ChatGPT Ads for Travel and Hospitality: A Planning-Stage Playbook
Travelers now plan whole trips inside ChatGPT. This playbook shows hotels, tour operators and travel brands how to appear in those planning conversations without overpromising.
Trip planning has always been a research marathon: ten tabs of hotel reviews, three blog posts about the best neighborhood, a spreadsheet of flight times. ChatGPT compresses that into a conversation. Someone types "plan a four-day trip to Lisbon with a toddler in October" and gets an itinerary, neighborhood suggestions and packing advice in one answer.
That planning conversation is where travel ads now live. Since February 2026, OpenAI has shown ads to logged-in adult users on ChatGPT's Free and Go tiers in the US, placed below the answer and clearly labelled as sponsored. Self-serve Ads Manager opened in May 2026, and the program has expanded to more countries through the year. For hotels, tour operators, vacation rentals, travel gear brands and experience companies, this guide covers how to show up usefully at the planning stage.
Where travel fits in the ChatGPT conversation
Travel intent unfolds over days or weeks. A traveler moves through stages, and each one produces different questions:
| Stage | Example question | Who can answer with an ad |
|---|---|---|
| Inspiration | Where should we go for a warm week in February? | Destination resorts, tour operators |
| Planning | What's the best area to stay in Kyoto for a first visit? | Hotels, vacation rentals |
| Logistics | How do I get from Rome airport to Trastevere? | Transfer services, rail passes |
| Preparation | What should I pack for Iceland in March? | Travel gear and apparel brands |
| In-trip | What can we do in Barcelona on a rainy afternoon? | Experiences, museums, food tours |
The mistake most travel advertisers make is running one generic "book now" ad across every stage. A person asking where to go for a warm week is not ready to book a specific hotel. A person asking how to get from the airport to their neighborhood is ready to buy a transfer today. Match the ad to the stage.
Who the audience is
Ads appear only to logged-in adults on Free and Go plans. Paid tiers such as Plus, Pro, Business and Enterprise are ad-free. Go is a low-cost tier available in 170+ countries, though ad availability depends on where OpenAI has launched ads, which as of late 2026 is reported to be 40+ countries.
For travel, that audience skews toward leisure travelers planning their own trips, which suits independent hotels, guesthouses, tour companies, travel gear and experiences. Corporate travel management tools that sell to companies are a weaker fit; business decision-makers are more likely to use paid, ad-free plans. Check our guide on who sees ChatGPT ads for the full breakdown.
Context hints for travel
Context hints are plain-language descriptions of the conversations your ad belongs in. A keyword list like "Lisbon hotels" misses the point. A good hint describes the traveler, the situation and what they need.
Examples for a hypothetical boutique hotel in Lisbon:
couple planning a first trip to Lisbon looking for a walkable neighborhood with good restaurantsfamily with young children planning a Lisbon city break and wondering where to stay that is quiet but centraltraveler comparing Lisbon neighborhoods like Alfama, Chiado and Príncipe Real for a long weekend
For a travel backpack brand:
person preparing for a two-week trip with only a carry-on bag and wondering how to pack lighttraveler worried about airline carry-on size limits on budget airlines in Europe
For a food tour company in Mexico City:
first-time visitor to Mexico City wanting to try street food safely with a local guidecouple planning a Mexico City trip who want an evening activity focused on local food
Each hint narrows to a moment where your product is a natural next step. Write three to six per ad group. The context hints generator can give you a starting list to edit, and our guide to context hints explains why situation beats keyword.
Chat card examples for travel brands
The chat card gives you a 3 to 50 character headline, about 100 characters of body, an image and a landing URL. In travel, the image does a lot of work, so use a real photo of your property or experience rather than generic stock.
Boutique hotel
- Headline:
Stay in Chiado, Steps From Lisbon's Cafés - Body:
18 rooms in a restored 1890s building. Rooftop terrace, breakfast included, free cancellation.
Food tour
- Headline:
Mexico City Street Food With a Local Guide - Body:
3-hour evening walk, 7 tastings, small groups of 10 or fewer. Vegetarian options available.
Carry-on backpack
- Headline:
A 40L Carry-On Built for One-Bag Travel - Body:
Opens like a suitcase, fits most airline size limits, laptop sleeve and lockable zips.
Airport transfer
- Headline:
Fixed-Price Transfers From Rome Fiumicino - Body:
Private car to your hotel, driver tracks your flight, price set when you book.
All four sit under 50 characters for headlines and 100 for bodies. Each one carries a concrete, checkable detail rather than adjectives like "luxurious" or "unforgettable." Notice the backpack says "most airline size limits," not "all," because airline rules vary and an overreaching claim invites complaints. You can check copy against the limits with the ad character counter.
Pricing accuracy and policy pitfalls
Travel is not on OpenAI's restricted list the way health or finance is, but several general rules hit travel harder than other categories.
