1 / 22

Using predictive analytics to drive game personalisation

Using predictive analytics to drive game personalisation. February 2012. Agenda. Who we are What is a nalytics? Predictive modelling and player segmentation Building personalised e xperiences A big b rother future?. ` Games Analytics. 30 + years games industry experience

daw
Télécharger la présentation

Using predictive analytics to drive game personalisation

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Using predictive analytics to drive game personalisation February 2012

  2. Agenda • Who we are • What isanalytics? • Predictive modelling and player segmentation • Building personalised experiences • A big brother future?

  3. `GamesAnalytics 30 + years games industry experience 15+ years dedicated to online & mobile games 15+ years data analytics experience with finance and retail sectors

  4. `ChangeTheGame

  5. So what is Analytics? Analytics is the process of developing optimal or realistic decision recommendations based on insights derived through the application of statistical models and analysis against existing and/or simulated future data – wikipedia Analytics is not

  6. It’s also not easy • Challenges • Big data • Complex player behaviours • Multiple monetisation mechanics • Overly focusing on whales • Making the data drive value • Never mind being expensive, resource and data intensive…slightly mind-bending and probably just a fad…

  7. The problem with Dashboards… • They almost never give you the information you actually need to action anything useful • They tell you about the average player • They tell you old information • They always look like this

  8. A/B Testing • Trial two versions and see which is most popular • Pick the most popular and roll it out to everyone • Repeat. • One size fits all • The Horizon Effect ….

  9. Funnel Analysis • Originated from web analysis • Great for linear progress and identifying ‘leaks’ • Cohort analysis • Multiple gameplay routes • Multiple monetisation mechanics • Works for simple social games • By its nature does not recognise multiple player types or non-linear gameplay

  10. Next Generation Analytics • Behavioural Segmentation • Social Analytics • Predictive Modelling • Real time in game messaging

  11. Z

  12. Who are my players? • A game’s player base is made up of lots of different player types • Each person is experiencing the same game differently • Understanding player behaviours is vital

  13. `Player Segmentation %Volume %Paying 7Day Ret CAC 7% 0.55% 36% $0.75 25% 1.30% 26% $2.21 6% 2.34% 57% $4.40 31% 0.89% 22% $1.75 Virality Potential 12% 0.86 59% $3.57 Early Enthusiasts 14% 0.97 21% $1.94 Confident Completers 5% 0.19 9% $2.38 Social Involver Sporadic SemiEngaged Losing Momentum Revenue Potential Need Guidance Borderline Incompetent

  14. The Power of Prediction • Once you understand your different players… • …You can start to predict what they want • and use this information to deliver immediate player value

  15. Building Predictive Models • Core predictive models in SAS & R • Multi-variant models can include 100+ separate variables • Each model allows you to target a set of users precisely • High propensity to take up the offer

  16. `Predicting Purchase Behaviour Variable Contribution • Score model at key points in the game • Start of Session • Start of Mission • After Mission Failed • Select players who have high model score (high likelihood to purchase) • Send message with offer/incentive 24 Hours + Gameplay High % GiftedItem Total Stamina 5000+ Level 7-12 Fighting Events Accepted Invite High PVP Low Mission Completion

  17. `Predicting Defection 150 Events Start 500 Events Country Age Gender Early Defectors Apply Model at 150 Events. Treat High Scores with Targeted Messages Level Momentum, Average Seconds Per Event, Socialness, Features Consumed Analysis Period Defectors Detailed Events: Quests Completed, Purchase Behaviour, Organising Tasks, Specific Missions Engaged Defectors and Engaged players behave differently in 1st 20 minutes Predict likelihood to defect and invoke retention activities before it is too late

  18. Using Data Effectively • Games data collection benefits from huge amounts of rich behavioural information • (When event collection is applied correctly) • Each individual player creates a complex decision path • Information can be mined and used to optimise gameplay

  19. The Players’ Views • Game design is generally focused on creating a great game • We need to look at it from the player perspective • Predictive analytics enables you to understand and identify behaviours to adapt gameplay to the player’s personal profile • Creating great personalised gameplay experience

  20. Big Brother…. • Data protection • Privacy • Exploitation of user profiles

  21. The Adaptive Game • Analytics provides a huge opportunity to deliver personal gaming experiences • Using the power of data for good • A new concept in game design • An incredibly powerful way of dynamically altering games • Adapting a game to players behaviour in real time • Player satisfaction delivers increased revenues

  22. Any Questions?

More Related