Shift change

What we were looking at

Gastfreund is, more or less, Slack for hotels. A set of communication products that a property uses to talk to its guests, and that staff use to talk to each other. By the time I left it was a suite of seven products, thousands of hotels, a forty-person engineering and product team. When this story starts it was smaller, and the bit of the business I was focused on was the hotel core: the thing the front desk and housekeeping and F&B would actually be opening every day if everything was working as intended.

The core was in trial with existing customers. Properties that already used us in other parts of the suite and had agreed to try the new thing. The kind of trial cohort you'd pick if you got to pick.

The problem we actually had

Trial usage was way below what we'd expected. Not slightly below. Significantly, in a way that wasn't going to fix itself with the usual round of tweaks.

The frustrating part was that the dashboards mostly said fine. Accounts were live, things were being clicked, the occasional flurry of activity. It was only when you stopped looking at the dashboards and started looking at the actual day in an actual hotel that you saw what was happening. Staff would open the app sometimes, half-fill something in, then go back to whatever they'd always done. Their own phones, a WhatsApp group, paper, the shift book at the desk.

The person who'd agreed to the trial was sitting in a back office or a manager's chair. The one being asked to actually open the app forty times a shift was the 6am front desk person, and the 2pm housekeeping lead, and the F&B manager prepping breakfast. None of them had agreed to anything in particular. There was no reason for them to care unless we gave them one.

What I did

Diary studies, on site, across multiple properties: front desk, housekeeping, F&B, the lot. Not a workshop, not a survey, the actual rhythm of the actual job, over enough shifts to see what people did when nobody was telling them to do anything specific.

A lot came out of it. Most of it was useful in a small way. One thing was useful in a different way.

The Übergabe. The handover between shifts.

Every department had one, and they didn't really look like each other (the front desk handover is not the housekeeping handover is not F&B's morning lead briefing), but every department had one, and every department treated it as the moment that mattered most in their day. It's where everything someone might otherwise forget gets pushed across to the next person who needs to remember it. The unfinished room. The guest in 314 who'd changed her breakfast for tomorrow. The maintenance thing that still hadn't been done. The allergy. The complaint that hadn't been resolved. The routine checks that needed signing off.

The Übergabe is the most load-bearing twenty minutes in a hotel, and in almost every property we sat in, it was being done with a notebook and a hope.

What made this finding useful wasn't only that it was important. It was that the same shape of problem turned up in every part of the building, in slightly different clothes. Designing for one department's handover didn't isolate the work into a single feature for a single team. It pointed at a moment that mattered everywhere, in similar-but-not-identical ways. One thing to build, for every team in the property, with real leverage.

The magic moment

The thing trial usage had been missing, the reason a free trial wasn't converting into the kind of behaviour you'd expect from someone genuinely trying a thing out, was a moment. A specific one.

It was the moment a team ran a handover through the product and felt what the product had been trying to do all along. Not the features on the page. The felt sense that nothing was being lost between shifts. That if something was unfinished, the next person would know. That if the guest in 314 had changed her preference for tomorrow, it was already there. That if there was a routine check to be done, it was on the list, waiting. The shift change stopped being a thing they had to hold in their heads and a stack of paper, and started being a thing the software took care of, end to end.

That feeling, nothing is slipping, was what the software had been trying to deliver from day one. It was just very hard to communicate from a sales deck, or to feel from looking at a settings screen. You had to live one shift change with it before it made sense.

Once we got that moment to happen reliably, in the product, in their actual Übergabe, the usage curves started behaving the way trial usage is supposed to. Opening the app for everything else stopped being a chore. The handover became the thing the product was best at, and once it was, the rest of the suite earned its place by being there.

What it opened up

Trial usage caught up. Onboarding time dropped by more than forty hours, a 31% improvement, mostly because we stopped trying to teach staff features they didn't care about and could lead with the moment they did. The rebrand and a full new illustration set across the suite came out of the same direction; everything could finally point at the same idea. And alongside the product work I built out a design culture that hadn't existed before. A design system, accessibility standards, fast internal sprints, recurring workshops that put execs and floor staff in the same room, a bi-monthly newsletter to keep it visible.

This is the kind of work I love most. A product that looks fine on the surface, with usage that quietly isn't. The fix isn't more features or louder onboarding. It's finding the one repeated moment that everyone in the building genuinely cares about, and letting the product be excellent at exactly that. If you've got something shaped like that, I'd love to hear about it.