Vacation Rental Cancellation Statistics (2026): Bookings, No-Shows, OTA Drag

Hear an AI agent answer an inquiry and qualify a booking in real time: call +1 (218) 636-0234 (Jessica, US). No signup. More numbers on our contact page.

What "Cancellation Rate" Actually Measures in STR

Before the numbers, a definition. "Cancellation rate" in vacation rental reporting is almost always calculated as cancelled reservations divided by gross reservations over a measurement window. It is not the same as the share of nights that fail to materialise, and it is not the same as the guest no-show rate. Different data providers also treat host-initiated cancellations, guest-initiated cancellations, and "altered" bookings differently.

When comparing figures from D-EDGE, IntelliHost and Key Data, and the platforms themselves, the most important thing to check is the denominator. D-EDGE reports the share of booked revenue that is cancelled, IntelliHost and Key Data report the share of reservations, and Airbnb's S-1 reports the share of nights booked. Airbnb splits out refund policies in their host documentation but does not publish a public cancellation rate by policy type. Where I cite a number below, the source and the denominator are specified.

What Is Actually Published About Cancellation by Channel (Booking.com, Airbnb, VRBO, Direct)

Booking.com

• Nearly 44 percent of Booking.com hotel revenue cancelled in France in 2025 (D-EDGE). D-EDGE's March 2026 press release, via TendanceHotellerie, covering a constant panel of more than 5,000 French hotels, found that nearly 44 percent of revenue booked through Booking.com was cancelled in 2025, against a markedly lower level on the direct channel, while the all-channel rate fell from 24.6 to 22.6 percent. The most recent Europe-wide split D-EDGE has published is for 2022: 40 percent of Booking Holdings revenue cancelled, versus 17 percent for direct bookings, across 4,594 European properties. While these studies cover hotels, the same dynamic (free cancellation policies, pay-at-property, lower barrier to speculative booking) applies to STR listings on Booking.com.
• Almost 40 percent across all channels in 2018 (D-EDGE). D-EDGE's earlier panel of 680 European hotels found that almost 40 percent of on-the-books revenue was cancelled before arrival in 2018, with Booking Holdings channels among the highest. Booking Holdings reports its gross bookings net of cancellations, and its FY2025 10-K says its cancellation rate in 2025 was lower than in the prior year, without publishing the rate itself.
• Non-refundable rates are pulling cancellation rates down. D-EDGE attributes the 2025 fall in French hotel cancellations mainly to the rise of prepaid and non-refundable rate conditions, so offering a non-refundable plan alongside a flexible one works with the trend rather than against it.

Airbnb

• About 13 percent of nights booked were cancelled before the pandemic (Airbnb S-1). Airbnb's IPO filing states that nights and experiences cancelled in January 2020 equalled 13 percent of the gross nights and experiences booked that month. Airbnb has not published a cancellation rate since. The most recent large sample is IntelliHost and Key Data's 2026 analysis of 731,000 US reservations, which found short-term rental cancellation rates from 10 to 36 percent depending on the city, highest in the Orlando and Disney corridor where trips are booked far ahead. That sample is not split by channel.
• Cancellations exceeded new bookings in March and April 2020. Airbnb disclosed in its S-1 filing ahead of IPO (filed November 2020 with the SEC, available via Airbnb's investor relations page) that the nights and experiences cancelled in March and April 2020 exceeded the gross nights and experiences booked in those months. The disclosure is a useful stress-test benchmark for STR insurance and cash-flow planning.
• Stricter policies go with higher earnings. Airbnb does not publish cancellation rates by policy. In 2026 data from IntelliHost and Key Data covering 281,192 active two-bedroom listings, Super Strict listings earned about 1.64 times what Flexible listings earned at the same size, a correlation the authors note partly reflects which operators choose strict policies.

VRBO (Expedia Group)

• No published Vrbo cancellation rate. Expedia Group does not report a Vrbo cancellation rate, and we found no primary benchmark that gives one. VRBO's guest base skews toward larger whole-home, family, and longer-stay bookings.
• Booking windows depend more on the market than the platform. We found no primary source comparing Vrbo and Airbnb booking windows. IntelliHost and Key Data's 2026 analysis of 319,553 reservations found median lead times from 3 days in drive-to and city markets to 155 days in fly-to destinations, and their 2026 study of 460,814 reservations found that stays booked six or more months ahead cancelled at 37.5 percent, against 20 percent for last-minute bookings. Long windows mean more cancellations, not fewer.

