Our CEO Adrian takes the Lötschberg car tunnel between Canton Bern and Canton Valais at least every couple of months. He could drive around via Montreux or through the valleys, but the tunnel is faster — so he takes it. And every single time, buying the ticket online costs him more nerves than the drive itself.

He asked us to make it episode two of Roast My Product. We said yes immediately.

What Adrian actually has to do to buy a ticket

Here's the current flow, step by step:

He visits bls.ch, selects his journey with a date and time, and lands on a trip planner where he can check the departure timetable. Fine so far.

Then he selects his journey and gets redirected to a completely different website — different domain, different design, different everything. That's already a trust problem. But it gets better.

On this new site, he goes through five steps. Step one? Select his journey. Again. The same journey he just selected on bls.ch. Then he picks a time. Again. Adds it to a shopping basket. Step two: enter his credentials. Step three: payment. Step four: booking confirmation.

And then he waits for an email with his ticket, which he needs to show at the counter when he drives onto the car carriage. Hope it didn't land in spam.

That's thirteen actions to buy what is essentially a one-way car train ticket. For a journey he makes regularly. Every time.

The problems

Too many steps. This should be obvious, but thirteen steps to buy a train tunnel ticket is absurd. People give up. They drive to the tunnel entrance and buy the ticket at the counter for 3 CHF more — which doesn't even guarantee them a spot on the carriage. That's a real risk.

The redirect kills trust. Sending users to a completely separate domain mid-purchase is a UX anti-pattern. It looks like a phishing attempt. The URL changes, the design changes, and suddenly you're wondering if you're still on the official BLS site.

Duplicate selections. Asking someone to choose their journey and time twice is the kind of thing that makes users feel like the system isn't listening to them. It's not just annoying — it signals that the two systems don't talk to each other.

Too many options on screen. Tabs, buttons, sidebars — the interface throws everything at you at once instead of guiding you through a focused flow.

Email-based tickets create friction on both ends. BLS needs backend infrastructure to generate and send tickets. Users need to dig through their inbox at the tunnel entrance, possibly on a phone with poor reception in the mountains. If it lands in spam, you're standing at the barrier holding up traffic.

Our redesign

We went through the current flow several times and designed a simplified version. Three screens. That's it.

Screen one: choose your journey. Pick your route, pick your date, buy one ticket or multiple. Two ticket types — weekday and weekend. No trip planner detour, no timetable page, no redirect to another domain.

Screen two: log in or sign up. Existing BLS account? Log in. New user? Account creation in under 40 seconds.

Screen three: pay. One button. Done. Ticket available for download right there, and also sent to your email as a backup.

Thirteen steps down to three screens. The information BLS needs is exactly the same — we just stopped asking for it twice and cut the unnecessary detours.

The principle

Same thing we said in episode one: simplicity. Not "make it look minimal." Make it do less unnecessary work. Every extra click, every redirect, every repeated form field is a chance for the user to bail. The best UX is the one that gets out of the way.

If you're building a product and want this kind of honest UX teardown, that's what we do. Design, development, the whole thing — check out our CTO as a Service offering. Direct contact via Discord, regular sprint calls, and a product team that actually cares about your users' experience. Shouldn't take more than six minutes to know if it's the right fit.

Roast my Product Episode 2: BLS Car transportation flow rework

Latest blogs