On 31 October 2026, Search Engine Journal reported on new BrightEdge research into how travel brands show up in Google AI Overviews and ChatGPT. The study tracked airlines, hotels, cruise lines and booking platforms across booking, comparison and inspiration queries. The big names led almost every category, and the research also found a gap the big names were not filling.
The giants win most answers and lose the specific ones
The headline numbers, as reported by Search Engine Journal:
| Finding | Google AI Overviews | ChatGPT |
|---|---|---|
| Leading airline: American Airlines | 26.6% | 25.3% |
| Leading cruise line: Royal Caribbean | 24.8% | 28.8% |
| Leading hotel brand | Hilton, 28.3% | Marriott, 29.4% |
| Expedia against Kayak | 31.3% against 14.4% |
Source: BrightEdge research, as reported by Search Engine Journal, 31 October 2026.
Two more findings matter. Booking platforms appeared in 78% of budget-related answers but only 17% of the rest. And only 9% of the tracked prompts had booking intent at all. Where the big brands slipped was inspiration queries, the ones that need subjective human experience and judgement. There, YouTube, Reddit and travel blogs with personal expertise could compete.
The claim, and the part I’d test
The easy reading is “big brands own AI search, so the rest of us should hunt for leftover inspiration queries”. I don’t think that reading is wrong. I think it stops one step too early.
Ask why a giant gets named so often. An AI engine can’t see ad budgets. What it can see is that American Airlines, Hilton and Expedia are the easiest brands in travel to describe. Every engine has read thousands of pages that say the same things about them, consistently, with the same facts. When the question is “which airline flies this route”, the answer is already written for the engine. It can state what the brand is without guessing.
Now ask why the same giants lose on inspiration queries. A chain hotel’s pages say almost nothing specific about a stay. They say “luxury”, “comfort” and “unforgettable”. A travel blogger’s page says which room had the sea view, how far the night market was on foot and what breakfast cost. The engine quotes the second page because there is something in it to quote. The article calls that subjective human experience. I’d call it specific facts the giant never wrote down.
That is the pattern I keep seeing. The AI doesn’t pick the biggest brand. It picks the one it understands most clearly and can quote most precisely. Big brands are precise by accident, through sheer volume of coverage. A small site has to be precise on purpose.
Be the brand the engine can describe exactly
Take a family-run hotel, or a clinic, or a specialist retailer. Your pages were probably written to sound as good as the big players. That is the mistake. Sounding like a chain makes you a weaker copy of something the engine already knows. The engine has a confident answer for the chain and a vague one for you, so it names the chain.
The opening is to be the brand the engine can state more exactly than the giant. Write down what you are without marketing language: the exact services, the price range, where you are and how to reach you, who you are for and who you are not for, and what is measurably different about you. Numbers survive. Adjectives get dropped. “Family-friendly” gives the engine nothing it can repeat. For example, “rooms sleep four, cots on request, pool open until 9pm” gives it three facts it can use in an answer.
Then cover the long tail of questions. People ask their AI the same question in a hundred slightly different ways, and comparison and inspiration questions make up most of that spread. The BrightEdge finding that booking intent was only 9% of prompts tells you where the volume sits. The big brands are strong on that 9%. The rest is open to whoever answers the specific question best.
This is how AeroChat got into Google’s AI Overview. We wrote down exactly what it is, who it’s for and how it’s different, with no marketing language, then built comparison pages around the questions that actually produced an AI answer. Within three weeks it was in the same answer box as Tidio, Gorgias and Intercom, the three leaders in its category.
What to do
- Pick ten questions a customer would ask an AI before choosing a business like yours. Include comparison and inspiration questions, not only the buying ones.
- Run them in Google and ChatGPT and record who gets named and what the engine says about them. Date the results, because answers change week to week.
- Rewrite your core pages so that every sentence states something an engine could quote: a number, a name, a place, a limit.
- Publish one page per specific question rather than one broad page for everything.
- Re-run the ten questions a month later and compare.
Source: Search Engine Journal: Big Brands Dominate AI Search But Leave Openings For Smaller Sites