SlideShare a Scribd company logo
1 of 30
Download to read offline
© King.com Ltd 2015 – Commercially confidential
TV Marketing and big data:TV Marketing and big data:TV Marketing and big data:TV Marketing and big data:
Cat and Dog
or
Thick as Thieves?
Krzysztof Osiewalski
Senior Econometrician / Data Scientist
Marketing Data Science
Krzysztof.Osiewalski@king.com
Cyril Papadacci
Senior Econometrician / Data Scientist
Marketing Data Science
Cyril.Papadacci@king.com
© King.com Ltd 2015 – Commercially confidential
About King
Page 3
© King.com Ltd 2015 – Commercially confidential
We make great games
About King
• More than 185 fun titles played in over 200 countries and regions around the world.
• 364 million average monthly unique users (Q1 2015). • Studios in Stockholm, London,
Barcelona, Bucharest, Malmo,
Berlin, Singapore and Seattle.
• Offices in San Francisco,
Malta, Tokyo, Seoul and
Shanghai.
Page 4
© King.com Ltd 2015 – Commercially confidential
Some stats and facts
About King
1400Employees(approx)
Four global franchises:
Founded in 2003, studios in
Stockholm, London, Barcelona,
Malmo, Bucharest, Berlin,
Singapore and Seattle.
Global leader in
cross-platform casual
games.
Candy Crush Pet Rescue Farm Heroes Bubble Witch
Page 5
© King.com Ltd 2015 – Commercially confidential
Some stats and facts
1.000.000.000.000
Millions of players around
the world.
Approximately 1.6 billion average
daily game plays across our games
in Q1 ‘15
More than 1 trillion levels played!
• Games popular across platforms, and can be played anywhere, anytime on most devices.
• 3 games in the top 10 grossing games on the Apple App Store and on Google Play in the US in Q1 ‘15 .
• Our Saga games allow players to switch platform without losing their progress.
About King
Page 6
© King.com Ltd 2015 – Commercially confidential
The evolution of King
About King
• Founded in 2003
• Originally games were only
available through our site and
portals including AOL and
Yahoo!
Online skill Social Mobile
• Launched first game on
Facebook in Q2 2011
• Launched first game on
mobile H2 2012
Page 7
© King.com Ltd 2015 – Commercially confidential
Some of our games
About King
Page 8
© King.com Ltd 2015 – Commercially confidential
Data at King
Page 9
© King.com Ltd 2015 – Commercially confidential
Big Data
The definition of Big Data [Gartner, 2001]
“Big data” is high-volume, -velocity and -variety information assets that demand
cost-effective, innovative forms of information processing for enhanced insightenhanced insightenhanced insightenhanced insight
and decision makingdecision makingdecision makingdecision making.
The big data V’s:
• VolumeVolumeVolumeVolume quickly evolving (TB in 2012 PB today)
• VarietyVarietyVarietyVariety numbers, text, language, sounds, coordinates, etc…
• VelocityVelocityVelocityVelocity needs fast collection and processing
• VeracityVeracityVeracityVeracity inconsistencies over time, missing data, etc.
• ValueValueValueValue capturing business opportunities, optimizing ...
Page 10
© King.com Ltd 2015 – Commercially confidential
A bit about KingBig Data
Big Data serves business needs
• Infrastructure and analytics should ultimately help answer business
questions
• Provide decision makers with a better real-time understanding of
the business at a very granular level [importance of visualizationvisualizationvisualizationvisualization]
• Help measure the impact of actions in order to have a more data-
driven strategy
• Efficient use of data fosters a more agile approach to driving the
business
Page 11
© King.com Ltd 2015 – Commercially confidential
Our data is… growing
Data at King
Page 12
© King.com Ltd 2015 – Commercially confidential
Our data is… growing
Data at King
Page 13
Exceeds Qlikview
capacity
Exceeds Infobright
capacity
© King.com Ltd 2015 – Commercially confidential
A bit about KingData at King
Our data describes the activity of our player base
• The information that we gather typically looks like this:
• User #123456
• Installed Game G on date t0 from source S
• Played R1, R2,… game rounds on dates t1, t2,…
• Passed level L1 on date t1
• Failed level L2 3 times on date t2
• Sent m1, m2,…. messages on dates t1, t2,…
• Did n1, n2,… transactions on date t1, t2,…
Acquisition
Engagement/Retention
Skills/level difficulty
Virality
Monetization
Page 14
© King.com Ltd 2015 – Commercially confidential
Data value for King
- in marketing
Page 15
© King.com Ltd 2015 – Commercially confidential
A bit about KingWhat do we use our data for?
Several marketing areas benefit from the
rich data that we have
• PerformancePerformancePerformancePerformance MarketingMarketingMarketingMarketing
CLV/RPI modelling
Marketing campaign performance analysis (e.g. digital, TV)
Typical business questions:
• What was the impact on all KPIs of the last digital ad campaign?
• How much are these players likely to spend within next year?
• Up to what threshold can we pay for acquiring this group of users?
