ABNB 3Q25 Business Update: Growth Reaccelerates, but New Chatbot Risk

(Download a PDF of this post here).

3Q25 Update.

Airbnb reported 3Q25 results and the stock was down -2% the next day, however it is now flat since reporting. Shares are -6% YTD and -12% since IPOing at the end of 2020.

Revenues grew +10% y/y, which marked a deceleration from last quarter. However, GBV continued reaccelerated in 3Q to +14% y/y, driven by growth in both nights booked and average daily rates. Adjusting for FX, GBV was +12% y/y, which is a 300bps improvement from last quarter. While revenue growth lagged GBV, it is important to note that GBV tends to be a leading indicator as revenue is recognized after the trip is commenced, whereas the bookings are recognized when a stay is booked.

Brian Chesky noted on the call that they are focused on 4 key areas: 1) making the service better, 2) bringing Airbnb to more parts of the world, 3) expanding the Airbnb offering, and 4) integrating AI across Airbnb. 

Nights Booked also accelerated, rising +9% y/y, up 200bps sequentially. On the call, they noted that the acceleration was largely due to strength in the U.S. Notably, the new pricing feature of Reserve Now Pay Later (more on this later) helped drive booking growth in the U.S. for 3Q. To put that in perspective, peers Expedia and Booking Holdings also saw nights booked grow in 3Q. Expedia saw a +11% y/y, while Booking saw a +8% y/y respectively.

EBIT margins fell slightly to 39.7%, a 120bps decrease y/y, driven by an increase in G&A and a slight increase in R&D.

LTM free cash flow grew to $2.3bn, which they have continued to deploy into stock buybacks. In 3Q alone they repurchased $857mn shares. Since they introduced their share repurchase plan, they have bought back ~8% of shares outstanding. They still have $6.6bn remaining in their latest share buyback program.

Business Commentary.

Airbnb’s “Expansion Markets” continue to grow at double the rate of their core markets, a trend that has now continued for 7 consecutive quarters. Overall, nights booked remained steady in 3Q. Breaking it down by geography, LATAM grew in the low-20s, driven by Brazil which grew nights booked by over 20% and first-time bookers in Brazil by +17% y/y. Their recently launched interest-free installment payment plan option in Brazil saw usage more than double in the period, which helped drive bookings in 3Q.

In Asia Pacific, they saw mid-teens growth in nights booked. Japan growing strong for them as nights booked for domestic travel increased +27% y/y, marking the third consecutive quarter of acceleration. In addition, they noted that they are seeing progress in first-time bookers in Japan and India. Japan saw first-time bookers grow +20% y/y, a 5 point increase from last quarter, and India saw +50% y/y in first-time bookers. EMEA grew mid-single digits with a tougher comp due to the Paris Summer Olympics.

Lastly, North America grew mid-single digits, representing a sequential acceleration. The growth acceleration in North America was due to strong domestic travel, which was helped by their new payment offering of Reserve Now Pay Later. This allows the traveler to book a stay without being charged for it immediately. While this does increase cancellations a bit, Airbnb notes that it is net additive with more total completed bookings as it reduces friction for the initial booking because guests feel lower commitment.

Despite, launching this feature in the middle of the quarter, they are already seeing an impact on booking growth. They mentioned that they saw meaningful acceleration from 2Q to 3Q in nights booked in the U.S. because of this pricing offering. CFO Ellie Mertz on the call, stated that roughly 70% of the guests utilized the feature when offered.

Seeing North America reaccelerate was a positive as it has been a struggle for them in recent quarters. In our last update, we mentioned that there were three main levers in which Airbnb could reaccelerate growth 1) improving affordability via total price transparency and pricing tools, 2) expanding payment options to attract a wider demographic, and 3) targeting underpenetrated segments such as the U.S. Hispanic population and Heartland states. This quarter’s results suggest early progress on the first two, with affordability and payment flexibility driving higher bookings.

Another key development was Airbnb’s renewed focus on hotels. If you remember, they did an acquisition of HotelTonight back in 2019, which focused on boutique hotels, however it hasn’t been a notable part of the strategy, but that is beginning to change. They started hotel pilots in LA, NYC, and Madrid by partnering with boutique and independent hotels. The idea for Airbnb’s push into hotels is to increase the available options for guests in supply-constrained markets where homes are not the best use case (or short-term Airbnb’s are outlawed). As we noted in our last update, the shift to hotels was framed as a sort of back up option when adequate Airbnb supply isn’t available. However, on this call Brian Chesky’s tone seemed to have shifted. While they will still focus on independent hotels, they make it seem like hotels are more of a complement to Airbnb, rather than a last resort substitute. It could potentially dilute their brand, but it’s also an easier cross-sell. Booking’s success here suggest that offering the two categories to customers is a net benefit.

