BRENNEN

Industries / Construction and property / Real estate / Short-term rentals

Pricing and market data

This is setting a home's nightly price and fees, showing guests what they will pay, and gathering the booking and occupancy figures that prices are set from.

7

Milestones achieved, from 2015 to 2025.

2

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. 2015

    The site suggests a price

    At its host conference in Paris on 12 November 2015 Airbnb announced Smart Pricing, a tool that sets a listing's nightly price from Airbnb's own data on demand, in a limited roll-out. Airbnb engineers later described the system at KDD 2018: one model estimates the chance a listing is booked at a given price, and a second, trained with a loss function written for the purpose, sets the suggested price; it powers both Price Tips and Smart Pricing.

    thenextweb.comkdd.org

    Credit

    Airbnb (Peng Ye and colleagues)

  2. 2015

    Outsiders estimate occupancy

    Murray Cox founded Inside Airbnb in February 2015 to publish data gathered from Airbnb's public pages, city by city: listings, calendars and reviews, now free to reuse under a Creative Commons licence. Its occupancy model turns reviews into estimated bookings by assuming that half of all stays get a review. AirDNA, founded by Scott Shatford, has sold estimates of occupancy and revenue built from the same kind of data since 2015.

    LessonMost published occupancy figures for holiday rentals are estimates from scraped pages, and they rest on assumptions such as how many guests leave a review.

    insideairbnb.cominsideairbnb.comairdna.co

    Credit

    Murray Cox; Scott Shatford

  3. 2019

    Europe requires the full price

    EU consumer protection authorities, coordinated by Norway's consumer authority, Forbrukertilsynet, issued a common position on Airbnb's practices on 5 July 2018. Airbnb committed to show the total price of a booking including service and cleaning fees, and on 11 July 2019 the European Commission published the changes it had made; Booking.com and Expedia Group made EU-wide commitments on price display in December 2019 and December 2020.

    commission.europa.eukkv.fi

    Credit

    the EU Consumer Protection Cooperation network; the European Commission; Forbrukertilsynet

  4. 2020

    Platforms report nights to Eurostat

    Under agreements signed in early 2020, Airbnb, Booking, Expedia and, until the end of 2024, Tripadvisor send the EU statistics office, Eurostat, data on every stay booked through them. Eurostat publishes stays, nights rented and guest nights by region and month from 2018 onward, as totals across platforms only, so that no platform's share can be read off.

    ec.europa.eu

    Credit

    Eurostat; Airbnb, Booking, Expedia Group and Tripadvisor

  5. 2021

    Uneven uptake widens a gap

    A study of Airbnb's Smart Pricing in Marketing Science (September 2021) found that hosts who adopted it cut their average nightly rate by 5.7% and raised their daily revenue by 8.6%. Before the tool, White hosts earned $12.16 a day more than comparable Black hosts; among adopters the gap fell by 71.3%, but Black hosts were much less likely to adopt, so across all hosts the gap grew.

    LessonA tool that helps the people who use it can widen a gap if those it would help most do not take it up.

    doi.org

    Credit

    Shunyuan Zhang, Nitin Mehta, Param Vir Singh and Kannan Srinivasan

  6. 2023

    A dollar sign costs millions

    Australia's Federal Court ordered Airbnb Ireland to pay A$15 million in penalties, the Australian Competition and Consumer Commission announced on 20 December 2023. From January 2018 to August 2021 Airbnb had shown prices for Australian stays with a bare dollar sign and charged some guests in US dollars; about 63,000 consumers made more than 70,000 such bookings, over 2,000 complained, and Airbnb offered up to A$15 million in compensation.

    LessonThe currency is part of the price, and a default that fails for some users counts as a misleading price.

    accc.gov.au

    Credit

    Australian Competition and Consumer Commission

  7. 2025

    Total price shown by default

    From 21 April 2025 Airbnb showed every guest worldwide the price including all fees, before taxes, by default in search results, and removed the switch that had made this optional. It had tested total prices in parts of Europe, Australia, Canada and Korea from 2019 and added the optional switch in the US and about 200 other countries in 2022; Airbnb said 17 million guests had used it.

    techcrunch.com

    Credit

    Airbnb

Watch for

  1. An AI pricing model

    Airbnb says it is building "an entirely new pricing model" powered by AI, which Brian Chesky called "one of the biggest single levers for growth that we have". Its autumn 2026 update added a dynamic pricing tool for hosts who opt in from October 2026: it recommends a price range and moves the nightly price within it with daily supply and demand. Airbnb gave no figures for its effect.

