Airbnb Q2 2026 updates show hotels conversion gains and AI search tests boosting ads

Airbnb: hotel-conversion anecdotes and an update on AI search tests’ ad-product relevance

1) Market anecdotes: what’s improving in hotels conversion (and where)

Hotels initiative is progressing faster than initially expected

• Airbnb management characterized the hotel initiative’s progress as “significantly better than I expected,” citing high initial expectations around execution but surprise on the reception from hotels once outreach began. ¹

Supply ramp: starting in regulatory-constrained cities, then expanding broadly

• Airbnb said it began in “markets that are regulatory constrained” (an “obvious opportunity” to bring supply to markets where it’s “not always able to fulfill demand”). ²
• It then expanded supply acquisition efforts to “a top 20 number of cities,” while aiming to add inventory that is “accreted to the platform versus compete with homes,” with selectivity to ensure quality. ²

Conversion engine: hotels are finding demand and also increasing bookings of homes

While Airbnb did not provide specific city-by-city conversion-rate deltas in the excerpt, it did provide behavioral conversion evidence that links hotels exposure to incremental marketplace conversion:

• The basic marketplace theory stated by management is that as more supply of homes or hotels is added, “conversion rate of our traffic goes up” and bookings rise. ³
• Airbnb reported that “about 35% of people who come to Airbnb and book a hotel for the first time” return to book a home. ³
• Management also framed hotels as making the other stronger: “Hotels not only are bringing new guests to Airbnb, but…choosing to book a home…Essentially, one is making the other stronger.” ³

Evidence of relative momentum: hotels growth outpacing homes

• Hotels were described as a “single-digit % of nights booked” on the platform, but management said hotels are “currently growing about three times as quickly as homes.” ²
• A separate excerpt reiterates that hotels are still “a single-digit percentage of nights booked,” while hotel nights are “growing approximately three times faster than our homes business.” ⁴

Where it’s working: both supply-constrained and non-supply-constrained markets

• Airbnb said results are showing strength “in both supply-constrained markets and in non-supply-constrained markets,” suggesting the conversion improvement isn’t limited to regulatory-supply contexts. ¹³

Bottom line on “conversion improvement anecdotes” from the excerpt: Airbnb’s commentary supports (1) faster-than-expected hotel supplier reception (more hotels willing to list due to traffic/take rate), (2) hotel demand traction that creates repeat behavior into homes (35% of first-time hotel guests return for homes), and (3) hotels booking momentum outgrowing homes by ~3x—though it does not provide specific numeric conversion-rate uplift by individual city in the excerpt. ¹³²⁴

---

2) Update on AI search tests and potential role in an advertising product

Status of AI search tests: beginning this month with controlled traffic

• Airbnb said “we are beginning to put it in test this month,” and that the test will involve “a very small % of our traffic.” ⁵
• Management stated that based on results, the rollout would “expand…to more traffic over the course of this year.” ⁵

Customer experience design: why the company expects higher conversion

Airbnb’s AI search concept in the excerpt is explicitly positioned as conversion-enhancing:

• The product is described as providing “AI-generated in real time, personalized” highlights on the PDP (product description page), plus conversational querying (“You can ask the PDP through AI”). ⁶
• Management said the “journey…is going to be powered by AI” and it should “feel as…almost as conversational as a chatbot,” with the expectation of “much higher conversion rates.” ⁶
• Management further explained rollout mechanics to manage user friction: default remains “core search,” with a visible toggle for the AI search experience; the company expects that toggling users “convert very, very well,” while not imposing changes on all users initially. ⁷

Learning curve/risk management: months to retrain user behavior

• Airbnb acknowledged that many users come “often…a handful of times a year” and expect “a search box with a location,” implying “it’s going to take some time, months and months, to retrain the customer.” ⁷

How AI search could become an “advertising product” lever (logical linkage from the excerpt)

The excerpts do not explicitly describe a monetized ads product tied directly to AI search, but they do provide several concrete components that are relevant to how an ads product could work:

1. AI search is positioned as a conversion tool—ads typically monetize incremental conversion.

   • Airbnb repeatedly frames AI as driving conversion (e.g., “much higher conversion rates,” and better search/discovery reducing friction to convert traffic into bookings). ⁶⁸⁷
   • If an advertising product’s economic value depends on improving conversion of promoted listings/services, then the AI search system’s conversational/personalized retrieval could become a controllable placement mechanism. ⁶⁷

2. Personalization is central to the AI search system.

   • Management highlighted “very good personalization” driven by AI to show “just homes, just hotels, or both.” ⁵
   • AI search input is described as natural language (“type in natural language whatever I want”), and results content can be AI-generated (titles in a conversational style). ⁷
   • In an ad context, this suggests future ad relevance and targeting could be more intent-based (what the user wants) rather than only keyword/location-based filters. ⁵⁷

3. Toggle-based rollout implies measurement readiness for any monetization layer later.

   • Airbnb intends to test AI search with a small traffic slice and expand after observing results. ⁵
   • The toggle approach also allows clean comparisons between “core search” and “AI search” for measurable impact (conversion, engagement, downstream bookings), which is a common prerequisite for monetizing a new surface (including ads). ⁷⁵

What Airbnb said about “gating factor” and timeline for it to matter to an ads product

• In the Q&A prompt, an investor asked how AI search tests are going, when they expect AI search to be “a bigger part of the product,” and whether it’s “a gating factor to an ad product.” ⁹
• The company response in the excerpt confirms the testing timeline/expansion plan (start this month; small % traffic; expand over the course of this year) but does not explicitly answer the ads “gating factor” question with monetization-specific details in the provided text. ⁵⁹

Bottom line on AI search for ad-product potential (from the excerpt): Airbnb is running controlled AI search tests starting this month on a small traffic fraction, planning wider rollout through this year, and expects it to improve conversion via conversational/personalized search + PDP AI. ⁵⁶⁷ The excerpts support that AI search is likely to be commercially important because it is designed to improve match quality and conversion—attributes that typically underpin the economics of advertising—but the excerpt does not provide explicit statements about ad product mechanics, pricing, or direct monetization strategy for AI search. ⁶⁷⁵

---

3) Analyst-style synthesis (linking hotels and AI search to marketplace economics)

• Hotels are already demonstrating “category expansion” conversion behavior: first-time hotel guests have a high likelihood of returning to book homes (~35%). ³ Hotels are also growing faster than homes (~3x hotel night growth). ²⁴
• AI is being positioned as an execution accelerator for conversion across search and the guest journey (“reducing friction…converting more traffic into bookings”). ⁸ AI search is expected to produce “much higher conversion rates,” starting with a toggle to protect the default experience while learning customer behavior. ⁶⁷
• The ad-product relevance is plausible but not proven in the excerpt: AI search personalization and intent-driven conversational interfaces are the right ingredients for better targeting and potentially higher sponsored-result effectiveness, but the provided text does not confirm how ads will be integrated or monetize AI search specifically. ⁵⁷⁶

All key figures and timelines above are taken directly from the provided excerpts. ^21345678