• How do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaign????
Page 16
© King.com Ltd 2015 – Commercially confidential
A bit about King
What do we use our data for?
Different channels, different challenges
TVTVTVTV
(and outdoor, press, radio...)
DigitalDigitalDigitalDigital
advertisingadvertisingadvertisingadvertising
Mass audienceAnonymous
User level
Full history of conversion funnel at individual levelFull history of conversion funnel at individual levelFull history of conversion funnel at individual levelFull history of conversion funnel at individual level Measurement only on aggregated metricsMeasurement only on aggregated metricsMeasurement only on aggregated metricsMeasurement only on aggregated metrics
Page 17
© King.com Ltd 2015 – Commercially confidential
User acquisition – econometric ‘top down’ approach
Baseline basket and TV group installs
follow similar patterns in a period
without TV spend
Clear difference between baseline basket
and TV group installs in period of TV
marketing spend
What do we use our data for?
Page 18
© King.com Ltd 2015 – Commercially confidential
364 m
We have more players than the entire US populationWe have more players than the entire US populationWe have more players than the entire US populationWe have more players than the entire US population
320 m
King in numbers
A reminder about King
Page 19
© King.com Ltd 2015 – Commercially confidential
TV has effects on multiple player segments
Other components
User acquisitionUser acquisitionUser acquisitionUser acquisition
Page 20
© King.com Ltd 2015 – Commercially confidential
User acquisitionUser acquisitionUser acquisitionUser acquisition
ReactivationReactivationReactivationReactivation Active usersActive usersActive usersActive users
Page 21
TV has effects on multiple player segments
Other components
© King.com Ltd 2015 – Commercially confidential
Reactivation measurement
Continuously active playersContinuously active playersContinuously active playersContinuously active players
W08 W09W07W06
TV campaignTV campaignTV campaignTV campaign
New installsNew installsNew installsNew installsReturners/ReactivatedReturners/ReactivatedReturners/ReactivatedReturners/Reactivated
W10W05
EconometricsEconometricsEconometricsEconometricsAnonymous uAnonymous uAnonymous uAnonymous user level analysisser level analysisser level analysisser level analysis
Big Data Aggregate time series
Other components
Page 22
© King.com Ltd 2015 – Commercially confidential
A potentially heavy task
Analysis across the board
24 TV countries
7 games
TV activity since beginning 2013
Hundreds of campaigns
Page 23
© King.com Ltd 2015 – Commercially confidential
“Compression” with JSON
UserID week game_rounds
123456789 2014W51 13
123456789 2015W03 9
123456789 2015W04 1
123456789 2015W08 123
123456789 2015W09 444
123456789 2015W10 12
123456789 2015W11 13
UserID game_rounds_blob
123456789 {"2014W51":13,"2015W08":123,"2015W09":444,"2015W04":1,"2015W03":9,"2015W10":12,"2015W11":13}
Need for speed
Raw data model
“Compressed” version – unique UserID record
Page 24
© King.com Ltd 2015 – Commercially confidential
Complexifying can sometimes make things simpler
Need for speed
Unique UserID JSON table
Optimally structured & partitioned
Heavy map stage:
• 9TB
• Useless (in this case)
partitioning
Heavy map stage
Heavy reduce stage
Fixedone-offcost
Light map stage
• Use of partitions
• Filter at map stage
Trivial reduce stage
Smallvariablecost
MMMM
RRRR
MMMM MMMM
RRRR RRRR
MMMM
RRRR
MMMM MMMM
RRRR RRRR
MMMM MMMM MMMM
RRRR
Heavy reduce stage:
• Requires proper
distributing
• Requires scripting in
memory over each user,
given all his data
• Potentially unbalanced due
to multiple filters
Page 25
© King.com Ltd 2015 – Commercially confidential
Complexifying can sometimes make things simpler
Need for speed
Unique UserID JSON table
Optimally structured & partitioned
Heavy map stage:
• 9TB
• Useless (in this case)
partitioning
Heavy map stage
Heavy reduce stage
Fixedone-offcost
Light map stage
• Use of partitions
• Filter at map stage
Trivial reduce stage
Smallvariablecost
MMMM
RRRR
MMMM MMMM
RRRR RRRR
MMMM
RRRR
MMMM MMMM
RRRR RRRR
MMMM MMMM MMMM
RRRR
Heavy reduce stage:
• Requires proper
distributing
• Requires scripting in
memory over each user,
given all his data
• Potentially unbalanced due
to multiple filters
Page 26
100’s x100’s x100’s x100’s x
slowslowslowslow
1x slow1x slow1x slow1x slow
100’s x100’s x100’s x100’s x
fastfastfastfast
© King.com Ltd 2015 – Commercially confidential
Value of JSON for us
Increasing fixed cost, but reducing the variable one
Full analysis of a single campaign reduced to less than 10 minutes
Value of flexibility for the business
Crucial when need for testing different scenarios
Another level of confidence in the achieved ROI
Opening new horizons: halo effect, cross market effect
Page 27
© King.com Ltd 2015 – Commercially confidential Page 28
© King.com Ltd 2015 – Commercially confidential Page 29
Thank you
© King.com Ltd 2015 – Commercially confidential

