SlideShare uma empresa Scribd logo
Optimising user acquisition through LTV prediction
Róbert Magyar
Warsaw 14 -15 October 2019
Warsaw | 14-15 October 2019
Róbert Magyar
Data Science Lead
robert.magyar@superscale.com
“We’re forming growth partnerships with world’s top developers
to scale their games to maximum potential.”
Warsaw | 14-15 October 2019
● 70+ World Class Experts In-House
○ (Self-)Publishing Infrastructure
i. Business Intelligence
ii. Analytics & Data Science
○ Monetization
i. LiveOps Optimization
ii. Game & Monetization Design
○ User Acquisition
i. Creatives
ii. UA Campaign Management
iii. ASO
● Founded in 2016
● Bratislava, London, Berlin, Helsinki &
Prague offices
First Light Games
Our partners
Games
Warsaw | 14-15 October 2019
UA Optimization
Motivation
Warsaw | 14-15 October 2019
How can we improve User Acquisition?
Huge topic - lots of angles
● Lookalike redesign
● Different UA channels
● New creatives
● Different spending strategy
● Passively through game related efforts e.g. Economy Balancing
...
● Optimization of UA investments through LTV prediction - focus of this talk
Warsaw | 14-15 October 2019
LTV predictions as an actionable insight for UA team
Comes early in the
campaign lifecycle
Helps to identify
opportunities
Accuracy is
stable over time
Supports decision
making
Defines strategy of
UA investment
optimization
Warsaw | 14-15 October 2019
Waiting for data is not fun
Usual approach for building LTV prediction model:
● wait several weeks or months to gather data about the campaigns
● create LTV model
● use the model to estimate UA performance
.. but market changes over time, new competitors arise etc
Warsaw | 14-15 October 2019
The Question
How can we improve UA efforts with LTV predictions
● early in campaigns lifecycle?
● without need of waiting several months to understand performance trend?
=> by understanding monetization of the game and building predictions from
bottom-up
How to actually do it?
● utilizing cloud machine learning
Warsaw | 14-15 October 2019
Some games have steep early monetization which creates an advantage in
modeling.
Game monetization defines data needed for predictions
= Game monetization => defines how many days of data we
need and helps to understand the payback of UA
50% of revenue in first couple of days =
enough data to work with
Warsaw | 14-15 October 2019
Variation in campaign cohorts performance = accuracy loss
=> Accuracy of models is impacted by daily market changes,
this can be seen on significant movements in campaign’s cohorts performances
over time
Warsaw | 14-15 October 2019
Building accurate predictions
Focusing on daily cohorts => cohort’s distribution of
payers does not change over time which improves
accuracy while building up the predictions:
1. Predictions based on cohort level not on campaign
level
2. Aggregate predictions for each campaign
Idea is to go hierarchically to the lowest level possible
(cohort/ad/adset etc) while:
- Keeping enough players in cohorts
= Daily cohorts => logarithmic growth can be leveraged => no need to wait
several months for data
Warsaw | 14-15 October 2019
Keeping enough players in cohorts
Number of players needed is based on:
- Conversion and number of payers
Unusual growth of revenue in the cohort can be the clue of not having enough players to
work with.
Looks like step function - reason could be:
1. Game relying mainly on Liveops offers (not so good design)
2.Conversion for the cohort is not high enough
Warsaw | 14-15 October 2019
UA Optimization
LTV prediction in Action
Warsaw | 14-15 October 2019
Information we are able to use:
- Monetary : Conversion, ARPU, ARPPU, # of purchases, revenue, payers
distribution, probability of 2nd purchase etc
- Absolute value
- Relative change/growth over first couple of days
- per cohort or campaign and aggregated
- Behavioral (Engagement metrics) : playtime, retention, first day
drop-off, percentage of one-shots etc
Leveraging all the data
Warsaw | 14-15 October 2019
Business optimization goal
SuperScale Analytics + UA Stack
Data Sources
(downloaded daily)
Game Data
(Appsflyer, Google Firebase, ...)
Attribution Platform
(Appsflyer, ...)
UA Channel Data
(Facebook, Google UAC, Ad-n...)
Store Revenue Data
(G Play, iTunes)
SuperScale Analytics + UA Stack
Data Sources
(downloaded daily)
Data Storage Materialized Views
Google BigQuery
UA Campaign
Evaluation MVs
Game Data
(Appsflyer, Google Firebase, ...)
Attribution Platform
(Appsflyer, ...)
UA Channel Data
(Facebook, Google UAC, Ad-n...)
Store Revenue Data
(G Play, iTunes)
Google Cloud Platform Machine Learning
(stochastic algorithms, random forest,
k-means clustering, …)
SBDW ETL Processing
(User States,
Materialized Views)
SuperScale Analytics + UA Stack
Data Sources
(downloaded daily)
Data Storage Materialized Views Application
Google BigQuery
Data Validator
(store revenue vs app revenue,
AF revenue vs store)
ROI/ROAS Spreadsheets
UA Campaign
Evaluation MVs
Game Data
(Appsflyer, Google Firebase, ...)
Attribution Platform
(Appsflyer, ...)
UA Channel Data
(Facebook, Google UAC, Ad-n...)
Store Revenue Data
(G Play, iTunes)
FB Campaign Optimization
Data Visualisation
Campaign
LTV/ROI prediction
Google Cloud Platform Machine Learning
(stochastic algorithms, random forest,
k-means clustering, …)
SBDW ETL Processing
(User States,
Materialized Views)
Warsaw | 14-15 October 2019
Building predictions from a bottom-up
Daily cohort 1
Daily cohort 2
Daily cohort 3
..
UA DATA PREDICTIONS
for each cohort
AGGREGATION of cohorts predictions on a
higher level
..
Weekly
Monthly
Quarterly
Campaign
Overall
CPI
LTV
Conversion,
ARPU, ARPPU,
# purchases,
revenue,
payers
distribution
(absolute values,
relative values)
Campaign success
prediction
Weighting of data
points based on
recency
Warsaw | 14-15 October 2019
Cloud pipeline takes care of automatic scaling of projects.
Cloud prediction pipeline
Data warehouse Batch preprocessing Machine learning
BigQuery
Creating models for:
Daily, weekly, monthly cohorts
and campaigns
Additional analysis:
Geo analysis
Break-even analysis
Business Intelligence
Data Studio
Scheduler
Cloud Functions
Models storage
Google Cloud
Storage
Google
Dataflow
ML Engine on
AI platform
Raw GAME Data +
Campaign data +
LKLs data
Periscope
Warsaw | 14-15 October 2019
Supporting decision making - examples
Weekly
cohort
estimate
d LTV
IAP D7 ROAS performance per country Weekly cohort LTV performance estimation Campaigns breakeven analysis
Top country to target for LKLs Which strategy was the best When can we expect payback
Warsaw | 14-15 October 2019
- LTV/ROAS predictions can help to identify opportunities but timing is the
factor that makes or breaks any modelling
- Aggregating weekly and monthly predictions from daily ones can improve
accuracy significantly
- Weighting data points based on recency can improve accuracy of
estimations
Takeaways
Warsaw | 14-15 October 2019
Thank you for attention.
Questions?