Direct Bookings

• About 17 percent in European hotels in 2022, with no STR benchmark published. D-EDGE put direct hotel booking cancellations at 17 percent of revenue in Europe in 2022, against 40 percent for Booking Holdings, and its 2025 data for France again shows the direct channel markedly below Booking.com. We found no primary benchmark for direct-booking cancellations at STR property managers. Direct channels allow the property manager to set policy, collect the deposit, and build a relationship at booking, all of which reduce cancellation risk.
• Direct carries no OTA commission, but it has to be fed. The trade-off is traffic: direct bookings require marketing, SEO, and inquiry response capacity that most small operators lack.

Last-Minute Cancellations and Guest No-Shows

• Late cancellations behave like no-shows. A cancellation made the evening before arrival leaves a night almost as hard to resell as a guest who never turns up. This is the "virtual no-show" phenomenon: the guest did not show up, but the reservation was technically cancelled in-window. We could not find a published share of Booking.com cancellations made within 24 hours of check-in in SiteMinder's or D-EDGE's research, so no figure is given.
• Pay-at-property bookings default to higher no-show risk. When nothing is charged at booking, a guest who never arrives leaves the operator with nothing to collect unless the rate plan carries a card guarantee or prepayment. We found no primary SiteMinder or HotelsCombined measurement of pay-at-property no-show rates, so no range is given.
• Airbnb and VRBO have structurally low no-show rates. Because payment is captured at booking, a no-show is simply an unused stay; it does not affect host payout. This is one of the structural advantages OTAs with prepayment have over Booking.com for the host.
• Last-minute bookings are the most reliable ones. In 460,814 reservations analysed by IntelliHost with Key Data in 2026, stays booked six or more months ahead cancelled at 37.5 percent against 20 percent for last-minute bookings. Demand also arrives late: in the same firms' 2026 search panel, about 47 percent of the search interest a night ever receives lands in its final three weeks. That late demand converts only if inquiries are answered fast.

Inquiry-to-Booking Conversion and Response Time

The single largest operator-controlled lever on STR revenue is inquiry response time. This is not an opinion; it is baked into the Airbnb ranking algorithm.

• Airbnb's host dashboard prominently displays "response rate" and "response time" as ranking signals. Airbnb's own host resources and Superhost criteria require a 90 percent response rate, measured on replies within 24 hours, and Airbnb's search ranking guide says it looks at how quickly hosts respond to inquiries on listings that are not instantly bookable. Faster responses correlate with higher booking rates and better search placement.
• Guests often message several listings and book the first good answer. The often-repeated claim that nearly all inquiries go to whoever responds first could not be traced to Airbnb or to any study, so we give no figure for it. The direction mirrors the well-documented first-responder advantage in B2B sales: see our lead response time studies compilation for the underlying research.
• How fast operators actually reply is not published as a benchmark. We found no primary survey from Transparent or Key Data giving a median inquiry response time for independent STR operators, so take your own from your PMS or inbox message log.
• Faster replies go with materially higher revenue. In 2026 data from IntelliHost and Key Data covering more than 164,000 US two-bedroom listings, listings that replied within an hour earned the most; a few hours' delay went with about 12 percent less revenue, a day with about 20 percent less, and a few days with about 32 percent less. The authors note the link is correlational. It is consistent with the MIT and Harvard Business Review speed-to-lead findings for other industries.

Revenue Lost to Unanswered Inquiries

• Many inbound STR inquiries arrive outside local business hours. Evenings, weekends, and early mornings are windows when most small operators are not actively monitoring inboxes. We found no primary measurement of the share in Key Data's or AirDNA's published research, so check the timestamps in your own message log.
• Unanswered inquiries default to the next available listing. Airbnb and VRBO both auto-suggest alternative listings when a host delays or declines. Every minute of delay is a direct transfer of revenue to a neighbour who answered faster.
• At current average daily rates, a single missed 4-night inquiry on a US listing is roughly 650 to 1,550 USD in gross revenue. AirDNA's market data for the 12 largest US short-term rental markets (updated October 2026) shows average daily rates from 164 USD in Houston to 385 USD in Wilmington, North Carolina. Missing one such inquiry per week is roughly 34,000 to 80,000 USD annually for a single unit.