More Related Content

Viewers also liked

Daum내부 Hadoop 활용 사례 | Devon 2012
Daum내부 Hadoop 활용 사례 | Devon 2012Daum내부 Hadoop 활용 사례 | Devon 2012
Daum내부 Hadoop 활용 사례 | Devon 2012
Daum DNA
 

Viewers also liked (13)

Tv marketing/ Teleshopping
Tv marketing/ TeleshoppingTv marketing/ Teleshopping
Tv marketing/ Teleshopping
 
Maximize Performance of Your Campaigns with Sponsored Updates Partners
Maximize Performance of Your Campaigns with Sponsored Updates PartnersMaximize Performance of Your Campaigns with Sponsored Updates Partners
Maximize Performance of Your Campaigns with Sponsored Updates Partners
 
Daum내부 Hadoop 활용 사례 | Devon 2012
Daum내부 Hadoop 활용 사례 | Devon 2012Daum내부 Hadoop 활용 사례 | Devon 2012
Daum내부 Hadoop 활용 사례 | Devon 2012
 
Retail Marketing Advertising
Retail Marketing AdvertisingRetail Marketing Advertising
Retail Marketing Advertising
 
Future TV 2025
Future TV 2025Future TV 2025
Future TV 2025
 
Node.js를 사용한 Big Data 사례연구
Node.js를 사용한 Big Data 사례연구Node.js를 사용한 Big Data 사례연구
Node.js를 사용한 Big Data 사례연구
 
집단지성 프로그래밍 02-추천시스템 만들기
집단지성 프로그래밍 02-추천시스템 만들기집단지성 프로그래밍 02-추천시스템 만들기
집단지성 프로그래밍 02-추천시스템 만들기
 
컨텐츠 기반 A/B 테스트 구현 사례
컨텐츠 기반 A/B 테스트 구현 사례 컨텐츠 기반 A/B 테스트 구현 사례
컨텐츠 기반 A/B 테스트 구현 사례
 
An example of Aeropostale marketing strategy presentation
 An example of Aeropostale marketing strategy presentation An example of Aeropostale marketing strategy presentation
An example of Aeropostale marketing strategy presentation
 
딥러닝을 11번가 영상 검색에 활용한 경험 공유
딥러닝을 11번가 영상 검색에 활용한 경험 공유딥러닝을 11번가 영상 검색에 활용한 경험 공유
딥러닝을 11번가 영상 검색에 활용한 경험 공유
 
