Travel has one of the highest paid-search click-through rates and no standard direct-booking benchmark

Travel runs a 9.32% click-through rate at $2.14 a click, among the highest attention rates of any industry in the published search set, and converts at 5.83% for a $44.70 cost per lead. Underneath that, online travel agencies took roughly a fifth of all US travel gross bookings in 2025 and are projected to reach 21% by 2028, with hotels making up 63% of agency gross bookings at $100.3B. The forecast for 2026 has record room-nights sold in the first half, 11.4 million more than 2025, room revenue up $5.4B and gross operating profit per available room up 4%, against a 63.1% occupancy forecast.

Record volume, 63.1% occupancy, and no published figure for what a hotel's own booking engine converts at.

1. 01

   The commission range is the real cost per acquisition and almost nobody prices it that way

   Published guidance puts commissions at roughly 15% to 30% and above today, up from about 10% a decade ago, with some niche and regional agencies as low as 4%. That guidance is vendor-published and discloses no sample, so it is a planning range and not a measurement. Even at the bottom of it, intermediation is the largest single acquisition line in the business and it appears in no media report anywhere.

2. 02

   The number that would settle the argument is not published

   There is no verified published direct-booking conversion rate for hotels, no published hotel metasearch cost per click, and no current verified average daily rate or revenue per available room dollar figure in this set. Which means every claim that direct is cheaper — including the ones made by direct-booking vendors — is an assertion without a denominator. It is also entirely measurable on a single property's own data inside a week, which is why naming the gap is more useful than filling it with somebody's estimate.

3. 03

   Intermediation is a trend line, not a fixed cost

   A fifth of gross bookings in 2025, projected to 21% by 2028. A plan sized against today's channel mix underestimates the drift, and the properties that notice late are the ones that negotiated their commission rate once and filed it.

4. 04

   Record volume and 63.1% occupancy are the same story told twice

   Room-nights sold are forecast at a record while occupancy sits in the low sixties, which means supply and demand are both growing and neither is the binding constraint. Mix and rate are. That reorders the marketing brief: the useful question is which segment fills which week, not how to generate more undifferentiated demand.

5. 05

   A 9.32% click-through rate is a warning as much as a strength

   Travel intent is broad, emotive and cheap to attract. The 5.83% conversion rate and $44.70 cost per lead say the qualification is happening after the click rather than before it. The model moves dates, party size and flexibility ahead of the enquiry rather than after it, which lowers the click-through rate on purpose.

Travel paid-search CTR **~9.3%** (among the highest of 23 industries) at ~5.8% conversion. No *cross-industry* booking-engine benchmark exists; hotel-tech sources put direct-site conversion at **~2–4%** vs OTA **12–15%**.

The model instruments distribution before it touches media. Phase one builds the channel-by-channel net rate view: what a booking costs through each path once commission, media, incentive and booking-engine cost are included, which almost never exists in one place. Phase two measures the property's own booking-engine conversion rate by device and by market, because that figure is unpublished and every direct-booking decision depends on it. Phase three fixes the booking path itself, on the basis that a booking engine on a phone is where most of the loss occurs. Media comes last and is bought against segment and week rather than against generic destination demand.

1. 1

   Weeks 1–3

   Channel net-rate view

   Cost of a booking through each path, including commission, media, incentive and engine cost

   Owner: OmniFlow + property commercial

2. 2

   Weeks 2–4

   Booking-engine baseline

   Direct-booking conversion rate measured by device and market — the figure the published set does not contain

   Owner: OmniFlow

3. 3

   Weeks 4–9

   Booking path work

   Mobile booking path, rate presentation, cancellation clarity, parity with the listing offer

   Owner: OmniFlow

4. 4

   Weeks 8–16

   Media by segment and week

   Search and metasearch bought against the weeks and segments that need filling, not against destination demand

   Owner: OmniFlow

5. 5

   Month 6

   Review

   Net rate per channel, direct-booking conversion by device, cost per confirmed room night

   Owner: OmniFlow

The model reports cost per confirmed room night by channel, net of commission, and the property's own direct-booking conversion rate by device. The second of those does not exist as a published benchmark anywhere, which is exactly why it is the first thing built: without it, the direct-versus-intermediated decision is being made on the strength of vendor assertion. Commission ranges are used as a planning band and labelled vendor-published throughout, and no average daily rate or revenue per available room figure is modelled because none is verified in this set.

Modelled. Inputs: WordStream by LocaliQ, Google Ads Benchmarks 2026 (13,474 US search campaigns, April 2025 – March 2026); Phocuswright, US Online Travel Agency Market Essentials 2026; Cloudbeds, Guide to OTA Commission Rates (updated August 2026); CoStar / Tourism Economics US hotel forecast, August 2026. Modelled outputs are not a forecast or a guarantee of results. Cited: LocaliQ *2026 Search Benchmarks*; Book Better Direct / SiteMinder 2026.

Every figure below is published research, not a client result. They are the inputs to the arithmetic above, listed so it can be checked rather than taken on trust. The bracketed number points to the full citation at the end of this page.

These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.

Each figure on this page, the system it is read from, and the definition and window it is measured over.

Not applicable — this is a modelled engagement. Its weakest input is the commission range, which comes from a vendor guide with no disclosed sample and covers a spread wide enough — 4% to above 30% — that the choice of point inside it changes the answer entirely. The search benchmark is firmer but pools hotels with agencies, tour operators and transport, whose funnels have almost nothing in common. Both are replaced by the property's own contracted rates and its own engine data in the first month, which is also why that work is phase one rather than phase three.

A modelled capability, not a client account. Figures illustrate what the model produces and are labelled as modelled wherever they appear. See all our work

Full citations for everything cited on this page, with the sample and period each study covers, so you can go and read the original.