As the long National Day holiday wraps up, how many nights did you spend in a hotel this holiday?
On October 9, the Ministry of Culture and Tourism released 2026 National Day holiday data. Over the seven-day period, domestic trips nationwide reached 826 million, with total domestic tourism spending hitting 738.375 billion yuan, up 6.3% and 4.3% year-on-year on a comparable daily average basis, respectively.
Survey data from the China Hospitality Association on the National Day holiday shows that 72.7% of accommodation businesses saw occupancy rates rise or stay flat year-on-year, with an average increase of 8.2%; at the same time, average room rates generally declined, with an average drop of 8.7%. People are still willing to travel, but they have become more cautious with every spending decision. Should they stay an extra night for a better location and service? Should they choose a room upgrade? More and more consumers are weighing these questions repeatedly before booking a hotel.
Airbnb Chief Business Officer Dave Stephenson once did a simple calculation: out of every 10 people who travel, 9 choose to stay in a hotel, and 1 stays in an Airbnb. "If we could convert just one of them to stay in an Airbnb, our business would double." This "doubling vision" sounds easy enough to achieve, but for a mature market with already abundant supply, the biggest challenge is where this one incremental consumer can actually come from.
On October 8, Feizhu released its 2026 National Day holiday data: hotel fulfilled room nights surpassed historical records for four consecutive days, average daily fulfilled room nights during the holiday grew nearly 70% year-on-year, and hotel room nights in more than 100 cities nationwide doubled.
Over the past year, similar high growth has not been uncommon. From last year's National Day to this year's Spring Festival, then the summer vacation, and now the Mid-Autumn Festival and National Day, Feizhu's hotel business has consistently maintained high-quality growth during peak travel seasons. In a market environment where consumers are more cautious, how has Feizhu managed to achieve nearly 70% hotel room night growth? For the hotel industry, how can they make the guests who have already arrived stay? Now that AI has arrived, can it bring new demand and efficiency space to this traditional industry?
Seeking high-value customer groups. Hotel operations need "longer relationships."
The "2026 China Hotel Industry Development Report" released by the China Hospitality Association shows that as of the end of 2025, mainland China's hotel industry had 374,700 facilities and approximately 18.736 million rooms in total. Industry insiders say the hotel market overall is still a buyer's market. They believe that price remains important among consumer factors, but it is no longer the only answer. The industry is shifting toward finding growth through segmented customer groups, product innovation, and refined operations.
Against this backdrop, high-value customer groups have become one of the key pillars. In Feizhu's report card for this holiday, 88VIP members played a supporting role. Data shows that Feizhu's overall fulfilled room nights for high-star hotels grew more than 80% year-on-year. Meanwhile, the per capita hotel spending of 88VIP users on Feizhu was 34% higher than the platform average. For hotels, the value of this type of customer group lies not only in their willingness to pay higher room rates, but also in their greater potential for accommodation and derivative spending. For hotels operating with heavy assets, finding customer groups that better match their own products and price bands is becoming increasingly important.
Capturing high-value customer groups is only the first step. The question hotels need to solve is how to turn these customers into their own assets. Hotels and OTAs used to "love and fight" each other: OTA platforms provided traffic for hotels and converted a large number of orders, but if a consumer searches and compares prices again for every stay, then the transactions hotels get are cut into very fine fragments, which is not conducive to hotels building long-term customer relationships. Facing the hotel industry's anxiety over losing private-domain customers, Feizhu's "second official website" is trying to change this. Compared with the traditional distribution model, hotels can not only obtain orders through the platform, but also further build membership relationships and repeat purchase opportunities. In this way, what the platform brings is not just a one-time traffic conversion; hotels also have the opportunity to reduce their reliance on repeatedly buying traffic and price promotions.
Data shows that more than 70 branded hotel official flagship stores doubled their room nights, and these brands' newly added members through Feizhu grew more than 30% year-on-year. "If the platform only matches a transaction, the relationship between the hotel and the platform is like a 'one-time' one. But if the hotel can also retain these guests and convert them into its own members, the two sides will move from a one-time business to long-term cooperation," an industry insider said.