20141223 머하웃(mahout) 협업필터링_추천시스템구현
20141223 머하웃(mahout) 협업필터링_추천시스템구현20141223 머하웃(mahout) 협업필터링_추천시스템구현
20141223 머하웃(mahout) 협업필터링_추천시스템구현
 
TV 2020 - The future of television
TV 2020 - The future of televisionTV 2020 - The future of television
TV 2020 - The future of television
 
[4차]왓챠 알고리즘 분석(151106)
[4차]왓챠 알고리즘 분석(151106)[4차]왓챠 알고리즘 분석(151106)
[4차]왓챠 알고리즘 분석(151106)
 

Similar to TV Marketing and big data: cat and dog or thick as thieves? Krzysztof Osiewalski & Cyril Papadacci, King

Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdfTwilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
ShavitBenitzhak
 
Digital_Signage_Finacial_Network_BIA
Digital_Signage_Finacial_Network_BIADigital_Signage_Finacial_Network_BIA
Digital_Signage_Finacial_Network_BIA
Greg Weaver
 
Igt powerpoint just canvas
Igt powerpoint just canvasIgt powerpoint just canvas
Igt powerpoint just canvas
David Engelmann
 

Similar to TV Marketing and big data: cat and dog or thick as thieves? Krzysztof Osiewalski & Cyril Papadacci, King (20)

Modern Product Data Workflows: How King Crushes New Product Development using...
Modern Product Data Workflows: How King Crushes New Product Development using...Modern Product Data Workflows: How King Crushes New Product Development using...
Modern Product Data Workflows: How King Crushes New Product Development using...
 
Modern Product Data Workflows: How King Crushes New Product Development using...
Modern Product Data Workflows: How King Crushes New Product Development using...Modern Product Data Workflows: How King Crushes New Product Development using...
Modern Product Data Workflows: How King Crushes New Product Development using...
 
Tipping Point for CRE Tech - Brandon Weber, VTS
Tipping Point for CRE Tech - Brandon Weber, VTSTipping Point for CRE Tech - Brandon Weber, VTS
Tipping Point for CRE Tech - Brandon Weber, VTS
 
Mastering Digital Channels with APIs
Mastering Digital Channels with APIsMastering Digital Channels with APIs
Mastering Digital Channels with APIs
 
TIBCO Spotfire: Data Science in the Enterprise
TIBCO Spotfire: Data Science in the EnterpriseTIBCO Spotfire: Data Science in the Enterprise
TIBCO Spotfire: Data Science in the Enterprise
 
Maintaining A Profitable User Acquisition Strategy | Patrick Witham
Maintaining A Profitable User Acquisition Strategy | Patrick WithamMaintaining A Profitable User Acquisition Strategy | Patrick Witham
Maintaining A Profitable User Acquisition Strategy | Patrick Witham
 
WTF is RTB? - WTF Programmatic UK, 11/11/14
WTF is RTB? - WTF Programmatic UK, 11/11/14WTF is RTB? - WTF Programmatic UK, 11/11/14
WTF is RTB? - WTF Programmatic UK, 11/11/14
 
Webinar: Calling Card and How to Start Calling Card Business | ASTPP
Webinar: Calling Card and How to Start Calling Card Business | ASTPPWebinar: Calling Card and How to Start Calling Card Business | ASTPP
Webinar: Calling Card and How to Start Calling Card Business | ASTPP
 
#3 DataBeersBCN - "Big Fun Data" by Xavier Guardiola
#3 DataBeersBCN - "Big Fun Data" by Xavier Guardiola#3 DataBeersBCN - "Big Fun Data" by Xavier Guardiola
#3 DataBeersBCN - "Big Fun Data" by Xavier Guardiola
 
Igt powerpoint final
Igt powerpoint finalIgt powerpoint final
Igt powerpoint final
 
GAMING INDUSTRY ENABLER. WE PROVIDE WHITE LABEL & CUSTOM GAMES PLATFORMS
GAMING INDUSTRY ENABLER. WE PROVIDE WHITE LABEL & CUSTOM GAMES PLATFORMSGAMING INDUSTRY ENABLER. WE PROVIDE WHITE LABEL & CUSTOM GAMES PLATFORMS
GAMING INDUSTRY ENABLER. WE PROVIDE WHITE LABEL & CUSTOM GAMES PLATFORMS
 
