"Book direct, it's always cheaper" is one of those travel tips everyone repeats. Hotels push it, OTAs push the opposite. I had the data to check it, so I did.
Disclosure: the data comes from a Google Hotels scraper I built and sell on Apify, and an AI assistant helped me draft this post. The numbers are from runs I did on November 17, 2026.
The setup
- 5 cities: Rome, Paris, London, Barcelona, New York
- Same stay everywhere: January 15–17, 2027, 2 adults, prices in USD
- The top 20 hotels Google Hotels shows for each city, with every room offer it lists per booking site
That's 100 hotels. 85 had room offers. 23 had both an offer from the hotel's own website and at least one from a booking site (Booking.com, Expedia, Hotels.com, Travelocity, Priceline, Agoda and a few others). Those 23 are the sample.
For each hotel I compared the cheapest total on the official site with the cheapest total on any booking site.
Is it cheaper to book directly with the hotel?
| Result | Hotels |
|---|---|
| Official site cheaper | 8 |
| Same price | 8 |
| Booking site cheaper | 7 |
Basically a three-way split. The median difference was exactly 0%.
The gaps were not small when they existed:
| Hotel | City | Official site | Cheapest booking site | Difference |
|---|---|---|---|---|
| Paramount Times Square | New York | $349 | $436 (Travelocity) | −20% |
| Petit Palace Boqueria Garden | Barcelona | $273 | $335 (Expedia) | −19% |
| Premier Inn London Woolwich | London | $148 | $179 (Expedia) | −17% |
| H10 Montcada | Barcelona | $293 | $351 (Hotels.com) | −17% |
| Crowne Plaza Paris République | Paris | $469 | $401 (goseek) | +17% |
| Kimpton Fitzroy | London | $778 | $638 (dealbase) | +22% |
| The Cumberland | London | $399 | $322 (dealbase) | +24% |
| Life Hotel | New York | $508 | $406 (Booking.com) | +25% |
So "always book direct" is wrong about a third of the time, and "OTAs are always cheaper" is wrong about two thirds of the time.
Booking direct or Booking.com, Expedia: what the pattern looks like
A few things stood out:
- Chains with their own booking engines often matched or beat the OTAs. H10, Premier Inn, NH Collection, Novotel, Hyatt, DoubleTree were either cheaper direct or exactly the same.
- When an OTA won by a lot, it was often a smaller seller (dealbase, goseek) rather than Booking.com or Expedia.
- Same price on many sites is common. That's hotel rate parity in action: many contracts ask the hotel to show the same rate everywhere.
One honest caveat on method: room names are written differently on each site, so I compared the cheapest offer of any room, not the identical room. Two rooms called "Standard Double" and "Classic Room" may not be the same product.
What is hotel rate parity, and how do you check it?
Rate parity means a hotel shows the same price for the same room on its own site and on every OTA. Revenue managers check it daily, because a cheaper OTA steals direct bookings (and the hotel pays it a commission), while a cheaper direct rate can break a contract.
Checking it by hand means opening a dozen tabs per hotel per date. Checking it with code is one call:
import os
from apify_client import ApifyClient
OTAS = {"Hotels.com", "Expedia.com", "Travelocity.com", "Booking.com",
"Priceline", "Agoda", "Trip.com", "dealbase.com", "goseek.com"}
client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("monkey.d.scraping/google-hotels-room-offers").call(run_input={
"hotels": ["H10 Montcada Barcelona", "Hotel Catalonia Ramblas Barcelona"],
"checkIn": "2027-01-15",
"checkOut": "2027-01-17",
"currency": "USD",
"maxOffersPerHotel": 60,
})
offers = list(client.dataset(run.default_dataset_id).iterate_items())
for hotel in sorted({o["hotelName"] for o in offers}):
rows = [o for o in offers if o["hotelName"] == hotel]
direct = [o["stayTotalDisplayed"] for o in rows if o["provider"] not in OTAS]
ota = [o for o in rows if o["provider"] in OTAS]
if direct and ota:
best = min(ota, key=lambda o: o["stayTotalDisplayed"])
gap = (min(direct) - best["stayTotalDisplayed"]) / best["stayTotalDisplayed"]
print(f"{hotel}: direct ${min(direct):.0f} vs {best['provider']} ${best['stayTotalDisplayed']:.0f} ({gap:+.0%})")
Each offer row also carries the room name, nightly price, free cancellation, cancellation deadline and breakfast when the site states them, so you can compare like for like before shouting "parity breach".
The five-city run cost about $0.18 in total. Schedule it daily on Apify and you have a free-ish rate parity checker.
So, should you book direct?
Check both, every time. In my sample the two options were a coin flip with a third side. If you're a traveller: compare the official site with the cheapest OTA for your exact dates. If you run a hotel: someone is probably doing this to you already.
FAQ
Is it cheaper to book directly with the hotel or Booking.com? In this sample, the official site was cheaper in 8 of 23 hotels, the same in 8 and more expensive in 7. It depends on the hotel and the date.
What does rate parity mean in hotels? The same room at the same price on the hotel website and on every online travel agency.
How do I check rate parity automatically? Collect the offers per booking site for your hotels and dates (the script above), then compare the official site with the cheapest OTA.
Where does the data come from? Google Hotels, which lists the price each booking site quotes for a hotel. I use my Google Hotels API on Apify.
If you want the raw numbers for a city, ask in the comments and I'll run it.
Top comments (0)