Mais conteúdo relacionado

Mais procurados

How much data is needed to calculate LTV?
How much data is needed to calculate LTV?How much data is needed to calculate LTV?
How much data is needed to calculate LTV?
Eric Seufert
 
Intro to liveops
Intro to liveopsIntro to liveops
Intro to liveops
Crystin Cox
 
Game monetization: Overview of monetization methods for free-to-play games
Game monetization: Overview of monetization methods for free-to-play gamesGame monetization: Overview of monetization methods for free-to-play games
Game monetization: Overview of monetization methods for free-to-play games
Andrew Dotsenko
 
How to start working with LTV measurement in mobile gaming? How to move to ad...
How to start working with LTV measurement in mobile gaming? How to move to ad...How to start working with LTV measurement in mobile gaming? How to move to ad...
How to start working with LTV measurement in mobile gaming? How to move to ad...
GameCamp
 
Live ops in mobile gaming - how to do it right?
 Live ops in mobile gaming - how to do it right? Live ops in mobile gaming - how to do it right?
Live ops in mobile gaming - how to do it right?
GameCamp
 
Developing an effective LTV model at the soft launch and keeping it valid fur...
Developing an effective LTV model at the soft launch and keeping it valid fur...Developing an effective LTV model at the soft launch and keeping it valid fur...
Developing an effective LTV model at the soft launch and keeping it valid fur...
GameCamp
 
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
Eric Seufert
 
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
Mozza
 
Growth Lessons from Early Snapchat
Growth Lessons from Early SnapchatGrowth Lessons from Early Snapchat
Growth Lessons from Early Snapchat
Mozza
 
A/B testing in ad monetisation
A/B testing in ad monetisationA/B testing in ad monetisation
A/B testing in ad monetisation
GameCamp
 
Startup Investing 101 for Angels with FoundersHK | Edith Yeung
Startup Investing 101 for Angels with FoundersHK | Edith YeungStartup Investing 101 for Angels with FoundersHK | Edith Yeung
Startup Investing 101 for Angels with FoundersHK | Edith Yeung
Edith Yeung
 
Hyper-casual in a Hypercompetitive Market - Voodoo
Hyper-casual in a Hypercompetitive Market - VoodooHyper-casual in a Hypercompetitive Market - Voodoo
Hyper-casual in a Hypercompetitive Market - Voodoo
GameAnalytics
 
Product Management For Free-to-Play Games w/ AC&A's Senior PM
Product Management For Free-to-Play Games w/ AC&A's Senior PMProduct Management For Free-to-Play Games w/ AC&A's Senior PM
Product Management For Free-to-Play Games w/ AC&A's Senior PM
Product School
 
GDC Talk: Lifetime Value: The long tail of Mid-Core games
GDC Talk: Lifetime Value: The long tail of Mid-Core gamesGDC Talk: Lifetime Value: The long tail of Mid-Core games
GDC Talk: Lifetime Value: The long tail of Mid-Core games
Tamara (Tammy) Levy
 
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
Eric Seufert
 
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
Jessica Tams
 
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
Jessica Tams
 
Personalisation as the key to optimising your game's revenue & LTV.
Personalisation as the key to optimising your game's revenue & LTV.Personalisation as the key to optimising your game's revenue & LTV.
Personalisation as the key to optimising your game's revenue & LTV.
GameCamp
 
Game design - Retention
Game design - RetentionGame design - Retention
Game design - Retention
Erez Yerushalmi
 
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
Jessica Tams
 

Mais procurados (20)

How much data is needed to calculate LTV?
How much data is needed to calculate LTV?How much data is needed to calculate LTV?
How much data is needed to calculate LTV?
 