Telco2 business models and opportunities briefing may 2013
Telco2   business models and opportunities   briefing may 2013Telco2   business models and opportunities   briefing may 2013
Telco2 business models and opportunities briefing may 2013
 
Going digital services
Going digital servicesGoing digital services
Going digital services
 
Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdfTwilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
Twilio_Segment Pitch - Liraz Rubinstein - Data Guild event.pdf
 
Success of foreign investment attraction by outsource/service companies.
Success of foreign investment attraction by outsource/service companies.Success of foreign investment attraction by outsource/service companies.
Success of foreign investment attraction by outsource/service companies.
 
ABC-8MillionWishes
ABC-8MillionWishesABC-8MillionWishes
ABC-8MillionWishes
 
Digital_Signage_Finacial_Network_BIA
Digital_Signage_Finacial_Network_BIADigital_Signage_Finacial_Network_BIA
Digital_Signage_Finacial_Network_BIA
 
"Valuez", connecting digital businesses
"Valuez", connecting digital businesses"Valuez", connecting digital businesses
"Valuez", connecting digital businesses
 
Igt powerpoint just canvas
Igt powerpoint just canvasIgt powerpoint just canvas
Igt powerpoint just canvas
 
Igt powerpoint just canvas
Igt powerpoint just canvasIgt powerpoint just canvas
Igt powerpoint just canvas
 

More from huguk

More from huguk (20)

Data Wrangling on Hadoop - Olivier De Garrigues, Trifacta
Data Wrangling on Hadoop - Olivier De Garrigues, TrifactaData Wrangling on Hadoop - Olivier De Garrigues, Trifacta
Data Wrangling on Hadoop - Olivier De Garrigues, Trifacta
 
ether.camp - Hackathon & ether.camp intro
ether.camp - Hackathon & ether.camp introether.camp - Hackathon & ether.camp intro
ether.camp - Hackathon & ether.camp intro
 
Google Cloud Dataproc - Easier, faster, more cost-effective Spark and Hadoop
Google Cloud Dataproc - Easier, faster, more cost-effective Spark and HadoopGoogle Cloud Dataproc - Easier, faster, more cost-effective Spark and Hadoop
Google Cloud Dataproc - Easier, faster, more cost-effective Spark and Hadoop
 
Using Big Data techniques to query and store OpenStreetMap data. Stephen Knox...
Using Big Data techniques to query and store OpenStreetMap data. Stephen Knox...Using Big Data techniques to query and store OpenStreetMap data. Stephen Knox...
Using Big Data techniques to query and store OpenStreetMap data. Stephen Knox...
 
Extracting maximum value from data while protecting consumer privacy. Jason ...
Extracting maximum value from data while protecting consumer privacy.  Jason ...Extracting maximum value from data while protecting consumer privacy.  Jason ...
Extracting maximum value from data while protecting consumer privacy. Jason ...
 
Intelligence Augmented vs Artificial Intelligence. Alex Flamant, IBM Watson
Intelligence Augmented vs Artificial Intelligence. Alex Flamant, IBM WatsonIntelligence Augmented vs Artificial Intelligence. Alex Flamant, IBM Watson
Intelligence Augmented vs Artificial Intelligence. Alex Flamant, IBM Watson
 
Streaming Dataflow with Apache Flink
Streaming Dataflow with Apache Flink Streaming Dataflow with Apache Flink
Streaming Dataflow with Apache Flink
 
Lambda architecture on Spark, Kafka for real-time large scale ML
Lambda architecture on Spark, Kafka for real-time large scale MLLambda architecture on Spark, Kafka for real-time large scale ML
Lambda architecture on Spark, Kafka for real-time large scale ML
 
Today’s reality Hadoop with Spark- How to select the best Data Science approa...
Today’s reality Hadoop with Spark- How to select the best Data Science approa...Today’s reality Hadoop with Spark- How to select the best Data Science approa...
Today’s reality Hadoop with Spark- How to select the best Data Science approa...
 
