SnowTeach is the newest thing we've shipped at Maladeta Studio, and it's also the first thing we've shipped that doesn't work like the others. NaturPark manages natural spaces for the team that runs them. MeteoTrail gives race organizers weather data for one route, on one day. Both are single-tenant, in the sense that matters here: one client, one deployment, one set of needs we design around. SnowTeach is different. It's a marketplace — ski and snowboard schools and instructors list classes and activities in the snow, and customers book them online. On paper it sounded like a small step from what we already knew how to build. It wasn't.
1. From one client to two sides of a market
With NaturPark, we sit down with a park manager, figure out what a park actually needs to run, and build for that team. There is one customer in the room. With SnowTeach there's no one customer in the room — there are two, and they don't want the same thing. Schools and instructors need to manage availability, pricing and who's teaching what, without babysitting a tool for it. Skiers and snowboarders need to find a class, compare it against a handful of others, and book it with enough confidence to pay before they've met anyone.
That split shows up in decisions that would never have come up before. A school wants some control over how it looks next to competitors; a customer wants an easy, fair comparison across schools. Neither side is wrong. It's just the first time we've had to design for two answers to the same question instead of one.
2. Trust becomes a product problem, not just a school's problem
In single-tenant software, if something goes wrong on the ground, it's the client's problem to handle, and our software's job is mostly to help them see it and log it. A marketplace doesn't get that shortcut. If a booking on SnowTeach goes badly, the customer's experience is with SnowTeach first, and with the school second. That means trust and quality control, which used to be someone else's job, become work the product itself has to do.
- A way for customers to see who they're booking with before they pay — verified school and instructor profiles, not just a listing
- Reviews and ratings that carry weight, so a bad experience is visible rather than buried
- Clear, consistent cancellation and refund rules that don't depend on which school you happen to book with
- A way to flag when something goes wrong that a real person looks at, not a support inbox that swallows it
3. The chicken-and-egg problem, honestly
The last thing that's genuinely new is the business problem underneath the product one. A marketplace with no schools has nothing to book, so customers don't come back a second time. A marketplace with no customers gives schools no reason to spend time listing their classes properly. Both of those are true at once at the start, and no amount of good UI fixes either of them on its own — this is a liquidity problem, not a design problem.
The honest answer, at least so far, is that we didn't try to solve it everywhere at once. We started narrow, in the mountains we already know, working directly with schools we could actually call, instead of betting on self-serve sign-ups to fill both sides on their own. It's slower than it looks in a pitch deck.
SnowTeach is still young, and what building it has mostly taught us so far is how different a marketplace's problems are from the single-tenant software we're used to. If it works — if enough schools and enough skiers end up booking through the same platform — the question I keep coming back to is whether there's a more general version of this underneath it: not just a marketplace for snow-sports classes, but for other outdoor activities running into the same two-sided problem. I don't know the answer yet. That's honestly the point of writing this down.