They believe that homes and hotels can live together and that they won’t see a large cannibalization between the two. Adding hotels is also their attempt to become a “one-stop shop for travel” and they hope that it could have knock off effects for services, experiences, and whatever else they dream up.

AI remains a big focus. They noted that they have a dozen AI work streams underway and rolled out an AI customer support assistant in chat, which helps users perform common actions such as changing reservation dates or canceling with greater ease. It was initially launched in the U.S but they will now expand to other countries and more than 50 languages next year. They found that it reduced people’s need to contact a human agent by 15%. In addition, they are testing AI powered search. Similar to ChatGPT or Gemini, users in the Airbnb AI search will have personalized responses via understanding the users prior history and preferences. The goal is to combine AI search and AI customer service into an AI travel concierge which will help you book your travel end to end. Brian Chesky notes in the quote below that in the end, specialization will win, which will lead them in becoming a “one-stop shop for travel.”

At the start of October, OpenAI launched travel apps in partnership with competitors Booking Holdings and Expedia, with Airbnb absent. Essentially what this allows users to book flights and find rooms within ChatGPT. However, Airbnb took another approach. Brian Chesky clarified that while they are not opposed to integrating with chatbots like ChatGPT and will do so in the future, Airbnb’s integration at the time wasn’t ready due to four integration principles they want to adhere to.

This alludes to a growing contention with AI Chat apps like ChatGPT who are disintermediating marketplaces from controlling the customer experience and losing their direct customer touchpoint. Airbnb does not want to just be a database that ChatGPT surfaces results alongside all of the other options. But if Google’s success in air travel and hotels showed us anything, it is that users like to see as many options as possible. However, Airbnb has been able to maintain a direct traffic to their site and apps despite Google’s grip on being the first place many potential travelers start their search. The hope is that Airbnb can similarly draw direct traffic to their app versus ChatGPT or another AI chatbot. While the status quo would be that Airbnb users continue to think Airbnb when they want an alternative accommodation, this is a key risk in the future as users internet habits change with AI.

Turning to their other bets, Airbnb Services & Experiences continued it’s momentum in the quarter, albeit off of a very small base. Brian Chesky noted that in 3Q almost half of the people who booked an experience did not have an Airbnb stay. In the period they received double the amount of applications (110,000) from potential hosts to be a part of services and experiences. To increase usage in the app and time spent on the app, they added a feature where other guests can interact with one another before and after an experience. Despite the initial success, they noted that they believe it will take 3-5 years for Services and Experiences to be a notable part of the business.

Despite this success, in an AlphaSense expert call, a former head of partnerships at Airbnb gave more context on some underlying issues Experiences has.

The expert believes that Airbnb has the challenge of earning user’s trust in terms of these experiences, which in turn lead to lower conversions. Also, the space is highly fragmented and people already go to other sites like Yelp to find these services and experiences. In our opinion, trust isn’t the real problem. It is more habit, lack of top of mind awareness, and low offering density. If people are willing to book a room in a strangers apartment, then Airbnb is clearly trustworthy.

They have a classic two-sided marketplace problem where they don’t have enough offerings to draw enough users and the low usage of services doesn’t incentive enough service providers to join. They are solving a clear user problem, which is that there is no great way to seamlessly book and compare local services. Yelp is a mess of an app, loaded with sponsored ads, an antiquated search function, and inconsistent in-app functionality for each business (I.e. most send you to their poorly built website to book or call instead of embedding messaging and scheduling directly into the app).

Google Reviews is not much better as you still have to go to each businesses website or call for most functionality. If you want to order food, you can do it all in app on DoorDash or Uber Eats, but if you want to book a hairdresser appointment you usually have to search on Google or Yelp and then click into the hairdresser’s website and call them. If the hairdresser does happen to have an online scheduling capability, it isn’t linked to a preexisting account of your requiring you to fill out basic information and payment. This all adds friction to process.

A SuperApp could eliminate all of this. However, Airbnb still lacks the frequency to push users to regularly try Services, which is limiting their ability to grow the service provider part of the marketplace. The expert below thinks this could potentially be solved with a large ad budget to raise awareness.

We previously talked about how Yelp could help them solve a lot of these problems. Thinking about why that is, can help an investor understand Airbnb’s current service’s weakness. This is because Yelp has the opposite problem Airbnb’s Services push has: Yelp has ample traffic, but no idea how to monetize it well. It is still very commonly used to search for local services—exactly the focus of Airbnb’s Services. Not only would they benefit from Yelp’s 180mn monthly users, which they could migrate into a combined app, but the repository of reviews and millions of businesses on Yelp would give their Services push instantly more credibility. Yelp would become the main reason that users open the app and then they could overtime push more service providers to accept a “fully integrated offering” which allows for instant booking, on-platform payment, and messaging. In return, they could boost these service providers in the search results, the same way Amazon boosts Prime products in their results. Airbnb could no doubt improve the Yelp experience too with their design aesthetic.