Jonathon Southam: Venture Capital, Funding & Pitching
Jonathon Southam: Venture Capital, Funding & PitchingJonathon Southam: Venture Capital, Funding & Pitching
Jonathon Southam: Venture Capital, Funding & Pitching
 
Signal Media: Real-Time Media & News Monitoring
Signal Media: Real-Time Media & News MonitoringSignal Media: Real-Time Media & News Monitoring
Signal Media: Real-Time Media & News Monitoring
 
Dean Bryen: Scaling The Platform For Your Startup
Dean Bryen: Scaling The Platform For Your StartupDean Bryen: Scaling The Platform For Your Startup
Dean Bryen: Scaling The Platform For Your Startup
 
Peter Karney: Intro to the Digital catapult
Peter Karney: Intro to the Digital catapultPeter Karney: Intro to the Digital catapult
Peter Karney: Intro to the Digital catapult
 
Cytora: Real-Time Political Risk Analysis
Cytora:  Real-Time Political Risk AnalysisCytora:  Real-Time Political Risk Analysis
Cytora: Real-Time Political Risk Analysis
 
Cubitic: Predictive Analytics
Cubitic: Predictive AnalyticsCubitic: Predictive Analytics
Cubitic: Predictive Analytics
 
Bird.i: Earth Observation Data Made Social
Bird.i: Earth Observation Data Made SocialBird.i: Earth Observation Data Made Social
Bird.i: Earth Observation Data Made Social
 
Aiseedo: Real Time Machine Intelligence
Aiseedo: Real Time Machine IntelligenceAiseedo: Real Time Machine Intelligence
Aiseedo: Real Time Machine Intelligence
 
Secrets of Spark's success - Deenar Toraskar, Think Reactive
Secrets of Spark's success - Deenar Toraskar, Think Reactive Secrets of Spark's success - Deenar Toraskar, Think Reactive
Secrets of Spark's success - Deenar Toraskar, Think Reactive
 
Hadoop - Looking to the Future By Arun Murthy
Hadoop - Looking to the Future By Arun MurthyHadoop - Looking to the Future By Arun Murthy
Hadoop - Looking to the Future By Arun Murthy
 
Fast real-time approximations using Spark streaming
Fast real-time approximations using Spark streamingFast real-time approximations using Spark streaming
Fast real-time approximations using Spark streaming
 

Recently uploaded

+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Safe Software
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 

Recently uploaded (20)

Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
 
Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)Powerful Google developer tools for immediate impact! (2023-24 C)
Powerful Google developer tools for immediate impact! (2023-24 C)
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
Apidays New York 2024 - The Good, the Bad and the Governed by David O'Neill, ...
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu SubbuApidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
Apidays Singapore 2024 - Modernizing Securities Finance by Madhu Subbu
 
GenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdfGenAI Risks & Security Meetup 01052024.pdf
GenAI Risks & Security Meetup 01052024.pdf
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 

TV Marketing and big data: cat and dog or thick as thieves? Krzysztof Osiewalski & Cyril Papadacci, King