As high-value customer groups on the platform become more certain, more hotels are willing to join, which in turn gives consumers more choices. Hotel merchants newly joining Feizhu this year saw their average daily fulfilled room nights during the Mid-Autumn Festival and National Day holiday increase by another 135% compared with the peak period of the May Day holiday. It can be seen that this cycle on the Feizhu platform has already begun to operate. After Feizhu's capabilities on both the consumer side and the supply side were strengthened in both directions, it is now accelerating in both directions.
AI may transform the industry model. Independent hotels have reached a new starting line.
This National Day, the Ministry of Culture and Tourism's 2026 National Day holiday summary showed that projects such as AI-assisted travel are enriching tourist experiences, and big data monitoring and AI scheduling are increasingly entering the cultural tourism industry. Ms. Zhang's family used AI to help book tickets during this National Day trip. In previous years, she had to take care of her parents walking less, her children wanting an amusement park, and she and her husband needing very good hotel signal because they might receive video work meetings at any time. "In the past, I would scroll through notes on social platforms, then switch to maps to see how to get a taxi, and finally filter room types on OTA platforms, switching back and forth in review apps," Ms. Zhang said. After the whole process, she was already too tired to be in the mood to enjoy the trip.
Travel consumption is naturally a complex decision, involving multiple constraints such as time, place, number of people, preferences, and budget. The traditional model requires consumers to do the permutations and combinations themselves across different platforms. This creates an industry paradox: OTAs' product pools are constantly expanding, but users' decision-making efficiency has not improved in sync, and is even declining.
AI is changing this transaction chain. In August this year, Feizhu launched a new-generation travel AI product called "Feizhu Bangbang," whose core is to solve in one conversation what used to be fragmented, multi-entry search behavior. This looks like merely a change in interaction from a "search box" to a "dialog box," but the logic behind it has changed. Users no longer enter "Shenzhen hotel" and then flip through lists to compare prices themselves; instead, they say, "Take elderly people and children to Shenzhen for three days, the elderly can't walk too much, the children want to go to an amusement park, budget around five thousand." What AI needs to understand is not keywords, but a complete travel scenario. Needs that used to be scattered across different apps may be fully handled through one entry point.
During the just-concluded National Day holiday, the interaction volume of "Feizhu Bangbang" increased 158% month-on-month, flight and hotel bookings increased 110% month-on-month, and fulfilled items including check-in, itinerary changes, and refunds and changes increased nearly 100% month-on-month. Behind this set of data is another meaning worth noting: what Feizhu Bangbang covers is no longer just "helping users pick a hotel," but the complete chain from pre-trip planning to in-trip fulfillment. The more complete the chain, the deeper AI's understanding of user needs, and the more accurate the recommendations.
It is not hard to see that as it gradually matures, AI may bring "operational equality" to hotels. Under traditional marketing logic, traffic and rankings largely determine a hotel's business momentum, but in the AI era, AI can "actively" recommend the most matching hotels to consumers based on consumer needs. For example, if an independent hotel clearly displays information in dimensions such as "quiet room configuration," "parent-child care details," or "pet-friendly facilities," when user demand comes in, AI can accurately capture it and push it to matching users. After AI intervenes, what determines who gets recommended is no longer ranking weight, but the degree of match between the product itself and user needs. The ceiling that was previously locked by scale and budget may be opened.
This will also guide hotels to shift their operational focus from traffic operations back to the product itself: not studying how to boost rankings or bid for keywords, but thinking about what makes their hotel good, who it suits, and what needs it can solve. When the entire industry's attention returns from "grabbing traffic" to "making products," those who benefit are not only independent hotels, but the entire accommodation ecosystem.
The "doubling" mentioned by Airbnb is very tempting, but returning to the essence of business, in a mature market, every step forward requires rethinking and using clearer solutions to leverage growth. The new perspective Feizhu provides this time is to solve consumers' choice difficulties and hotels' operational difficulties together: Feizhu brings target groups such as 88VIP to hotel merchants, and AI then helps consumers find the right hotel faster, while also helping hotels find the right guests more accurately. When matching efficiency improves, transactions will naturally happen where they should. For hotels that have truly put effort into products and services, this brand-new competitive environment may be the biggest opportunity.