    Achieved whenAn outside study compares revenue and occupancy for hosts using the tool with similar hosts who do not, as Zhang and colleagues did for Smart Pricing.

    fool.comrentalscaleup.com

    Claimed by

    Airbnb (Brian Chesky), August and September 2026

  2. One fee, keener prices

    Airbnb is moving all listings to a single 15.5% service fee in place of separate host and guest fees. Its finance chief, Ellie Mertz, said in August 2026 that about half of active listings were on it and that it puts "a kind of downward pressure on pricing"; Brian Chesky said prices "have become more competitive".

    Achieved whenOutside price data for the same homes show guests' totals fell after each home moved to the single fee.

    fool.comrentalscaleup.com

    Claimed by

    Airbnb (Ellie Mertz and Brian Chesky), August 2026

Open

  1. 01

    Check scraped occupancy estimates

    A public comparison, region by region, of occupancy estimates made from scraped listing pages against Eurostat's figures from the platforms' own records. The estimates would include Inside Airbnb's review-based model, a published dataset of daily occupancy from the pricing firm Wheelhouse, and any vendor figures a team can license. The comparison must allow for Eurostat's totals covering three platforms while most scraped sets cover one.

    The brief
    Credit · unclaimed
  2. 02

    Track shown versus paid prices

    A repeatable check that records, for the same homes and dates, the first price a guest sees in search results and the total on the last page before payment, on Airbnb, Vrbo and Booking.com in several countries, and publishes the gaps by platform, country and type of fee.

    The brief
    Credit · unclaimed

Open 01 · the brief

Check scraped occupancy estimates

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A public comparison, region by region, of occupancy estimates made from scraped listing pages against Eurostat's figures from the platforms' own records. The estimates would include Inside Airbnb's review-based model, a published dataset of daily occupancy from the pricing firm Wheelhouse, and any vendor figures a team can license. The comparison must allow for Eurostat's totals covering three platforms while most scraped sets cover one.

Why it is still open

Eurostat publishes guest nights for NUTS 3 regions, the EU's statistical areas of roughly county size. AirDNA's own comparison with Eurostat (May 2026) was made country by country on indexed monthly patterns, where it found a 0.97 correlation, and its totals ran about 38% higher than Eurostat's (445 million against 321 million nights in the EU in 2025), a gap it puts down to wider coverage. Inside Airbnb's model rests on an assumed review rate of 50%. The one independent check found, by the economists Vinod Agarwal, James Koch and Robert McNab, compared AirDNA's listing data for Virginia from 2015 to 2019 with hotel-industry definitions and found its occupancy biased upward by 60% on average; it did not use platform records. Searches of the web, Crossref and GitHub found no check against Eurostat at regional level.

airdna.coinsideairbnb.comdoi.orgec.europa.eu

Built with what exists

Eurostat's datasets tour_ce_oan3 and tour_ce_omn12 through its free data API; Inside Airbnb's city files; the Wheelhouse dataset of daily occupancy for 500 markets from 2017 to 2022 (Aastveit and colleagues, September 2026); Python with pandas.

arxiv.org

  1. First step

    Download tour_ce_oan3 and the Inside Airbnb files for five European cities and compare their nights for 2024 and 2025.

  2. Done looks like

    A published table for at least 40 NUTS 3 regions over at least three years giving each estimate, the Eurostat figure and the percentage gap, with the review rate that would make Inside Airbnb's model match, and the code.

Open 02 · the brief

Track shown versus paid prices

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A repeatable check that records, for the same homes and dates, the first price a guest sees in search results and the total on the last page before payment, on Airbnb, Vrbo and Booking.com in several countries, and publishes the gaps by platform, country and type of fee.

Why it is still open

EU commitments since 2019 and the US fee rule require the total price up front, and an Australian court has fined Airbnb over a misleading price display, yet Airbnb's default total still leaves out taxes. No systematic audit of holiday-rental platforms turned up: the consumer group Which?'s 2023 study of drip pricing, where fees are added during checkout, covered airlines, event tickets and takeaway food, and the EU consumer authorities' check of 314 online traders during the 2025 Black Friday sales, published on 26 March 2026, found drip pricing at 10% of them without naming any lodging site.

techcrunch.comwhich.co.ukcommission.europa.eu

Built with what exists

A web agent such as Browser Use (see the Note "Deleting 80% to start over") or Playwright scripts driving Chrome, set to stop before any payment page; full-page screenshots; an AI model that reads images, such as Claude Fable 5.2 (see the Note "Three working devices"), to read fees off the screenshots, with a person checking a sample.

  1. First step

    Pick 25 homes listed on both Airbnb and Vrbo in one city and record both prices by hand for one set of dates.

  2. Done looks like

    300 home-and-date pairs on each of three platforms in four countries, each with screenshots of the first and final price, and a published table of the median and largest gaps.