  • 1.
  • 2. © King.com Ltd 2015 – Commercially confidential TV Marketing and big data:TV Marketing and big data:TV Marketing and big data:TV Marketing and big data: Cat and Dog or Thick as Thieves? Krzysztof Osiewalski Senior Econometrician / Data Scientist Marketing Data Science Krzysztof.Osiewalski@king.com Cyril Papadacci Senior Econometrician / Data Scientist Marketing Data Science Cyril.Papadacci@king.com
  • 3. © King.com Ltd 2015 – Commercially confidential About King Page 3
  • 4. © King.com Ltd 2015 – Commercially confidential We make great games About King • More than 185 fun titles played in over 200 countries and regions around the world. • 364 million average monthly unique users (Q1 2015). • Studios in Stockholm, London, Barcelona, Bucharest, Malmo, Berlin, Singapore and Seattle. • Offices in San Francisco, Malta, Tokyo, Seoul and Shanghai. Page 4
  • 5. © King.com Ltd 2015 – Commercially confidential Some stats and facts About King 1400Employees(approx) Four global franchises: Founded in 2003, studios in Stockholm, London, Barcelona, Malmo, Bucharest, Berlin, Singapore and Seattle. Global leader in cross-platform casual games. Candy Crush Pet Rescue Farm Heroes Bubble Witch Page 5
  • 6. © King.com Ltd 2015 – Commercially confidential Some stats and facts 1.000.000.000.000 Millions of players around the world. Approximately 1.6 billion average daily game plays across our games in Q1 ‘15 More than 1 trillion levels played! • Games popular across platforms, and can be played anywhere, anytime on most devices. • 3 games in the top 10 grossing games on the Apple App Store and on Google Play in the US in Q1 ‘15 . • Our Saga games allow players to switch platform without losing their progress. About King Page 6
  • 7. © King.com Ltd 2015 – Commercially confidential The evolution of King About King • Founded in 2003 • Originally games were only available through our site and portals including AOL and Yahoo! Online skill Social Mobile • Launched first game on Facebook in Q2 2011 • Launched first game on mobile H2 2012 Page 7
  • 8. © King.com Ltd 2015 – Commercially confidential Some of our games About King Page 8
  • 9. © King.com Ltd 2015 – Commercially confidential Data at King Page 9
  • 10. © King.com Ltd 2015 – Commercially confidential Big Data The definition of Big Data [Gartner, 2001] “Big data” is high-volume, -velocity and -variety information assets that demand cost-effective, innovative forms of information processing for enhanced insightenhanced insightenhanced insightenhanced insight and decision makingdecision makingdecision makingdecision making. The big data V’s: • VolumeVolumeVolumeVolume quickly evolving (TB in 2012 PB today) • VarietyVarietyVarietyVariety numbers, text, language, sounds, coordinates, etc… • VelocityVelocityVelocityVelocity needs fast collection and processing • VeracityVeracityVeracityVeracity inconsistencies over time, missing data, etc. • ValueValueValueValue capturing business opportunities, optimizing ... Page 10
  • 11. © King.com Ltd 2015 – Commercially confidential A bit about KingBig Data Big Data serves business needs • Infrastructure and analytics should ultimately help answer business questions • Provide decision makers with a better real-time understanding of the business at a very granular level [importance of visualizationvisualizationvisualizationvisualization] • Help measure the impact of actions in order to have a more data- driven strategy • Efficient use of data fosters a more agile approach to driving the business Page 11
  • 12. © King.com Ltd 2015 – Commercially confidential Our data is… growing Data at King Page 12
  • 13. © King.com Ltd 2015 – Commercially confidential Our data is… growing Data at King Page 13 Exceeds Qlikview capacity Exceeds Infobright capacity
  • 14. © King.com Ltd 2015 – Commercially confidential A bit about KingData at King Our data describes the activity of our player base • The information that we gather typically looks like this: • User #123456 • Installed Game G on date t0 from source S • Played R1, R2,… game rounds on dates t1, t2,… • Passed level L1 on date t1 • Failed level L2 3 times on date t2 • Sent m1, m2,…. messages on dates t1, t2,… • Did n1, n2,… transactions on date t1, t2,… Acquisition Engagement/Retention Skills/level difficulty Virality Monetization Page 14
  • 15. © King.com Ltd 2015 – Commercially confidential Data value for King - in marketing Page 15
  • 16. © King.com Ltd 2015 – Commercially confidential A bit about KingWhat do we use our data for? Several marketing areas benefit from the rich data that we have • PerformancePerformancePerformancePerformance MarketingMarketingMarketingMarketing CLV/RPI modelling Marketing campaign performance analysis (e.g. digital, TV) Typical business questions: • What was the impact on all KPIs of the last digital ad campaign? • How much are these players likely to spend within next year? • Up to what threshold can we pay for acquiring this group of users? • How do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaignHow do we measure the ROI of a TV campaign???? Page 16