As the expert alluded to (and in our opinion), in order to build adoption, Airbnb should probably abandon their 20% take-rate on services and wait until a consumer habit is built before attempting to monetize it. Their focus needs to be on becoming a strong demand generator for service providers first and doing everything they can to build up the supply-side of service providers.

Without a stronger base of users who have a reason to open the app regularly, it will be tough to have enough traffic to cross-sell to service providers. While in our opinion Yelp could solve this, and cost less than one year of cash flows at $2bn, which could make it worth the experiment (and would probably be cheaper than acquiring the usage via advertising).

Airbnb’s Services serves a real need by helping a variety of providers find demand, which is why we like it, but it is just a hard thing to build without more user frequency. Meituan, which is similar to what Airbnb wants to build, was able to drive most of their frequency through food delivery. Ironically, Meituan actually also is a leading hotel booking provider, but the sequence matters. They used their traffic from users who ordered food delivery 1-2x a week to cross-sell hotel rooms—not the other way around.

Taking a step back though, Airbnb’s core business had a good quarter. Core markets reaccelerated from new growth initiatives like Reserve Now Pay Later and International continued to see success with more and more users booking on Airbnb in Asia. Their growth in GBV helps build confidence that they can grow revenues in the low teens longer term after many quarters of being unhappy with their results.

In terms of valuation, Airbnb currently has an enterprise value of $65bn after backing out almost $12bn in net cash. If you back out stock based comp and interest income from free cash flow, they still have generated about $2.3bn in FCF LTM. This gives them a 28x FCF multiple. If you think of mature margins of 40% though, which is very plausible for a marketplace business, they trade at just 16x EV/ NOPAT.

Below are some notables from the call.

Call Notes

  • GBV increased +14% y/y, nights booked rose +9% y/y (strength in the U.S. and ADR)
  • 4 key areas of focus: making the service area better, bringing ABNB to more parts of the world, expanding the offering, and integrating AI across the app

Feature Releases

  • Introduced Reserve Now Pay Later In the U.S., which helped drive licenses booked in 3Q and they are going to roll it out to more areas next year. (Saw meaningful acceleration in the US from 2Q to 3Q, due to the launch of Reserve now Pay later payment offering).
    • “So about 70% of people that we offer Reserve Now, Pay later take us up on that offering. Now to your question of cancellations, we obviously tested this product pretty extensively before launching it in the U.S. to ensure that the benefit of the incremental lift in bookings from the payment offering was not more than offset by increased cancellations. So yes, there are increased cancellations, but we’re highly confident that the net impact of the product is a lift to the net bookings.”
  • Also, announced that they improved the context of the map, which they believe will increase the user experience, which will then keep people on the the app and increase conversion. (They noticed that many users would leave the ABNB app to search on Google or Apple maps due to a lack of context, which then leads to a lower probability that that user would come back to the app to book).
  • Updated their cancellation policy: guests can cancel free up to 14 days, and all guests can cancel free within 24 hours if they booked more than 7 days before check-in

International

  • Average nights booked in expansion markets have grown at double the rate of core markets. First-time bookers were up 20% y/y in Japan and +50% y/y in India.
    • Nights booked remained steady or accelerated sequentially
    • LATAM: grew low 20s, APAC: mid-teens, both NA and EMEA: up mid-single digits
    • Saw meaningful acceleration in the US from 2Q to 3Q, due to the launch of Reserve now Pay later payment offering.

Services and Experiences

  • Service experiences are bringing more people to the platform, Brian Chesky noted that in 3Q almost half of the people who booked an experience did not have an Airbnb stay. The supply of service experiences is also scaling along with demand. Chesky noted that they received double the amount of applications from potential hosts (110,000). Introduced a feature where you can see other guests who are attending your service experience and you can contact them and keep in touch with them in the app, in the hope to increase usage in the app.
  • “So we’re focused on Paris, focused on L.A. We’re seeing a lot of traction in these markets. And as they grow, we’re going to be rolling out to dozens of cities around the world, this expanded playbook and then hundreds more cities. So it’s going to take 3 to 5 years, I think, for service experiences to become a material part of our business, but I’m very, very bullish on them.”

