The Billboard Effect: Why OTAs Fund Direct Bookings, Not Steal Them

Booking.com shows up in 95.3% of AI hotel searches tested in a 2026 hospitality study. That sounds like the OTA won. It didn't, not entirely: when an AI model actually names a property, 91.1% of those links point straight to the hotel's own website. The discovery layer and the booking layer are not the same thing, and confusing them is the single most expensive mistake in hotel marketing right now.

This is the billboard effect, and it has quietly become the most important concept in hotel distribution for 2026. Not because it's new. Cornell first documented it in 2009. It's important now because AI search just gave it a second, much bigger stage.

What Is the Billboard Effect in Hotel Marketing?

The billboard effect describes what happens when a traveller discovers a property on Booking.com, Agoda or Traveloka, then books it directly on the hotel's own website instead. The OTA listing works like a physical billboard: it puts the property in front of someone who wasn't looking for it by name, and some percentage of those people go straight to the source afterward.

Cornell University's 2023 update to its original study confirmed the effect "is still alive and well." The mechanism hasn't changed. What has changed is the scale of the discovery layer doing the advertising, and, in 2026, that layer increasingly includes a chatbot.

The Numbers Behind the Loop

Three separate research groups have measured pieces of this cycle, and they converge on a consistent picture:

• Kismet's analysis of Google travel survey data found that 52% of travellers visit a hotel's official site after discovering that hotel on an OTA.
• Triptease reports that up to 60% of travellers who eventually book on an OTA visited the hotel's own website first, meaning the traffic sometimes flows the other direction too.
• Cornell's own data found that roughly 54% of guests who ultimately booked through an OTA had clicked the hotel's brand website at some point during research.

None of these numbers describe a clean, one-directional funnel. Guests bounce between the OTA and the brand site multiple times before deciding. What matters commercially is which channel gets the booking at the end, and that's the part a hotel can actually influence.

Then AI Search Made the Billboard Bigger

Here's where 2026 changes the maths. A 1,548-prompt study of AI hotel recommendations found that OTAs and metasearch sites account for roughly 44% of every citation an AI model makes when answering a hotel question, and a property's own website accounts for about 6%. Booking.com specifically was cited in 95.3% of tested hotel queries in a separate 450-query hospitality dataset. On the surface, that looks like total OTA dominance of the AI layer.

It isn't the whole story. HotelRank.ai's January 2026 study of more than 31,000 hotels found that once GPT-5.2 actually names a property in its answer, 91.1% of the links it provides point to the hotel's own website. Even Perplexity, the model in that study with the strongest OTA lean, still sent 74.7% of its links direct. AI reads from the OTA to build its answer. It sends the traveller to you.

Why This Matters More Than the Citation Rate Itself

Most hotel marketing conversations about AI stop at "are we being cited." That's the wrong finish line. A property obsessing over its 6% website citation share while ignoring its Booking.com and TripAdvisor completeness is optimising the smaller number and neglecting the one that actually drives the AI's answer. Get named accurately, on the strength of a complete OTA and editorial presence, and the direct link tends to follow on its own.

What Each Channel Actually Costs You

The billboard effect is only worth chasing if the maths behind it holds up, and in Indonesia specifically, it does. Booking.com and Agoda typically charge 15%-20% commission per booking. Traveloka runs 18%-24%. A well-run direct booking channel, by contrast, costs roughly 2%-8% of revenue once SEO, booking engine and CRM investment are accounted for. Indonesia also happens to rank second globally for OTA dependence among independent hotels, at 77.9% of bookings, which means the commission bill here is larger than in most comparable markets, and the upside from shifting even a modest share direct is larger too.

Read across that table and a pattern shows up in every row: the channel that costs the most to be discovered on is rarely the channel that ends up keeping the commission. The job isn't to abandon OTAs, and it certainly isn't to abandon AI citation. It's to make sure the discovery layer is complete enough to be trusted, and the booking layer, your own website, is fast, clear and ready to close the loop once the guest arrives.

How to Actually Capture the Billboard Effect

1. Win the Brand-Name Search That Follows the OTA Visit

If a guest saw your property on Booking.com and then searches your name on Google, your own site should be the first organic result, with your Google Business Profile verified and linked. This is table stakes, and a surprising number of independent Indonesian properties still lose this specific search to their own OTA listing.

2. Make the OTA and AI Citation Layer Complete, on Purpose

Since AI reads Booking.com, TripAdvisor and Traveloka to build its answer, an incomplete listing on any of them is a gap in the story AI can tell about you. Full amenity data, current photography and a healthy review response rate aren't cosmetic. They are the raw material the citation is built from.

3. Don't Let Rate Parity Stop You From Competing on Value

Parity clauses restrict the price you can display, not the perks you can offer. Loyalty pricing, bundled extras and free cancellation on direct bookings all sit outside most parity language, and they are exactly the kind of value-add that turns a billboard-driven visit into a direct booking instead of a repeat OTA click.

4. Measure the Loop, Not Just the Click

Standard last-click analytics gives 100% of the credit to whichever channel closed the booking, which erases the OTA's role in discovery entirely. Track branded search volume as a leading indicator of billboard exposure, and compare booking-engine-level direct revenue across periods of strong versus weak OTA and AI visibility. The billboard effect is invisible in a dashboard that only looks at the final click.

Frequently Asked Questions

What is the billboard effect in hotel marketing?

It's the pattern where a traveller discovers a property on an OTA like Booking.com or Traveloka, then books it directly through the hotel's own website instead of through the OTA. The OTA listing functions like a billboard: it creates awareness, and some share of that awareness converts direct.

Does the billboard effect still work in 2026, now that AI search exists?

It works at a larger scale. AI models cite OTAs like Booking.com to build their answers, but once a property is named, the majority of links an AI provides point to the hotel's own website rather than back to the OTA. The billboard got a bigger stage, not a smaller one.

Should we stop paying for OTA listings if we want more direct bookings?

No. OTA and AI citation completeness is what creates the discovery that leads to the direct booking in the first place. The goal isn't to disappear from OTAs, it's to make sure your own website and branded search presence are strong enough to capture the guest once they come looking for you by name.

How much does the billboard effect actually cost in commission?

Indonesian hotels typically pay 15%-20% commission to Booking.com and Agoda, and 18%-24% to Traveloka, against a direct-channel cost of acquisition of roughly 2%-8% once SEO and booking engine investment are factored in. Indonesia ranks second globally for OTA dependence among independent hotels, at 77.9% of bookings, so the commission exposure here is higher than in most markets.

What's the fastest way to convert OTA-driven awareness into a direct booking?

Win your own brand-name search, keep your Google Business Profile complete and verified, and offer parity-compliant value adds like free cancellation or loyalty pricing that give a guest a reason to book direct once they land on your site. None of this requires touching the price the OTA displays.

For a deeper look at how AI citation layers behave differently across languages and platforms, our Hotel GEO service covers the ecosystem work in full, and our Hotel SEO service covers the destination content and booking-engine fixes that convert the traffic once it arrives. Both are explored at length in Tessar Napitupulu's Cited or Silent, which maps how AI citation and direct conversion interact across industries beyond hospitality.