You sell a 12-day Vietnam trip at around 3,800 EUR per person. Your blog post about street food in Hanoi picks up traffic every month. Your Vietnam trip page picks up almost nothing, and the terms you care about are held by a marketplace, a guidebook publisher and a competitor who runs the trip worse than you do.
That gap is not proof that SEO fails for tour operators. It is what happens when a site is built as a brochure with a booking button attached, while the person spending 4,000 EUR searches their way through a decision that runs for weeks: whether to go at all, then where, then when, then how long, then small group or private, and only at the very end who to book it with. Almost none of those searches contain your product name. If trip pages are the only substantial pages you own, you are visible for the last step of that sequence and paying someone else for every step before it.
Below: mapping that journey onto pages, why thin trip pages lose, internal linking, the technical faults specific to tour sites, and what AI answers really change. It ends with an order to do it in.
Map the journey, then give each stage its own page
The traveller who books your Kenya safari did not start at "kenya safari 10 days". They started nearer "best time to see the great migration", spent a few weeks on "kenya or tanzania for a first safari", and only then typed something that looked like a product. Long consideration windows are the opportunity, not the obstacle: every step in one is a query that somebody gets to own.
Each stage wants a different page type. Destination pages win the head term and route people onward: when to go month by month, how long you really need, a fair budget per person. Experience pages target what people want to do rather than where they want to go, which is how a lot of travellers actually think: walking with luggage transfers, a first safari with young children, diving for beginners. Those terms tend to be less contested than destination head terms, because guidebooks and marketplaces are organised around places, and the person searching them has already decided what kind of trip they want. Itinerary pages are the product. For attractions and day tours, Google Things to do is a further surface, and Google's Actions Center documentation encourages attractions, tour operators and activity providers to explore connecting through a connectivity partner first rather than integrating directly.
So sort your keyword research into the three tiers below, then map every page you already have onto them. Tier one is the enjoyable stuff to write. Tier three tends to come straight out of your reservation system. Tier two is where the trip is actually decided, and it is the tier worth counting first when you audit your own site.
- Inspiration, no trip in mind: best time to visit Patagonia, where to go in January for sun, is Jordan safe for families.
- Shaping and comparison, the trip is forming: Kenya or Tanzania for a first safari, is 10 days enough for Vietnam, small group versus private guided.
- Trip-shaped, ready to enquire: 10 day Vietnam tour, small group Galapagos cruise, private Japan tour cherry blossom.
Why thin tour pages lose to OTAs and content sites
Open your best-selling trip page and count the words a crawler can actually read. Not the text baked into the hero image. Not the day-by-day accordion that only fetches its content when a visitor clicks. Not the departures table injected by a script after load. If what survives that test is a short summary, a price-from line and an enquiry form, you are asking Google to rank a page that says less about the trip than a single reply from your sales inbox, written in language that would fit any operator running the same circuit. The marketplace listing for the same route has review text, question-and-answer threads and photographs from people who went.
Google's guidance on creating helpful, people-first content asks directly: "Does the content provide substantial value when compared to other pages in search results?" and "Is this the sort of page you'd want to bookmark, share with a friend, or recommend?" "Morning: transfer to Cusco, afternoon: at leisure" fails both, because it contains nothing that only you could have written.
Depth here is not padding. It is what your sales team already types out by hand, five times a week: what day four feels like when the altitude hits, how far you walk each day and on what surface, the real group size and who leads it, what is included down to domestic flights and guide tipping, why you route it anticlockwise, and what happens if a departure does not fill. Every one of those paragraphs earns long-tail rankings and removes an objection before the enquiry ever reaches you.
Link the guides to the trips they are meant to sell
On most operator sites the guides and the trips are strangers. The guide ranks, the trip page converts, and the only thing joining them is a related trips widget below the fold that nobody scrolls to. Fix it with links inside the body copy, placed at the point where the reader's question has just been answered, with anchor text that names the trip.
Here is the arithmetic. Every number in it is invented so you can see the shape of the calculation, so please do not read any of these rates as a benchmark: put your own in. Say a guide on Great Migration timing settles at 600 organic visits a month, and four in every hundred readers click through to the linked safari itinerary. That is 24 visits. Say that itinerary turns six per cent of visits into enquiries, so roughly 1.4 a month, 17 a year. Say you close one in four, so four bookings. At 7,000 EUR for a couple that is 28,000 EUR of revenue, and on a 15 per cent margin around 4,200 EUR of gross margin from one article that keeps earning next season too. Run the same sum with your own click, enquiry and close rates and you will know whether an article is worth briefing before you commission it.
That only works if you can see it happen. Organic enquiries convert weeks or months later, after two phone calls, a revised itinerary and a deposit, by which time last-click reporting has long since credited the brand search someone did the day they paid. Store the first-touch source on the enquiry record in whatever CRM you run, ours is OpenVoy, so the November article still gets the credit when the March deposit lands.
- Link from guide to trip in the first third of the article, not only at the end.
- Use descriptive anchors: our 9-day Northern Tanzania safari, not click here.
- Keep a sheet of which guide points at which trip, to spot articles with no commercial destination.
The technical basics that actually bite tour sites
Filters are the worst offender. Departure month, duration, budget band, activity level and group type multiply into an effectively infinite set of URLs. Google's crawling documentation on faceted navigation is blunt about the cost: crawlers typically work through a very large number of those URLs before deciding they are useless, which it calls overcrawling, and that time is time not spent discovering pages that matter. Its recommended fix is to disallow those URLs in robots.txt when you do not need them indexed. It also notes that Google Search generally does not support URL fragments in crawling and indexing, which is why filters built on fragments stay out of the way.
