Direct booking strategy
The Billboard Effect: Does Listing on Booking Sites Bring You Direct Bookings?
Yes, according to Cornell research. In a 2009 test, four hotels got 7.5% to 26% more reservations through their own channels while listed on Expedia, not counting Expedia's own bookings. A 2011 Cornell study found three to nine brand-website bookings for each Expedia booking. The effect is real, but it only pays if guests can find and book you directly.
Owners who want to pay booking sites less often hear the same reply: being listed on Expedia and Booking.com brings you direct bookings too. That claim has a name, the billboard effect, and it comes from research at Cornell University's School of Hotel Administration. This guide sets out what the three studies found, read on Cornell's own pages on October 12, 2026, what the numbers do and do not prove, and what to do about it.
What is the billboard effect for hotels?
It is the extra bookings a hotel gets through its own website, phone and front desk because travelers first saw it on a booking site. Cornell professor Chris Anderson coined the term in 2009, comparing booking sites to billboards along a highway: many travelers visit them early in their search, learn about a hotel's location, rates and reviews, and then book through a channel the hotel controls (2009 Cornell Hospitality Report).
The effect is measured separately from the bookings the booking site makes itself. So the question it answers is: if your hotel disappeared from Expedia, would your direct bookings fall too?
What did Cornell's 2009 billboard effect study find?
From October to December 2008, Expedia switched four hotels run by JHM Hotels on and off its site in cycles of seven to eleven days, 40 days on and 40 days off in total. While on, the hotels appeared at the top of the first page; while off, they could not be found on Expedia even by name. Anderson then compared the reservations each hotel took through every channel except Expedia.
| Hotel | Daily reservations, listed | Daily reservations, not listed | Increase |
|---|---|---|---|
| Branded hotel 1 | 39.4 | 36.6 | 7.5% |
| Branded hotel 2 | 54.8 | 49.7 | 9.1% |
| Branded hotel 3 | 39.1 | 34.2 | 14.1% |
| Independent hotel | 28.2 | 22.3 | 26% |
Reservations made through channels other than Expedia. Source: Cornell Hospitality Report, Vol. 9, No. 16, October 2009, Exhibit 1.
The independent hotel saw the biggest lift. Anderson wrote that its data were the cleanest, because the branded hotels had sister hotels nearby that stayed on Expedia, so travelers could still find their brand there during the "off" days. He also cautioned that the bigger benefit to the independent may partly come from that difference in the test. All four hotels also recorded a slightly higher average daily rate while listed, from 0.3% to 3.9%.
The test covered four hotels, one of them independent, over three months in 2008, and Expedia ran the listings for the study. It is the best controlled test there is, but it is small and old. Treat 7.5% to 26% as a sign that the effect exists, not as a figure to plan your budget on.
What did the 2011 and 2017 studies add?
Two follow-ups looked at how travelers search before they book.
- 2011. Anderson studied 1,720 reservations at InterContinental Hotels Group brands for June to August of 2008, 2009 and 2010. Cornell's summary says that for each reservation an IHG hotel received at Expedia, the brand's own website received three to nine more, made on the hotel's site but created or influenced by the Expedia listing. The average guest made twelve visits to a booking site before booking, viewing 7.5 pages per visit (Cornell's summary of the 2011 study).
- 2017. Anderson and Saram Han used comScore data on more than 50,000 people and their 2015 online behavior, including 13,000 travel reservations. Cornell's announcement says the average hotel shopper made 25 visits to travel sites before buying (Cornell's 2017 announcement). The report's summary says the influence of booking sites on the final reservation may be lower than before, but the effect still occurs because many people visit a booking site before booking direct (Cornell's summary of the 2017 report).
So the finding has held across three studies, but it has weakened, and the newest data is from 2015. We found no newer published test that switches hotels on and off a booking site.
Does the billboard effect make booking-site commission cheaper?
Anderson argued that it does, on paper. His 2009 example: a hotel pays 15% commission, the booking site brings 10% of its reservations, and 60% come through direct channels. If the billboard effect is 20%, the listing also creates about 10 points of those direct reservations (60 minus 60 divided by 1.2). Spread the commission over both, and the effective rate halves to 7.5%.
