Transcript
UPCOMING TALKS
Changes in Mobile Play & its Analytic ImplicationsPhilip Tuck - Head of Business Insight, Gaming Realms
The Trials of Building a Custom Analytics Solution Andres Tallos - CEO & Founder, Fun Fungus
PANEL DISCUSSION
Analytics and Adserving: The FutureExperts from Rebellion, Bee7, InMobi and deltaDNA will look at the issues surroundingthe future of analytics and adserving
Regionalism and play
Session time and style.
● Our mobile customers withdraw 50% less net than desktop on near identical like for like deposits● They reverse 33% more money than desktop● They also Churn 10% slower
Demographics
● Young, evenly split male/female mobile only customers
● Dual desktop and mobile most valuable, majority young and female over all segments
Changes to CRM
● Changes in session time has created multiple experiences throughout the day – constant touch and churn points require real time algorithms to deal with
● Mobile bonusing very different animal to desktop: well targeted small bonuses creating much larger results than seen previously; badly targeted anything leading to high unsubscribe rates and high churn
● Geo targeting algorithms based around consistent daily touch points exceptionally strong
The trials of building a custom analytics solution
Andres TallosCEO Fun Fungus
andres.tallos@funfungus.com9
© Mintsprint Ltd. 2015
Content
● The context● The challenge● The solution● The learnings● The conclusion
© Mintsprint Ltd. 2015
The context
mintsprint
My background
© Mintsprint Ltd. 2015
The challenge
● Goal: Massive multiplayer dogfighting game, able to run on 3G and low-end devices
● Analytics tool had to adapt to:○ All game logic running server-side○ Custom-built technology○ Rapid iterations on game mechanics
© Mintsprint Ltd. 2015
The solution
● Custom built analytics tool
● Web-based, built with Javascript (Bootstrap + Google Visualization API), PHP & MySQL
● Key features:○ Real-time player activity monitor○ High-level & detailed KPI dashboards,
updated in almost real-time○ Auto-generated reports (eg. marketing
campaign results)
© Mintsprint Ltd. 2015
The learnings (pros)
Perceived advantages of building our own analytics:
● It forced us to clearly define what we wanted to learn from our analytics
● We got to know exactly what each KPI means (eg. what counts as one player?)
● Observing real-time player behaviour turned out to be really insightful for game design
● Fully customized to our needs
● No need for 3rd party SDKs
© Mintsprint Ltd. 2015
The learnings (cons)
Perceived disadvantages:
● Significant resource investment
● Hard to get right! (especially scalability & performance)
● Hard to integrate marketing attribution data
● We ended up missing out on features that 3rd party solutions offer as standard (eg. A/B testing, segmentation…)
© Mintsprint Ltd. 2015
The conclusion
Our takeaways:
● Getting it right takes a lot of effort, and it feels like reinventing the wheel. We probably won’t do it again
● We will, however, try to plan our analytics as if we had to build the tool from scratch, to ensure we are hypothesis-driven rather than tool-driven
● Real-time player monitoring (aka. ‘Big Brother’) will forever be #1 in our ‘analytics wishlist’
Stephan Ansari, Business Development
Director, Europe at InMobi
John Rankin, Managing Director at
Bee7
Mark Robinson, CEO at deltaDNA
Matt Jeffery, Head of Digital Games
at Rebellion
Discussion
Analytics & Adserving:The FutureLead by:Chris Wright, CTO at deltaDNA
Panelists
JOIN IN THE CONVERSATION PARTICIPATE IN THE NEXT GIAF
Analytics for Games events@deltadna.com
top related