THE IMPORTANCE OF THE
FUTURE

Lloyd Melnick

Casual Connect
2012
Agenda

Background

Current state of analytics

Analytics 2.0

Summary
BACKGROUND
Lloyd Melnick
FiveOneNine Games




    Capitol     FiveOneNine
                              EW Scripps
 Broadcasting      Games
Our Business Strategy


    GAME                   GAME
  DEVELOPER              PUBLISHER
 • Campaign Story       • Publishing
   launches next          Program
   month                  launched this
                          week
Analytics and gaming



                       Statistical Tools
                       Predictive Models
                       Data Mining




                        In-Game
                        Analytics Tools




                                   7
Analytics and Gaming

     In-Game
                       Statistical Tools
     Analytics
                        (Predictions)
   (Dashboards)
 • What                • What will
   Happened?             Happen?
Customer Lifetime Value
Most Important Metric



                  Future     Better
   Current                  Business
                Predicted
   Lifetime                 Decisions
                 Lifetime
    Value
                  Value       $$$$
CURRENT STATE OF ANALYTICS
In-Game Analytics Tools

                    DAU

                              Session
     K-Factor
                               Times

                  Lifetime

     Virtual
                                2-day
     Goods
                              retention
      Sold
                    7-day
                  retention




               Visualization of Game Metrics
                                               11
In-Game Analytics
                          • Easy Visualization of high level metrics
 Executive Dashboards


                          • Ability to address immediate concerns
   Ad-Hoc Reports


                          • Where exactly is the problem?
   Query/Drilldown


                          • What actions are needed?
        Alerts



      Kontagent, Mixpanel, Honeytracks, CollectTM & MeasureTM
      (by GamesAnalytics)                                              12
ANALYTICS 2.0
STATISTICAL Analysis
                          • Are there statistically significant
   Correlations and CHI     associations among factors?
            SQ

                          • Are there statistically significant
    T-tests & ANOVA         differences among groups in usage or
                            monetization?

                          • What are the factors (i.e.: gender, age)
   Regression Analysis      that “significantly” impact revenue and
                            by how much?

                          • How are the high value players
     Outlier Analysis       monetizing differently than most
                            players?




           Excel, R, SAS, SPSS, STATA, SWRVE, PredictTM
                                                                       14
Forecasting



                          • How much revenue will we bring in next
                            quarter?
  Time Series Analysis    • How many users will we have in the
                            future (near term)?




               Excel, R, SAS, SPSS, STATA
                                                                     15
Predictive Modeling

                           • Who is more likely to monetize?
        Logistic           • Who is more likely to react to in-game
  Regressions, Decision      messaging?
       Trees, etc.



                           • What will be the lifetime of a user?
                           • How long will it take for users to
    Survival Analysis        monetize? (time to first purchase)
                           • What factors impact the retention of
                             the users?




           R, SAS, SPSS, STATA, PredictTM, MeasureTM
           (by GamesAnalytics)                                        16
Data Mining
                         • Are there clear “segments” among our
      Clustering           users that could be approached
 (Segmentation) Analysis differently?


                           • Are there items that sell “together”?
   Association Analysis



                           • How do users feel about our games?
                           • What are the main topics of conversation
 Text Mining (Consumer       for our Twitter followers?
  Sentiment Analysis)      • Are comments on our Facebook page
                             mostly positive or negative?




            SAS Enterprise Miner, SPSS Modeler, Weka, PredictTM
                                                                        17
Simulation



                      • How long do tasks in our game take to
                        play on average?
   Monte–Carlo        • What happens if we tweak the rules of
    Simulation          the game?




        Excel, R, Risk Solver, SAS, SPSS, STATA
                                                                18
Optimization


                           • What is the optimal price for the virtual
                             goods?
   Price Optimization



                           • What is the optimal allocation of
                             resources for supporting the game?
   Linear Programing




          R, Risk Solver, Oracle Cristal Ball, SAS
                                                                         19
SUMMARY
Huge opportunities to use analytics
better
Analytics is the foundation of business and games deep
insight

They help you optimize production and marketing
decisions

They help improve the game and the user experience

Adding predictive modeling to in-game analytics lays the
foundation for additional optimization of business decisions
Thank you!




