How Parimatch Use Data-Driven Approaches to Launch Experimental Betting Projects

Parimatch is changing the betting industry, and data is the key to these changes. Together with Head of BI and Analytics Parimatch Anna Gayvoronskaya, we’ll tell you how we applied data-driven approaches to launch the session game Footboss — a new gamified way to involve customers in the betting world.

Parimatch Betting Experiment: game Footboss

Footboss is a simple soccer-themed IDLE RPG game integrated into the Parimatch app: the player doesn’t need to register anywhere else if he already has an account in the system.

Games are designed for relaxation and chilled-out gameplay — conventionally, there are 5 minutes for which you can manage to go through a couple of bosses, acquire new skills and get prizes. Among the awards were Samsung Galaxy Note 10, iPhone 11 Pro Max, MacBook Air 13″, Sony PS4, iPad Air 10.5″, and other branded gifts from Parimatch.

Parimatch natively integrated betting into the game: the player could place a couple of bets and speed up the progress, but this was completely unnecessary.

How Footboos Engaged Players

The global goal of Footboss is to refresh the world of betting: the game was conceived as an alternative to the boring campaigns and promotional offers for Euro 2020. In addition to focusing on high odds, bonuses, and competitions in marginality, the goal was to give players variety and new emotions.

Thanks to Footboss, the player returned to Parimatch not only because he wanted to place a bet or check the odds but to take a break, win a few fights and get a prize in a process. The game developers specially made it session-based so players would not get stuck in it for hours and returned to their business after a couple of boss fights.

Footboss’s main gaming adventure is to free ten soccer stadiums from the giant invaders. The game’s key character is the Coach, whose task is to prepare his players for effective penalties and non-standard decisions day after day.

At first glance, the game seems simple, but with each new level, the player discovers many directions to deepen and develop their characters. Thanks to the random distribution of awards and achievements, the journey became unexpected: different things, tasks, heroes always dropped out. And then there’s the added interest of loot boxes: a kind of “kinder-surprise for adults.”

Ways to Use Data-Driven Approaches

Nowadays, in business, not enough to be guided by feelings or intuition. We must base global steps on data and analytics because analytics answers the question, “what can we do to make this product better?”

Before the launch of Footboss, a BI team completed the game’s mathematical model, which was the basis for the player’s interactions with Parimatch products. Also, data specialists developed a mathematical system for the equiprobable drop of prizes, according to which the player could get a prize regardless of how much money he spent on the game.

*quote*

Placing some bets could speed up the game’s passage, but this does not affect the dropout of prizes.

The compilation of models and comparison with the bases of active players of Parimatch helped to attract the number of players to Footboss, calculate the possible increase in GGR and the costs of the project.

For developers, data analytics prepared dashboards of the game progress, the number of participants, and their activity.

We frequently analyzed players’ engagement during the promotion period and made updates based on this data, for example, extra-levels or new characters.

We carried out the work with data at all stages of the project, including calculating the retrospective results. Here’s the data we’ve collected for future versions of the game:

• the number of players
• player engagement rate
• number of fights
• number of completed quests
• the average and the maximum duration of a gaming session
• number of open loot boxes
• game speed run
• the number of registrations on the site and application
• conversion of landing pages
• email open rate
• costs of prizes per active player
• the financial result of an experiment

Data to Highlight the Pandemic Damage

It’s hard to describe 2020 projects without mentioning the pandemic impact. The crisis and quarantine dealt heavy damage to the entire market, and the Footboss campaign was no exception. Everything went wrong: change of concept due to the cancellation of Euro 2020, anti-crisis measures in the company, limiting the launch of the game to 1 region out of 7 planned.

We calculated not only positive results but also what we received less due to the pandemic and crisis:

• loss of audience and GGR due to the cancellation of Euro 2020
• losses due to launch in only one region of the brand’s presence
• losses due to out of sync with the marketing campaign
• losses from the lack of integration of other company’s entertainment

Data-Driven Approaches for Future Experiments

The fundamentals of data-driven approaches lie in that data can be collected not only for retrospective activities but also throughout the entire process. You can build mathematical models at the beginning of the experiment, compare them with real performance, make adjustments, and achieve a positive effect for the business.

If earlier everyone in betting analyzed only the results of campaigns, now the data is used at all stages. We plan and build the mathematical models, make adjustments based on the results we obtained, and then we conduct a retrospective for future projects. Working with data has now become a continuous process from post facto analysis of results.

For Footboss data-driven, the approaches helped to calculate the number of potential players, the profit from their attraction, and the losses caused by the pandemic.

The first launch of Footboss opened the door to a new approach to attracting players to the world of betting. As CPO Parimatch Sergey Berezhnoy said: “We have built a stadium where we held the first championship.”

So new adventures await the players, and we are developing new quests and game levels — and all game activities will improve thanks to the data from previous versions of the project.