Second, duplicate itineraries. The same trip published as a seven-day, a nine-day, a private version and a 2027 departure gives you four near-identical pages splitting the same signals between them. Either consolidate to one canonical trip page where length and style are options within it, or make each version genuinely different in content, not just in the number at the top.
Third, sold-out seasons. Deleting a departure page throws away every link and every ranking it earned over a season. Keep the URL, mark it sold out, show the next departure and offer a waitlist. Seasonality is the one thing in this business you can predict: treat those pages as assets that go quiet in the shoulder months and come back, not as clutter.
- Check the day-by-day text sits in the HTML, not injected after a click or a script.
- Disallow filter and sort parameters in robots.txt; keep one clean crawlable URL per trip.
- Give every trip a stable URL with no year or departure date in it.
- Redirect retired trips to the closest live equivalent, not the homepage.
Local expertise is the one thing a marketplace cannot copy
Google's people-first content guidance leans on three questions: who made the page, how it was made, and why it exists. On experience, expertise, authoritativeness and trust it says trust is the most important of the four, and it gives knowledge that comes from having actually used a product or visited a place as an example of what it is looking for. If you have a ground team, local guides and a founder who has walked the route, that is a structural advantage over a marketplace, and most operators leave it entirely off the page.
The fix is not a trust badge in the footer. Name the author and say how many times they have run the route. Quote the guide in Cusco or the fixer in Amman by name, and say what they told you. Use your own photographs, including the unglamorous ones: the lodge at 6am, the minibus, the queue at the permit office. State the downsides, because an operator who admits day three is a seven-hour drive reads like someone who has done the drive.
- Byline plus a two-line bio with first-hand credentials on every guide and trip page.
- A last-reviewed date on permits, visas and seasonal timings.
- Reviews shown on the specific trip page they refer to, not pooled on a testimonials page.
Multiple markets without three copies of the same trip
Selling into the UK, Germany and the Netherlands is not one site in three languages. School holidays fall in different weeks, so peak departures move. Flight routings and add-on nights differ. Price expectations differ. And the search terms are not translations of one another: a German traveller looking for a walking holiday uses vocabulary, trip lengths and comfort categories that do not map cleanly onto the English phrasing, so the keyword research has to be redone per market rather than run through a translator.
Technically, Google's documentation on localized versions gives three ways to declare alternates, HTML link tags, HTTP headers or XML sitemap entries, and says the three are equivalent from its perspective. Pick one and stick to it. The rules that catch people out are that every version must list itself as well as all the others, that if two pages do not both point at each other the tags are ignored, and that x-default is there as the fallback for users whose language settings match none of your versions. The same page notes that localized versions count as duplicates only if the main content is left untranslated.
Machine-translated itineraries are therefore the worst of both worlds. They read as foreign to exactly the market you are trying to win, at the same time as handing an easy advantage to a local operator writing about the same route in its own voice.
What AI answers change, and what they do not
Google's documentation on AI features is direct: there are no additional requirements to appear in AI Overviews or AI Mode, and no special optimisations necessary. To be shown as a supporting link, a page needs to be indexed and eligible to appear with a snippet. The same page describes a query fan-out technique, where multiple related searches are issued across subtopics and data sources to build a response, which rewards covering a subject properly over publishing one shallow page per keyword.
What changes is which content still earns a click. Purely factual queries, how long the Inca Trail is, whether Vietnam needs a visa for a British passport, what a gorilla permit costs, are the ones an answer can resolve without anyone visiting a website. If your content plan rests on those, expect erosion. Queries that turn on judgement and trust, whether ten days is enough, which operator runs the route properly, whether this suits a nervous first-timer travelling alone, still send people to pages, because the answer depends on who is answering.
Google also says that when people click from search results pages with AI Overviews, those clicks are higher quality, meaning users are more likely to spend more time on the site. Change what you measure accordingly. Fewer but better-qualified visits is a perfectly good outcome for a business selling 4,000 EUR trips a handful at a time, and it reads as failure only if the number you report every month is sessions.
The mistakes I see most, and what to do instead
These are the ones I run into again and again on operator sites.
- Chasing the destination head term alone. Instead, win the shaping queries where the decision is made, then link across to the trip.
- Blog posts with no commercial destination. Instead, decide which trip an article is meant to sell before you brief it.
- One page per departure date or per year. Instead, keep one evergreen trip page with departures as data within it.
- Publishing volume instead of depth. Instead, make your five best trip pages more useful than the marketplace listing for the same route.
- Treating multilingual as translation. Instead, research each market separately.
- Judging SEO on sessions. Instead, judge it on enquiries and bookings attributed to first touch.
Where to start, in order
Start with what you already have. Pull twelve months of enquiry records and list the questions people asked before they paid a deposit: whether October is too wet, whether the children will cope with the driving days, whether ten days is enough, why you cost more than the operator down the page. That is your tier two keyword list, written by people who actually bought. Then check which pages already bring organic visits, and where each one sends the reader next. Adding body links from existing guides to the right trip pages is an afternoon of work that usually moves revenue before a word of new content is written.
Second, fix the leaks: filters generating crawlable URLs, itinerary text hidden behind scripts, four versions of the same trip competing with each other, departure pages deleted at the end of a season. None of it is glamorous, and all of it caps the return on everything you do afterwards.
Third, rebuild your three highest-margin trip pages properly. Real detail, a named guide, your own photographs, honest limitations, the things your sales team explains on every call. Use those three as the template, and measure enquiries per visit on them before you roll the format out across the range.
Only then publish, and go after the shaping queries first: less contested than the inspiration terms everyone fights over, and much closer to the money. One well-researched comparison piece a month, properly linked to a trip, will out-earn a weekly post that goes nowhere. Then give it two seasons before you judge it, because the booking it produces in March started with a search in November.