2. Miroslav Pikhart
Geewa.a.s.
Head of BI
miroslav.pikhart@geewa.com
Why your game needs a data scientist?
Plus some use cases from Smashing Four
3. About Smashing Four - core game
● Free-to-play PvP game
● Teams of four heroes fight in the arena
● Characters resemble pool balls
● Each hero has an ability
4. About Smashing Four - economy
● Players get orbs for winning or buy them in shop
● Orbs contain random hero cards and gold
● Players collect these resources to upgrade heroes
and reach higher arenas
● Higher arenas lead to higher rewards, completing
the loop
5. Why your game needs a data scientist
and what personalized content can do for your revenue
6. Why your game needs a data scientist
and what personalized content can do for your revenue
7. Why your game needs a data scientist
and what personalized content can do for your revenue
8. How data can help improve your game
● Learn when and why players are leaving
○ Improve new player experience
● Understand how players play your game
○ Keep possible strategies balanced
○ Forecast how many players will use new content
● Find out what players want the most
○ Tailor progression to your players’ needs
○ Increase revenue through content personalization
12. Identifying when players leave your game
There’s more...
● The problematic goes beyond a simple funnel
○ Not only you need to know where you lost your players
○ It also helps to know why - is this part too long? Too boring?
● Similar, more ‘advanced’ use case are churn models
○ Basically predicting whether a player is likely to leave your
game
○ Using features such as spend, winrate, matches per day...
13. Sometimes things go off balance
● Another area where data shine is keeping the game in balance
● Sometimes things interact in a way you did not anticipate
○ This happens often, especially if your game is complex
● If anything suddenly changes - you have to know why
15. Using data to drive revenue increase
● With all the data you’re getting, you know all about your players
● That puts you into the perfect position to leverage all this data
● Why would you try to sell the same thing to everyone if you know
that they are not the same?
16. Intermezzo - Special offers in Smashing Four
● Essentially a chance to buy bundled orbs and
currency at a discount
● Only available at limited times
● Positive impact on conversion
19. Data usage case study - Hero orb
● Goal was to create a new product to improve monetization
● Instead of pushing richer special offers, we wanted to offer our
players something that they personally would like
● In order to reach wider audience, low price-point to start with
21. Hero orb - what’s behind it
● There’s a relatively simple algorithm behind the
scenes that scores heroes based on:
○ How often players use the hero
○ How often they request him from their clan
○ How much they upgrade him
22. Lesson learned #1 - Define your tests carefully
● When we first created it, we tried to A/B test the feature:
○ One group got the hero orb offer for $5 USD
○ The other got the very same offer for 500 gems (in-
game currency that’s worth exactly $5 USD)
23. Lesson learned #1 - Define your tests carefully
● Then we looked at the conversion to payer...
24. Lesson learned #1 - Define your tests carefully
● Then we looked at the conversion to payer...
25. Lesson learned #1 - Define your tests carefully
● Based on the figure just shown, the product
originally prefered to go with gems, however...
26. Lesson learned #1 - Define your tests carefully
● Based on the figure just shown, the product
originally prefered to go with gems, however...
27. Lesson learned #1 - Define your tests carefully
● The initial graph was significantly skewed by ‘old’ players
● These only wanted to find a way to spend their ‘free’ gems
● This shows how important it is to set your metrics precisely
28. Lesson learned #2 - Don’t test for the sake of testing
● This is the most common mistake I have encountered so far
● Sometimes people just love A/B tests too much
○ They are easy to set up and perform
○ Even if they sometimes offer almost no information
● Knowledge of statistics is a big asset in this area
30. Lesson learned #2 - Don’t test for the sake of testing
● Specify the hypothesis you want to prove before running any
test
● Come up with some KPIs you can measure that are both
relevant and not corrupted
● Always have a control group to measure against
31. Hero orb - the results
● The product turned into our best performer and still runs strong
34. Summary
● With the rising trend of GaaS, data became significantly more important
● Data helps you improve both player experience and your revenue
● There’s (almost) never enough data
○ But it’s not that easy to differentiate the vital information from the rest
● Collecting data is only the first step
○ Then you need to interpret them and use the knowledge
35. So why would you want a data scientist?
● You are technically able to collect data down to a single click level
○ Having someone who can draw significant conclusions is a boon
● You can get answers to business-related questions like:
○ How much is this player likely to spend in my game?
○ How many players will actually reach endgame content?
○ How overpowered is this new item we designed?
○ If I create this new product, how many players will see it?
36. Why would you want to work in game dev?
● The whole field is only a few years old
○ There are problems yet to be completely solved
○ There aren’t agreed upon best practices for everything
○ You can still discover something innovative
● The amount of available data is huge
○ You’re working with large, complete data sets
○ The amount of possible features is (almost) endless