Smarter game discovery for a safer playing experience
Personalisation plays a key role in modern e-commerce, especially in services with broad product portfolios. When customers face too many options and no guidance, they often default to what’s familiar or abandon the service entirely.
In online casinos, the stakes are even higher. Slot games have different volatility levels, and choosing the wrong one can easily lead to a poor experience, especially for new or casual players.
As a responsible game provider, Veikkaus wanted to help users find games that actually match their preferences and playing style, creating a safer, more intuitive experience that also supports long-term engagement.
In my role as a Lead Designer for Veikkaus, I worked with a multidisciplinary team to ensure that customers weren’t left to navigate the game offering blindly.
The client
Veikkaus is a state-owned gambling company in Finland offering everything from lotto and sports betting to online slot games. Veikkaus emphasizes responsibility and aims to create a safe and entertaining experience across all touchpoints. Each year, the company channels over a billion euros back into society.
My role
Lead UX/UI Designer
Context
Veikkaus Casino offered a wide selection of slot games, but the browsing experience wasn’t working, especially for new users. There was little to guide players through the content, and many struggled to choose between unfamiliar games. This often led to two types of behaviour: sticking to a few well-known titles or leaving the service altogether.
For new or casual users, the problem was more than just an overwhelming choice. Slot games have different volatility levels, and without clear guidance, players could easily start with a game that didn’t match their playstyle, risking fast losses and frustration right from the start.
From a business perspective, this had a clear impact. When users couldn’t confidently explore the game portfolio, they engaged less, discovered fewer games, and were less likely to return. Helping users find the right games wasn’t just a UX issue; it was directly tied to customer retention and long-term value.
Objective
Help customers find games they enjoy to reduce drop-offs, improve first impressions, and increase long-term engagement. While also creating a proof of concept for product personalisation using machine learning.
Our approach
We mapped the customer journey with data scientists to identify the pain points and opportunities. We took a step-by-step approach, testing and measuring each idea before scaling it further.
Early Focus: Supporting first-time users
We started by focusing on first-time users, as they are most at risk of leaving. We built a model to identify common patterns among first-time users who returned. From there, we created a curated game set that was likely to resonate. A bandit algorithm selected which games to show to each user based on what had worked in similar cases.
We also used AB testing to determine the optimal number of game options to choose from. As too many would overwhelm, too few wouldn’t engage.
Structural fixes to support discovery
The existing casino UI relied on manual filters, which were mostly ignored by all but power users. Most visitors stayed on the landing page, which featured a few hand-picked games followed by an A–Z list of all titles. This was not a great first impression.
We simplified the structure, introduced a curated front page, and moved the full list of games to a separate subpage. Filters and subpages still exist, but now with a helpful sorting algorithm like a shop assistant who knows exactly what to show first when you say, “I’m looking for jeans.”
Driving engagement through smart recommendations
The previous landing page was based on a simple grid layout. All games were shown identically, with no variation, and no guidance. We wanted to offer something more engaging.
We introduced a new UI component called a themed carousel that lets users explore collections like “Based on Movies” or “Because You Played Kultajaska”. Some collections are manually curated and then filtered by relevance. Others are built entirely by AI. For example, the “Because You Played” carousel uses a bandit algorithm to analyze session behavior and surface related games.
As part of the project, we discovered through AB tests that strong visuals convert to a broader game selection per session. This was especially true with graphics that show characters or themes rather than just logos. Based on these results, Veikkaus is planning to bring richer game visuals into their personalization strategy going forward.