Lloyd.melnick@fiveoneninegames.com

Twitter: Lloyd Melnick

http://lloydmelnick.com/

The Importance of the Future

  • 1.
    THE IMPORTANCE OFTHE FUTURE Lloyd Melnick Casual Connect 2012
  • 2.
    Agenda Background Current state ofanalytics Analytics 2.0 Summary
  • 3.
  • 4.
  • 5.
    FiveOneNine Games Capitol FiveOneNine EW Scripps Broadcasting Games
  • 6.
    Our Business Strategy GAME GAME DEVELOPER PUBLISHER • Campaign Story • Publishing launches next Program month launched this week
  • 7.
    Analytics and gaming Statistical Tools Predictive Models Data Mining In-Game Analytics Tools 7
  • 8.
    Analytics and Gaming In-Game Statistical Tools Analytics (Predictions) (Dashboards) • What • What will Happened? Happen?
  • 9.
    Customer Lifetime Value MostImportant Metric Future Better Current Business Predicted Lifetime Decisions Lifetime Value Value $$$$
  • 10.
  • 11.
    In-Game Analytics Tools DAU Session K-Factor Times Lifetime Virtual 2-day Goods retention Sold 7-day retention Visualization of Game Metrics 11
  • 12.
    In-Game Analytics • Easy Visualization of high level metrics Executive Dashboards • Ability to address immediate concerns Ad-Hoc Reports • Where exactly is the problem? Query/Drilldown • What actions are needed? Alerts Kontagent, Mixpanel, Honeytracks, CollectTM & MeasureTM (by GamesAnalytics) 12
  • 13.
  • 14.
    STATISTICAL Analysis • Are there statistically significant Correlations and CHI associations among factors? SQ • Are there statistically significant T-tests & ANOVA differences among groups in usage or monetization? • What are the factors (i.e.: gender, age) Regression Analysis that “significantly” impact revenue and by how much? • How are the high value players Outlier Analysis monetizing differently than most players? Excel, R, SAS, SPSS, STATA, SWRVE, PredictTM 14
  • 15.
    Forecasting • How much revenue will we bring in next quarter? Time Series Analysis • How many users will we have in the future (near term)? Excel, R, SAS, SPSS, STATA 15
  • 16.
    Predictive Modeling • Who is more likely to monetize? Logistic • Who is more likely to react to in-game Regressions, Decision messaging? Trees, etc. • What will be the lifetime of a user? • How long will it take for users to Survival Analysis monetize? (time to first purchase) • What factors impact the retention of the users? R, SAS, SPSS, STATA, PredictTM, MeasureTM (by GamesAnalytics) 16
  • 17.
    Data Mining • Are there clear “segments” among our Clustering users that could be approached (Segmentation) Analysis differently? • Are there items that sell “together”? Association Analysis • How do users feel about our games? • What are the main topics of conversation Text Mining (Consumer for our Twitter followers? Sentiment Analysis) • Are comments on our Facebook page mostly positive or negative? SAS Enterprise Miner, SPSS Modeler, Weka, PredictTM 17
  • 18.
    Simulation • How long do tasks in our game take to play on average? Monte–Carlo • What happens if we tweak the rules of Simulation the game? Excel, R, Risk Solver, SAS, SPSS, STATA 18
  • 19.
    Optimization • What is the optimal price for the virtual goods? Price Optimization • What is the optimal allocation of resources for supporting the game? Linear Programing R, Risk Solver, Oracle Cristal Ball, SAS 19
  • 20.
  • 21.
    Huge opportunities touse analytics better Analytics is the foundation of business and games deep insight They help you optimize production and marketing decisions They help improve the game and the user experience Adding predictive modeling to in-game analytics lays the foundation for additional optimization of business decisions
  • 22.