Genesis: from nothing to two travel apps

Voyra started on 22 January 2026 as a single commit and a hosting config. Nothing else. No accounts, no map, no design, just an empty repo and an idea that planning a trip should not mean juggling six tabs and a group chat. Five weeks later there were two working apps online: a day-by-day trip planner, and a map where you drop stops and build a route that hands itself across to the planner. This is the story of that first stretch, written as it happened.

Week one: a planner that exists

The first thing I wanted was the simplest honest version of the idea, a trip laid out across its days, in a browser, that I could actually open. By the end of the first weekend that existed. A real, deployable web app with a day-by-day Timeline view, set up so it loads cleanly and the links work no matter which page you land on first.

It was about as raw as software gets. The author name on the very first commit was wrong and had to be corrected after the fact, which tells you how fresh the ground was. But the core principle was already there from line one, and it has never moved since. The itinerary is the single source of truth. Everything else in Voyra reads from it. The map, the timeline, the today view, the bookings. Nothing competes with the trip you're actually planning.

The week after, I called it version one. Honestly that label took a round of wrestling the build into shape before it would deploy. The "this is complete" moment and the "fix the build so it actually ships" moment landed back to back. That gap between done and not quite is most of what early building is.

Week three: it can belong to you

A trip that only lives in one browser tab is a note, not a plan. So the next big step was accounts. Sign-in arrived, and with it the thing that makes a planner feel real, a trip could belong to you, not just to the machine you happened to open it on.

I made a deliberate choice here that has stuck. You can preview a trip before you sign in. You only need an account when you want to save it. No wall in front of someone just looking around.

Getting account-based saving to behave took a lot of trial and error in a single day. That is the unglamorous truth of it. The same week, a second app was born: Voyra Explore. A map where you drop stops, build a travel route, and let it sort the order of those stops by distance. And, crucially, send a finished route across to the planner. Two apps, one trip. The map is for discovery and shape. The itinerary is where decisions land.

The map is a view, not a decision-maker. Voyra organises the trip you're already planning. It never decides it for you.

Week four: the one I keep coming back to

Week four is the week the whole philosophy of Voyra showed up, in a single small reversal that I still think about.

Explore briefly gained real road-following routes through a mapping service. Proper roads, proper turns, the lines bending along actual highways. It looked clever. And then I took it out, on purpose, and went back to plain straight lines between stops.

The reasoning is the entire point of Voyra in miniature. A route should read the same whether you drive it, fly it, or take the train. A bent road line quietly lies to you the moment your next leg is a flight. A straight line says one honest thing: here, then there. Clarity beat the flashier feature, and that instinct, remove the clever thing and keep it calm, became the spine of everything I have built since.

The rest of the week was quieter craft. Explore's place search got much better at finding the location you actually meant. You could set how many nights you stay at each stop and drag to reorder. The planner learned to auto-fill trip dates and run consecutive days for you, after a couple of corrections before it felt right. Expired sign-in links started failing gracefully instead of dead-ending. Small things, but small things are most of trust.

Week five: a front door, and a hint of together

The last week of this window is where Voyra first started thinking about strangers, not just me.

The planner got a public landing page with a live, interactive demo, with sign-in gating the real app behind it. The idea was simple: let someone touch the thing before they commit to anything. Signed-in travellers started landing straight on their own dashboard instead of the marketing page. And the date handling got tidied so a route imported from Explore carries its days through cleanly.

Then a first seed of something bigger: Explore gained an early board-based way to plan a route together with other people. It was rough, and group planning would take many more months to become real, but the direction was set. A trip is often a "we", not just a "me".

Five weeks. From an empty repo and a corrected author name to two apps online, accounts, a map that talks to a planner, a demo you can touch, and the first deliberate choice to keep things calm rather than clever. The foundations were there. They just weren't permanent yet. Everything still lived in the browser, which meant clearing your history could clear your trip. Fixing that, moving every trip into the cloud and making it follow you across devices, is the next month's whole story.

If you want somewhere calmer to plan your next trip, start planning. It is live and free to use.