Every visitor who walks through your gate is holding a phone. Most of them will use it to photograph a view they cannot explain, read a plaque three people are already standing in front of, and leave knowing less than you wish they did. The place is extraordinary. What surrounds it — the signage, the leaflet, the guide who can only be in one spot at a time — has not changed in thirty years.
What I do
I build the layer between your visitors and what you know about your site. That means augmented reality tours a visitor runs on their own phone, software that answers their questions in their own language before and during the visit, and the quiet machinery behind both: what to show, where, and to whom.
This is not a marketing app. It is the difference between a visitor who saw a view and a visitor who understood it — and, for you, between a ticket sold once and a place people recommend.
Augmented reality tours, in plain terms
An augmented reality tour puts what is no longer visible back into the view. A visitor holds up their phone at a lookout point and sees the valley as it was two hundred years ago. They stand at a ruin and watch the building complete itself. They point at a ridge and the animals that move through it in the season they are not visiting are named and followed. The phone knows where they are standing and what they are looking at; your knowledge supplies the rest.
Done properly it works within the limits your site actually has, and that is most of the engineering. Signal is poor or absent in the places worth standing, so the tour has to work offline. Visitors arrive with phones that are five years old and batteries that are half empty. The site itself cannot be re-engineered around the software: no beacons drilled into a heritage wall, no cables across a reserve.
A virtual tour is the same content reaching people who are not there yet — a school on another continent, a traveller deciding between your attraction and someone else's, a visitor with mobility needs who can now see the part of the site they could never reach. It is the same build, released to a much larger audience.
One I am building
Vitruvian AI · in build, launching late 2026
Hindsite, at Kolmanskop
Hindsite is the first application I am building inside Vitruvian AI, my own firm. A visitor will unlock the tour at the gate and walk it on their own. The narration starts when they reach a stop, the plaques answer when tapped, and the camera fills the ridge back in. Out there in the Namib none of it can lean on a network, so none of it does.
Heritage sites run the same build as their own app — their name, their colours, their staff writing the stories.
See HindsiteThe engineering behind it is the engineering documented in the case studies: systems that had to work in the field, on ordinary devices, for people who were never going to read a manual. The rest of my working software lives in the lab.
The kind of work this is
- Tours a visitor runs on their own phone, working offline, with the story attached to the place it belongs to.
- Answers to visitor questions before and during the visit, at any hour, in the language they asked in.
- Virtual access for the people who cannot come — schools, distant travellers, visitors whose mobility keeps them from the far side of the site.
- An honest picture of how people actually move through your site: where they linger, where they turn back, and which parts of it nobody ever finds.
- A record of what your guides and curators know, held somewhere it cannot retire.
What a first project usually is
Not the whole site. One route, or one viewpoint that everybody stops at anyway, built properly and put in front of real visitors for a season. That is enough to learn what people actually do with it — which is never quite what anyone predicted — and small enough that a disappointing answer costs you very little.
The parts that follow are easier, because the hard work was done first: the way content is authored, how it reaches a phone with no signal, and who inside your organisation keeps it current after I am gone.
Questions operators ask
What is an augmented reality tour?
A tour a visitor runs on their own phone, where what they see through the camera is added to rather than replaced: the valley as it looked two hundred years ago, the ruin completed, the animals named on the ridge they are pointing at. The phone knows where they are standing; your knowledge supplies everything else.
Do visitors need to download an app, or good signal?
A tour that depends on either will fail at exactly the places worth standing. Most of what I build runs in a phone's own browser and works offline once it has loaded, on ordinary phones several years old. Where a downloadable app genuinely serves the site better, that is a decision the assessment makes on evidence, not on fashion.
We are a heritage site. Can this be built without touching the place?
Yes, and it usually must be. Nothing is drilled, mounted or cabled into a protected site. The tour works from what a phone can already sense and from a map of your site made once, which is why this approach suits national parks, reserves and heritage places where physical changes are rightly difficult.
What does it cost to find out whether this is worth it?
From €9,500, as a fixed fee. The feasibility assessment takes three to four weeks and ends in a written verdict covering what your site could support, what visitors would actually use, and what building and running it would cost. If the honest answer is that your attraction does not need this, that is the verdict you get.
Where to begin
With the feasibility assessment: a fixed piece of work, from €9,500, that tells you whether what you are imagining is worth building for your site, your visitors and your budget — before any of it is built.
When the verdict is build, you keep the same pair of hands. I design and build the system myself and stay until it works through an ordinary season of yours.
If your site also has beds and a restaurant, the hospitality work sits alongside this one, and most places need a little of both.
Start a conversation