A historic building almost never has a page about itself in English. It has a booking engine, an opening-hours box, and a paragraph in the local language. So when a traveller in Toronto or Seoul types its name, something else answers. We spent the last month measuring what that something is, on our own pages, and the numbers turned out to be more interesting than we expected — and less flattering in one particular way.
What we measured
Cultural Heritage Online publishes 16,952 heritage pages. Each one is a single place: a building, a monument, a museum, a garden. They are written in English, sourced, dated, and permanent.
We took two groups and asked Google Search Console what actually happened to them over 28 days.
Museums and galleries. We publish 2,066 of them. 1,700 received search impressions — that is, Google showed them to a human being who was looking for something. Together they appeared 58,191 times and were clicked 511 times.
Historic hotels. We publish 360. 303 received impressions, 8,615 of them, and 76 clicks.
Read those two paragraphs again and you will find the uncomfortable part, because we did: the click-through rate is 0.88%. Most of the people who are shown the page do not open it.
We could have left that number out. We are leaving it in, because the interesting thing is not the clicks. It is where the pages sit.
Position is the number that matters
Among the hotel pages, the average position in Google results included 4.6, 5.2, 5.3, 5.9, 6.6. Among the museum pages: 5.6, 5.7, 6.0, 6.7, 7.1. Those are first-page positions, in English, for buildings whose owners publish nothing in English at all.
The Hôtel Tassel in Brussels — Victor Horta, 1893, the building usually named as the first Art Nouveau house in Europe — appeared 617 times in 28 days from our page alone, at an average position of 8.5. The Grand Hôtel Villa de France in Tangier sat at 5.3. Neither of them wrote a word of it.
This is the mechanism, and it is worth being plain about it: the page exists whether or not the building’s owner knows about it. Someone is going to answer the question “what is this building”. The only real question is who, and where that answer sends the reader next.
What is unusual about this, and why almost nobody else does it
Three things, and none of them is a marketing claim — they are simply how the archive is built.
It is one place per page. Not a city listicle with forty entries. A directory entry competes with every other directory entry for a generic phrase. A page about one building, with its architect, its date, its protection status and its sources, competes for the building’s own name — and that is a search almost nobody else is fighting for.
It is written in English and it stays. Most heritage material in Europe exists in the local language, inside a municipal site that is redesigned every four years and loses its URLs. Our pages have a permanent address. The oldest content in this archive has been continuously published since 1998.
It is editorial, not promotional. The pages carry sources, and corrections from the people who run the buildings are applied and credited. That is what makes them citable — by a reader, by a journalist, and increasingly by the AI assistants that people now ask before they book. A page that reads like an advertisement is not used as an answer by anything.
What an owner can actually do with a page they did not write
This is where the story usually stops for us and starts for them.
The building’s page can carry a marked link to the owner’s own booking page. Not a mention: a button, with tracking parameters attached, so that the visit appears in their analytics as a referral from culturalheritageonline.com. They do not have to take our word for anything. They open their own dashboard and count.
That single detail changes the category of the thing. Most visibility is sold as belief. This is sold as a number the buyer already owns.
Around it sit the other pieces: their events published on the dates they actually run, a walking route through the heritage nearby with their building as its destination, downloadable so their visitors can follow it on a phone on a day when there is no guide, and a quarterly written review of what people typed into Google to arrive — which, for most institutions, is the first time anyone has shown them the actual words.
The honest limits
We publish the page whether or not anyone pays us, and we do not remove it if they decline. A free page is a real page: sources, photographs, corrections, events. Many places stop there and it is a perfectly good outcome.
And we will not pretend the traffic is larger than it is. 0.88% is the number. What we are selling is not a flood of visitors; it is the position of the answer, the link out of it, and somebody who works on the words over time.
Frequently Asked Questions
Does my building already have a page?
Possibly. There are 16,952 of them, and they were written from public sources without asking anyone’s permission, because that is what an archive does. If yours is there, it stays there at no cost, and you can have anything on it corrected by writing one line to the editor.
What does it cost to be listed?
Nothing. The listing, the sources, the photographs, the corrections and the publication of your events are free and stay free. The paid partnership is a different thing: placement, an editorial case study written about you, the tracked link to your booking page, and continuing work on the search terms that bring people to you.
How do I know it works?
By looking at your own analytics, not ours. The link from your page to your booking page is tagged, so visits from us appear in your own reporting as referrals. And every quarter you receive the search terms people actually used, taken from Google Search Console, whatever they say.
What if it does not work?
If we do not deliver what is written in the plan within the first six months, the money comes back. We would rather refund one partner than describe a service we did not perform.