the hotel price dispersion study and the scoring code of its weekly snapshots. The hotels are the 100 of our Thanksgiving hotel prices repeat of October 27, 2026: in each of 25 cities that anchor large US hotel markets, the first 4 hotels, in Google's order on that city's newest study snapshot, that had a Thanksgiving going rate on October 12, 2026. Each hotel was priced by its Google property token, once for November 25 to 29, 2026 and once for December 23 to 27, 2026, each week within one day (UTC), for 2 adults and one room, as a traveler in the US, in US dollars: 200 searches in all (100 for Thanksgiving week and 100 for Christmas week), one a hotel a week.

For each hotel and week we read every listing Google Hotels returned. A listing counts when it is priced for 2 guests and gives its rate both before and after taxes and fees, so its price is known to be all in as that site reports it; the price is the nightly rate averaged over the 4 nights, and the public price where a sign-up price is shown beside it. A site's price is its lowest such listing; the hotel's own website counts as a site. A hotel is compared when Booking.com and at least one other site have such a price. Another site is lower when its price is 5% or more below Booking.com's; the gap is Booking.com's price minus the lowest other price. Shares carry 95% Wilson intervals and median gaps distribution-free 95% intervals; a share is given as a percentage only from 40 compared hotels and a median only from 10 lower listings.

Per-hotel data, both weeks (CC BY 4.0): holiday-hotel-prices-2026.csv. One row per hotel and week: Booking.com's price, the lowest other price and its site, the gap, and why a hotel could not be compared. Each hotel's measured prices from the study, all in: hotel prices by city.