People in Greece buy insurance because they have to. They use it only when they need to.
That sentence is the cultural premise underneath every screen I designed for three years on Anytime, the digital brand of Interamerican. It is also the thing most global UX patterns get wrong the moment they are dropped into a Southern European insurance market without translation.
When I started, I thought the design problem lived in the emergency moment. The accident, the breakdown, the claim. Three years in, I think differently. The emergency moment matters, of course it does. But the real win lives in the small daily moments, the touchpoints that turn an insurance app into a tool a Greek driver opens between renewals, without anything having gone wrong.
This piece is about how I got there. It is also about the working tension that catches every designer working outside the Anglosphere sooner or later: patterns are global, but localisation is where the work begins.
Three forces working together
Western insurtech UX, the Lemonade and Wefox lineage, runs on one assumption. That customers want to engage with insurance the way they engage with banking or fitness. Frictionless onboarding. Speed at every step. Gamified updates. Fewer screens between intent and action.
In Greek consumer insurance, that premise breaks under three forces working together.
The first is cultural posture. Greek customers do not see insurance as a service they engage with. They see it as something they pay because they are required to. A piece of paper that lives in a drawer until something goes wrong. Speed at the wrong moment reads as carelessness here. Confirmation friction reads as institutional weight. Frictionlessness, the central virtue of Western fintech UX, actively erodes trust in this market.
The second is regulatory weight. GDPR is enforced more conservatively in Southern Europe than in the Anglosphere. Greek insurance regulation, layered on top of EU directives like the IDD, requires specific disclosure patterns, signed acknowledgements, paper-trail equivalents. A confirmation screen designed for a Stripe-style checkout does not survive contact with these requirements.
The third is the trust model. Trust in Greek consumer insurance is rooted in institutional permanence. The brand has existed for decades. Your parents bought policies from them. The building on the corner has had the same logo since the 1980s. A digital product that signals startup energy or Silicon Valley polish does not read as trustworthy here. It reads as foreign.
Each of these can be designed around in isolation. Together they form a trap that funnels most Western patterns into the wrong choice. This is the localisation problem. It is not translation. It is rebuilding the assumption layer.
Road assistance, and what shock actually does
The flow I am proudest of on Anytime is Road Assistance. It is the one users open during or right after a real roadside emergency. A flat tire. An accident. A breakdown on a motorway at midnight. The user is stressed. The user is in a place they do not want to be. The user has never used the flow before, and they are using it now because something has gone wrong.
The Western instinct here is speed. Reduce clicks. Get the user from intent to action in under fifteen seconds. Push notifications, geolocation, AI triage, instant routing.
That instinct is wrong for this user.
I watched it happen once. Someone using the flow under genuine stress. What stuck with me was not what the textbooks tell you about stressed users. The textbooks talk about hands shaking and shorter attention spans. What I actually saw was that the person's common sense had shifted. Words they would normally read without thinking became hard to parse. Ordinary text turned into a wall. Comprehension dropped in a way you can only really understand when you watch it happen in front of you.
That observation rewrote the brief.
The right design for a Greek user in panic is not faster clicks. It is reduced cognitive load through clarity. The first screen has to ask only the question that matters most. Where are you. What happened. Everything else answered by default. Location detected automatically, with a manual override for users who do not trust the auto-detect, and many do not. Vehicle pre-populated from policy data. Service type narrowed to the four most-requested options, with "other" as the escape hatch. Confirmation patterned to feel like a person on the other end has received and acknowledged the request, not like a form has been submitted into the void.
The flow takes under thirty seconds in the worst real-world case. Not because it is optimised for speed, but because it is optimised for a person whose common sense has shifted. The cultural posture, the regulatory weight, and the trust model all push the design toward something that looks slower than a Western equivalent. In the moment that actually matters, it works faster.
The daily layer
For most of the project, I thought the emergency moment was the whole story. Get the rare moment right and you have done the job.
I changed my mind somewhere in year two.
The win is not in the rare moment. The win is in turning the app into something a Greek driver opens between renewals. Without an emergency. Without a claim. For reasons that have nothing to do with insurance as a category and everything to do with the small daily texture of owning a car.
The features that prove this idea are the ones I did not see clearly when the project began. Policy details a driver can pull up the second a traffic officer asks for them. Documents in one place, easy to find when you want to check something yourself or when someone asks you to show them. Coupons for fuel. Coupons for the mechanic. Coupons for glass repair. Deals and offers that arrive in the app instead of in a paper letter. Competitions. Small things, all of them. Each one a reason to open the app that has nothing to do with the negative emotional weight insurance usually carries.
This is the move I want other designers in Southern European markets to see clearly. The localisation work is not only about respecting cultural posture in the heavy moments. It is also about giving the user enough light moments inside the same product that the relationship to the brand changes shape over time. As we grow, technology becomes a partner in our lives. Having your car documents in one place, easy to find when you want to check something or when someone asks to see them, becomes useful in a way an insurance app has not historically been useful.
That is the turning point I did not see when I started.
What I am willing to say after three years
Patterns are global. Localisation is the turning point.
The pattern says insurance app equals claims app. The local instinct says no, here it can also be the place you keep your car documents, the place you find a coupon for fuel before a long drive, the place a small deal lands when you are not expecting one. The pattern is not wrong. It is just incomplete. The work is in the completion.
I am still in the middle of this project. There is more to ship. But three years in, the lesson I am willing to put my name on is this. Any designer working outside the markets the canon was written for is going to live inside the same tension I live in. The pattern wants you to do one thing. The user in front of you wants something different. The job is to know which one to trust in which moment, and to build a product where both can sit inside the same shell without contradicting each other.