Hotels

  • Launching Airbnb Hotels. Started a hotel pilot in LA, NYC, and Madrid. Partnering with boutique and independent hotels. They built a custom built product for hotels with new search filters and room types specifically designed for hotels. They are pushing to recruit hotels in key markets where the supply of Airbnbs are little to none and where hotels make sense for guests (ie. night stay in a city)
  • “I imagine Airbnb in the future is going to be the best one-stop shop for travel. And for us to be a one-stop shop for travel, I think we’re going to need not just homes, not just services, not just experiences. We’re going to need a portfolio of offerings on Airbnb. And so hotels is certainly one of those things.”

AI

  • They have a dozen AI work streams underway. Rolled out AI customer support assistant in chat, which helps users perform common actions such as changing reservation dates or canceling. Initally launched in the US but will now expand to other countries and more than 50 languages next year. They found that it reduced people’s need to contact a human agent by 15%.
  • “Our thesis of AI is that specialization will win in travel. That’s our theory, that specialization will win. We have a lot of unique capabilities. We understand travel, we have one of the best design teams in the world, so we can design custom interfaces. We do not think the way AI search will work in the world of travel is just text. It’s going to have to have rich user interface experiences. We’re adding a lot more verticals. So we do think Airbnb could be a one-stop shop for travel.”

Share Buybacks

  • Repurchased $857mn of stock, which brings LTM to >$3.5bn ($6.6bn remaining in authorization)

Personalization

  • So I think it’s really in the future, going to be much more about personalization. Right now, it’s a little bit more backfill filling and supply gap and prioritizing hotels when it’s the last minute, say, night stay. But I think over time, it’s really about personalized experience for every guest and understanding what kind of guests are you? What is your intent? And as we do that, I think we can expand our hotel offering a bit more.

More entrepreneurial approach to new business launches

  • I think we’ve decided to take a more entrepreneurial approach, and we’ve learned this from certain experiences where we can test and incubate businesses, and we can test them in city. And so I imagine going forward, we are going to be simultaneously testing half a dozen, maybe even a dozen new businesses at a time in one market. Now it doesn’t mean if we test in a market, we ever ship it, but it does mean that we can kind of develop a lot of pilots over — in parallel. And then we can decide when something is working, that we can scale it up. And that’s what we’re going to — that’s how we’re going to really approach a lot of these new businesses, kind of really starting with a pilot market. Hotels as an example. We’re starting with 3 pilot markets. We know hotels we’re going to expand. There’s no question. There are other businesses where we’re testing them out, and we’ll see where they go. Just to give an example of not a business we’re going to launch, but a segment that we’re seeing a lot of promise in is [lux]. One of our fastest-growing segments is high ADR. And we think that’s a big opportunity for us. So I’m not redoing that we’re going to launch something there, but I’m just showing there’s a lot of segments, a lot of supply types that we think could be really, really promising for us. And we’re going to be really going city by city. And then when we figure out the playbook, we’ll be expanding more aggressively.
  • The more competitive our prices, the more bookings we get, the more pricing tools we build, the lower the prices become. I think if you noticed over the last 2 years, hotel prices have appreciated faster than Airbnb. I think that’s a good testament to the pricing tools we’re making are working.

Chesky on Growth

  • I do not think our business is even close to mature. And it’s so funny. We started raising money and I think our first fundraising round was in 2009. And the question was how big is this market? And then the question came in 2011, how big is this market, 2012, 2014, all the way up until the IPO launch. And every step along the way, we’ve always kind of been surprised at how big this market is. And while I don’t know how big the Airbnb market is, again, if you just think about the size of our market, for every person who stays in a home, about 9 people stay in a hotel. I could imagine that the market could at least double the core market for homes. And the reason why is, number one, people love homes. That’s why they live in them.
  • A large percent of people increasingly are traveling with groups or families. Hotels have limitations for people staying as groups. One of the biggest challenges historically with Airbnb has been quality control, not having a front desk and just making sure we’re price competitive. As we’ve built a great supply machine and add better pricing tools, the prices are increasingly getting more effective. We now show total price upfront. There’s no cleaning fees that a guest ever sees. We think this is helping us become more competitive. We’re making the product significantly higher quality.
  • We’re also building a machine to recruit supply where we have homes. So there is a huge amount of opportunity for us to grow our core business. I don’t think it’s mature in any market. And I think we are looking at the possibility of reaccelerating our core business in every country in the world over the coming years. because I think there’s a new generation of travelers that aren’t supposed to be in hotels, especially young travelers, Gen Z and eventually Gen Alpha. And I think for many of them, Airbnb is going to be the go-to way to travel. And as social media is taking over for travel discovery,

*At the time of this writing, one or more contributors to this report has a position in Airbnb. Furthermore, accounts one or more contributors advise on may also have a position in Airbnb. This may change without notice.