The sum rests on three conditions that are easy to miss:
- The guest must find you. A traveler who leaves the booking site and searches your name will only book direct if your own website and rate show up first, and not another booking site's ad.
- The effect is not the same size every year. The 2017 study found it smaller than in 2009, and booking sites today cost more than 15% once add-ons are counted, usually 20% to 30% (see the real cost of an OTA booking).
- Paid programs are a separate question. The research measured being listed against not being listed. It did not test whether paying extra for a higher spot, through Preferred Partner or Visibility Booster, adds more direct bookings. See Booking.com's Preferred Partner program for that break-even.
How do you turn booking-site lookers into direct bookings?
Make the second search easy. Cornell's 2017 advice to hotels was to make sure their online presence is easy to find, attractive and holds up against the competition. In practice:
- Own your name on Google. Claim and fill in your Google Business Profile (see Google Business Profile for hotels), and get Google's free booking links so your own rate sits next to the booking sites'.
- Show the same price. Under Booking.com's standard US terms your public rate cannot undercut it, but a guest who sees the same total price on your website has every reason to book there. Our guide to rate parity for US hotels explains what the terms allow.
- Make booking on a phone quick. Test your booking page yourself, start to finish.
- Track where each booking comes from, so you can see the effect in your own numbers. See tracking which ad brought a booking.
How can you test the billboard effect at your hotel?
The 2017 study suggested the method: stop one action for a short period, such as preferred placement on a booking site or sponsored search on Google, and compare sales before, during and after. For an independent hotel, the cheapest version is to pause a paid program, not to leave the booking site.
| Step | What to do |
|---|---|
| 1. Pick one change | Pause one paid program, such as Visibility Booster, for two to four weeks |
| 2. Pick comparable dates | Compare with the same weeks last year, or with the weeks just before |
| 3. Count every channel | Direct bookings, booking-site bookings and revenue, not just the booking site's report |
| 4. Compare what you keep | Revenue after commission and discounts for each period |
Read your contract before pausing any program, and change one thing at a time so the result is clear.
If direct bookings hold up while the program is paused, you were paying for visibility you already had. To see what moving a share of bookings to direct is worth, try our savings calculator. For the full plan, read how independent hotels get more direct bookings.
MarginStay helps independent hotels bring in direct bookings, so we have a stake in this topic. The study figures come from the 2009 Cornell Hospitality Report and from Cornell's own summaries of the 2011 and 2017 reports, read on October 12, 2026. The 2011 and 2017 full reports sit behind a bot check, so we quote Cornell's published summaries of them.
Billboard effect: quick answers
What is the billboard effect in hotels?
It is the extra bookings a hotel gets through its own website, phone and front desk because travelers first saw it on a booking site such as Expedia. Cornell professor Chris Anderson named it in 2009, after a test in which four hotels took 7.5% to 26% more non-Expedia reservations while listed on Expedia.
Is the billboard effect still real?
Cornell's 2017 follow-up, using 2015 data on more than 50,000 people, found that the effect still occurs because many travelers visit a booking site before booking direct, but that a booking site's influence on the final reservation may be lower than in 2009. We found no newer published on and off test.
Does the billboard effect mean OTA commission is worth paying?
Partly. Being listed brings some direct bookings, so a listing is worth more than its own bookings show. But the research measured being listed, not paid programs for higher placement, and the effect only turns into direct bookings if guests can find and book your own website easily.
How do I measure the billboard effect for my hotel?
Change one thing for a short period and compare. Pause one paid booking-site program for two to four weeks, then compare direct bookings, booking-site bookings and revenue after commission with similar dates. Cornell's 2017 study suggested this kind of short stop-and-compare test. Check your contract first.
Turn booking-site lookers into direct bookings
Talk to us about making sure guests who find you on a booking site can find and book you directly, or see first what your booking-site bookings cost.
Sources
- Cornell Center for Hospitality Research: The Billboard Effect, Online Travel Agent Impact on Non-OTA Reservation Volume (2009)
- Cornell eCommons: Search, OTAs, and Online Booking, An Expanded Analysis of the Billboard Effect (2011)
- Cornell eCommons: The Billboard Effect, Still Alive and Well (2017)
- Cornell SC Johnson College of Business: Online travel sites are still important for hotel operators (2017)