Double-Booking and Channel-Sync Failures

• Double-bookings impose both relocation cost and platform penalties. Airbnb's host cancellation policy names double-booking as a cancellation the host is responsible for and charges 10, 25 or 50 percent of the reservation amount depending on how close to check-in it happens, with a 50 USD minimum, plus a calendar block on the affected dates and possible loss of Superhost status.
• Channel management is a standing operational risk. Every distribution channel added is another calendar that has to agree with the others, and operators running several channels without real-time calendar sync are the ones most exposed to double-booking incidents.
• Channel-manager adoption reduces but does not eliminate the risk. Even with Hostaway, Guesty, Hostfully, or Rentals United in place, manual overrides and rate updates create windows for collision. Operators in the VRMA community routinely cite a small but non-zero annual double-booking incident count.

OTA Commission Drag on Net Revenue

• Airbnb: host service fee of 3 percent (split-fee) or about 15.5 percent (host-only). Airbnb publishes both fee schedules in its Help Centre: under the split-fee structure most hosts pay 3 percent and guests pay 14.1 to 16.5 percent of the booking subtotal; under the single host-only fee most hosts pay 15.5 percent (the rest typically 14 to 16 percent), with no guest service fee. Airbnb's FY2025 10-K says it began moving to the single-fee structure in October 2025, and the single fee is mandatory for hosts using property management software.
• Booking.com: commission set in each property's contract. Booking.com's partner help describes commission as a percentage that varies by country and can vary by property type or location, so there is no single published STR rate to quote.
• VRBO: flat 12 percent commission including payment processing on pay-per-booking listings, per Vrbo's help centre. Taxes and security deposits carry only a 3 percent processing fee, and listings managed through property management software in some regions, including Europe, may pay 12 to 15 percent. Most existing subscription hosts move to the 12 percent pay-per-booking model on 29 October 2026, when Vrbo's new host terms take effect, and Vrbo no longer accepts new annual subscription listings.
• Compounded effect. Using D-EDGE's 2022 European hotel figures, a mix that loses 40 percent of booked revenue to cancellation (the Booking Holdings rate) and pays, for illustration, an 18 percent commission realises about half of gross bookings as net-of-commission revenue. A direct-led mix at D-EDGE's 17 percent direct cancellation rate and zero commission realises about 83 percent, before card processing.

Policy Mix, Seasonality, and Property-Type Breakdown

• Seasonality effect. How far cancellation rates swing between low and peak season depends on the market, and we found no primary AirDNA figure for the spread, so measure it from your own reservation history.
• Property type matters. Larger homes are booked further ahead: across 328,270 US reservations in IntelliHost and Key Data's 2026 data, studios booked a median of about 13 days out and five-bedroom homes about 41 days out, and longer lead times go with more cancellations. We found no primary figure for the cancellation gap by property type itself.
• Length of stay matters, though not always in the direction assumed. D-EDGE's 2019 European hotel panel found that larger reservations with longer lead times were more likely to be cancelled. We found no primary STR figure comparing cancellation rates by length of stay.

Implications for Vacation Rental Managers in 2026

The data points in one direction. If you operate a short-term rental portfolio and you want to lift realised net revenue, the three highest-leverage moves are (1) shift your channel mix toward direct and away from Booking.com, (2) respond to every inquiry faster than your competitors, and (3) enforce consistent cancellation policies that actually reduce the cancellation rate rather than adding friction without effect.

Items (1) and (3) are strategic and slow. Item (2), inquiry response speed, is the one lever you can change this week. The Airbnb algorithm explicitly rewards it. Listings that answer faster earn more. And the underlying economics, documented across every lead-response study ever published, are consistent: faster response always wins more of the inquiries you have already paid to acquire.

How an AI Voice Agent Captures Inquiries That OTA Messaging Misses

Most STR inquiries arrive through OTA message threads, not phone calls. So why does a voice agent matter for vacation rentals? Three reasons, all grounded in the data above.

First, the OTA message channel is not the only inbound channel. Every reputable STR operator runs a direct-booking site, a Google Business Profile, and a listed phone number for concierge and on-property requests. Inquiries on those channels arrive by phone, SMS, and form. A 24/7 AI voice agent answers these within seconds; a human team answers them in hours or not at all.

Second, a voice agent handles concierge calls (lockbox codes, check-in confusion, late-arrival coordination) without waking the property manager. This is operationally relevant to cancellation rates because some same-day cancellations are triggered by friction at arrival: a lost check-in instruction, an unanswered call at 22:00, a guest standing outside in the rain. An always-available voice agent resolves these before they escalate.

Third, a voice agent drives direct bookings. A guest who finds your direct site via Google, calls the number, gets an instant intelligent response, and receives a booking link or live quote is a guest you keep at 0 percent commission and a materially lower cancellation rate. See our deep-dive on AI voice agents for property management and vacation rentals and our comparison of Guesty AI and alternatives for how this slots into an existing PMS stack.