  • 17. © King.com Ltd 2015 – Commercially confidential A bit about King What do we use our data for? Different channels, different challenges TVTVTVTV (and outdoor, press, radio...) DigitalDigitalDigitalDigital advertisingadvertisingadvertisingadvertising Mass audienceAnonymous User level Full history of conversion funnel at individual levelFull history of conversion funnel at individual levelFull history of conversion funnel at individual levelFull history of conversion funnel at individual level Measurement only on aggregated metricsMeasurement only on aggregated metricsMeasurement only on aggregated metricsMeasurement only on aggregated metrics Page 17
  • 18. © King.com Ltd 2015 – Commercially confidential User acquisition – econometric ‘top down’ approach Baseline basket and TV group installs follow similar patterns in a period without TV spend Clear difference between baseline basket and TV group installs in period of TV marketing spend What do we use our data for? Page 18
  • 19. © King.com Ltd 2015 – Commercially confidential 364 m We have more players than the entire US populationWe have more players than the entire US populationWe have more players than the entire US populationWe have more players than the entire US population 320 m King in numbers A reminder about King Page 19
  • 20. © King.com Ltd 2015 – Commercially confidential TV has effects on multiple player segments Other components User acquisitionUser acquisitionUser acquisitionUser acquisition Page 20
  • 21. © King.com Ltd 2015 – Commercially confidential User acquisitionUser acquisitionUser acquisitionUser acquisition ReactivationReactivationReactivationReactivation Active usersActive usersActive usersActive users Page 21 TV has effects on multiple player segments Other components
  • 22. © King.com Ltd 2015 – Commercially confidential Reactivation measurement Continuously active playersContinuously active playersContinuously active playersContinuously active players W08 W09W07W06 TV campaignTV campaignTV campaignTV campaign New installsNew installsNew installsNew installsReturners/ReactivatedReturners/ReactivatedReturners/ReactivatedReturners/Reactivated W10W05 EconometricsEconometricsEconometricsEconometricsAnonymous uAnonymous uAnonymous uAnonymous user level analysisser level analysisser level analysisser level analysis Big Data Aggregate time series Other components Page 22
  • 23. © King.com Ltd 2015 – Commercially confidential A potentially heavy task Analysis across the board 24 TV countries 7 games TV activity since beginning 2013 Hundreds of campaigns Page 23
  • 24. © King.com Ltd 2015 – Commercially confidential “Compression” with JSON UserID week game_rounds 123456789 2014W51 13 123456789 2015W03 9 123456789 2015W04 1 123456789 2015W08 123 123456789 2015W09 444 123456789 2015W10 12 123456789 2015W11 13 UserID game_rounds_blob 123456789 {"2014W51":13,"2015W08":123,"2015W09":444,"2015W04":1,"2015W03":9,"2015W10":12,"2015W11":13} Need for speed Raw data model “Compressed” version – unique UserID record Page 24
  • 25. © King.com Ltd 2015 – Commercially confidential Complexifying can sometimes make things simpler Need for speed Unique UserID JSON table Optimally structured & partitioned Heavy map stage: • 9TB • Useless (in this case) partitioning Heavy map stage Heavy reduce stage Fixedone-offcost Light map stage • Use of partitions • Filter at map stage Trivial reduce stage Smallvariablecost MMMM RRRR MMMM MMMM RRRR RRRR MMMM RRRR MMMM MMMM RRRR RRRR MMMM MMMM MMMM RRRR Heavy reduce stage: • Requires proper distributing • Requires scripting in memory over each user, given all his data • Potentially unbalanced due to multiple filters Page 25
  • 26. © King.com Ltd 2015 – Commercially confidential Complexifying can sometimes make things simpler Need for speed Unique UserID JSON table Optimally structured & partitioned Heavy map stage: • 9TB • Useless (in this case) partitioning Heavy map stage Heavy reduce stage Fixedone-offcost Light map stage • Use of partitions • Filter at map stage Trivial reduce stage Smallvariablecost MMMM RRRR MMMM MMMM RRRR RRRR MMMM RRRR MMMM MMMM RRRR RRRR MMMM MMMM MMMM RRRR Heavy reduce stage: • Requires proper distributing • Requires scripting in memory over each user, given all his data • Potentially unbalanced due to multiple filters Page 26 100’s x100’s x100’s x100’s x slowslowslowslow 1x slow1x slow1x slow1x slow 100’s x100’s x100’s x100’s x fastfastfastfast
  • 27. © King.com Ltd 2015 – Commercially confidential Value of JSON for us Increasing fixed cost, but reducing the variable one Full analysis of a single campaign reduced to less than 10 minutes Value of flexibility for the business Crucial when need for testing different scenarios Another level of confidence in the achieved ROI Opening new horizons: halo effect, cross market effect Page 27
  • 28. © King.com Ltd 2015 – Commercially confidential Page 28
  • 29. © King.com Ltd 2015 – Commercially confidential Page 29
  • 30. Thank you © King.com Ltd 2015 – Commercially confidential