Prices must match the landing page
If the ad says "from $129 a night," the landing page should show $129 for real dates, not only for a single off-season Tuesday. Bait pricing is the fastest way to get disapproved and to waste clicks. If prices vary widely, leave the price out of the ad and lead with location or amenities instead.
No fake urgency
"Only 2 rooms left!" and countdown-style pressure are a common habit from online travel agencies. OpenAI's policies prohibit fake urgency and clickbait. If scarcity is real and verifiable on the page, describe it plainly; otherwise drop it.
Reviews and ratings
Quoting "4.8 stars" is fine only if the rating is real, current and visible on the landing page with its source. Do not invent review counts or round up.
Trademarks and other brands
Do not name booking platforms, airlines or competitor hotel chains in your ad copy. "Book direct and skip the booking-site fees" is fine; naming the site is not.
Insurance and financing
Travel insurance and "book now, pay later" financing touch financial products, which are a restricted category. Treat those ads with the extra care described in our guide to ChatGPT ads for consumer finance products.
Landing pages for travel traffic
A traveler who clicked from a planning conversation still has questions. Answer them on the page:
- Show where you are on a map and how far it is from the landmarks they care about.
- Put live availability and prices close to the top, with real dates prefilled where possible.
- State cancellation terms clearly, since flexibility is a major factor for trips booked weeks ahead.
- Include photos of rooms or experiences as they actually are.
- For tours, list duration, group size, meeting point and what is included.
The page should repeat the ad's promise in its headline. If the ad said "steps from Lisbon's cafés," the page should open with the neighborhood, not a generic welcome message.
Budgets and booking-window math
OpenAI's minimum is about $25 per day per campaign in the US. SecondWin recommends $50 to $200 per day to start. OpenAI's recommended starting bids work out to roughly $3 to $5 per click.
Travel has two quirks. First, order values can be high, which gives you room. Second, the time between click and booking can be long, so judging ads after three days can mislead you.
Here is a hypothetical example. Say a boutique hotel's average booking is three nights at $180, or $540, and the hotel keeps about $380 after fees, breakfast and cleaning costs. At $4 per click and a 1.5 percent booking rate, each booking costs about $267 in ad spend ($4 divided by 0.015). That leaves roughly $113 of contribution per booking, before any repeat stays.
A food tour at $85 per person with two people per booking ($170) and $100 of margin has less room. At the same $4 CPC, it needs a booking rate above 4 percent to break even ($4 divided by 0.04 is $100). Run your own figures through the ROAS calculator and set expectations by product before you scale.
Track bookings properly with the OpenAI pixel or the server-side Conversions API so delayed bookings still get credited. Booking engines often run on a separate domain, so confirm the purchase event fires on the confirmation page. Our conversion tracking guide covers setup.
Seasonality
Travel demand moves with seasons, school holidays and events. Plan budgets around the planning window, not the travel date. People research summer trips in spring and ski trips in early fall. Raise budgets as planning conversations pick up and pull back once the booking window has largely passed. Refresh context hints by season too: "warm winter getaway" hints in November, "family summer road trip" hints in March.
How SecondWin helps travel brands
SecondWin starts from what already works. Enter your URL and it reads your site to identify your property or product, your traveler and your niche. It then looks at the longest-running ads in that niche in Meta's Ad Library, since ads that stay live for 60+ days are usually the ones that keep booking, and pulls out the messages behind them: the promise, the proof and the offer.
It writes original ChatGPT ads around the questions travelers ask at each stage, with context hints and landing links, and checks every one for policy issues such as unmatched prices or fake urgency. After you pick a plan and add SecondWin as an Admin on your OpenAI ad account, it launches and manages campaigns while OpenAI bills ad spend directly to you. Get a free analysis of your travel site to see the traveler questions it finds, or compare SecondWin's plans. SecondWin is independent and not affiliated with OpenAI.
FAQ
Are ChatGPT ads good for hotels?
They can be, especially for independent and boutique hotels that want to reach travelers while they are choosing a neighborhood or comparing places to stay. Ads appear below ChatGPT's answer and do not change the answer itself. Hotels with clear location benefits, real photos and transparent pricing on a fast booking page tend to have the best chance of turning planning-stage clicks into bookings.
Can I show prices in a ChatGPT travel ad?
Yes, as long as the price is accurate and visible on the landing page for realistic dates. "From" pricing that only applies to a rare date is likely to frustrate users and can create policy problems. If your rates swing a lot by season, lead with location, amenities or flexibility instead, and let the booking page show live prices.
How long should I run a travel ad test?
Longer than most ecommerce tests. Travel bookings often happen days or weeks after the first click, so give a campaign at least two to four weeks and make sure conversions are tracked through your booking engine with the OpenAI pixel or Conversions API. Judging ads after a few days usually undercounts what they actually produced.
Can tour operators target specific destinations?
Yes, through context hints that describe travelers planning trips to that destination, such as "first-time visitor to Rome planning a day near the Vatican." Context hints describe situations rather than keyword lists, so write them around the traveler's plan and need. Ad availability also depends on the user's country, so confirm OpenAI serves ads where your target travelers live.