Intro to liveops
Intro to liveopsIntro to liveops
Intro to liveops
 
Game monetization: Overview of monetization methods for free-to-play games
Game monetization: Overview of monetization methods for free-to-play gamesGame monetization: Overview of monetization methods for free-to-play games
Game monetization: Overview of monetization methods for free-to-play games
 
How to start working with LTV measurement in mobile gaming? How to move to ad...
How to start working with LTV measurement in mobile gaming? How to move to ad...How to start working with LTV measurement in mobile gaming? How to move to ad...
How to start working with LTV measurement in mobile gaming? How to move to ad...
 
Live ops in mobile gaming - how to do it right?
 Live ops in mobile gaming - how to do it right? Live ops in mobile gaming - how to do it right?
Live ops in mobile gaming - how to do it right?
 
Developing an effective LTV model at the soft launch and keeping it valid fur...
Developing an effective LTV model at the soft launch and keeping it valid fur...Developing an effective LTV model at the soft launch and keeping it valid fur...
Developing an effective LTV model at the soft launch and keeping it valid fur...
 
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
Eric Seufert, GDC 2014: Profitably launching Jelly Splash to #1, a marketing ...
 
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
Organic Acquisition: How To Acquire A Million Users With Zero Marketing (mobi...
 
Growth Lessons from Early Snapchat
Growth Lessons from Early SnapchatGrowth Lessons from Early Snapchat
Growth Lessons from Early Snapchat
 
A/B testing in ad monetisation
A/B testing in ad monetisationA/B testing in ad monetisation
A/B testing in ad monetisation
 
Startup Investing 101 for Angels with FoundersHK | Edith Yeung
Startup Investing 101 for Angels with FoundersHK | Edith YeungStartup Investing 101 for Angels with FoundersHK | Edith Yeung
Startup Investing 101 for Angels with FoundersHK | Edith Yeung
 
Hyper-casual in a Hypercompetitive Market - Voodoo
Hyper-casual in a Hypercompetitive Market - VoodooHyper-casual in a Hypercompetitive Market - Voodoo
Hyper-casual in a Hypercompetitive Market - Voodoo
 
Product Management For Free-to-Play Games w/ AC&A's Senior PM
Product Management For Free-to-Play Games w/ AC&A's Senior PMProduct Management For Free-to-Play Games w/ AC&A's Senior PM
Product Management For Free-to-Play Games w/ AC&A's Senior PM
 
GDC Talk: Lifetime Value: The long tail of Mid-Core games
GDC Talk: Lifetime Value: The long tail of Mid-Core gamesGDC Talk: Lifetime Value: The long tail of Mid-Core games
GDC Talk: Lifetime Value: The long tail of Mid-Core games
 
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
Eric Seufert - GDC San Francisco 2016 Presentation - Taking Flight Again: The...
 
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
Case Study: Introducing LiveOps and F2P to Traditional Game Mechanics in Roll...
 
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
3 Useful Case Studies from Hyper-Casual Games and Ad-Funded Apps | Yaniv Nizan
 
Personalisation as the key to optimising your game's revenue & LTV.
Personalisation as the key to optimising your game's revenue & LTV.Personalisation as the key to optimising your game's revenue & LTV.
Personalisation as the key to optimising your game's revenue & LTV.
 
Game design - Retention
Game design - RetentionGame design - Retention
Game design - Retention
 
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
Improving LTV with Personalized Live Ops Offers: Hill Climb Racing 2 Case Stu...
 

Semelhante a Optimising user acquisition through LTV prediction

Mobile marketing 101 - January 2016
Mobile marketing 101 - January 2016Mobile marketing 101 - January 2016
Mobile marketing 101 - January 2016
Ratko Bozovic
 
Mobile marketing from analysis to launching a project
Mobile marketing from analysis to launching a projectMobile marketing from analysis to launching a project
Mobile marketing from analysis to launching a project
Heads&Hands
 
Market Research for Game Developers
Market Research for Game DevelopersMarket Research for Game Developers
Market Research for Game Developers
Benjamin Nolan
 
iSPORTiSTiCS investment opportunity 2020
iSPORTiSTiCS investment opportunity 2020iSPORTiSTiCS investment opportunity 2020
iSPORTiSTiCS investment opportunity 2020
Vinicius Gholmie
 
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
Databricks
 
Adways Session 4.App Promotion Trend in Japan
Adways Session 4.App Promotion Trend in JapanAdways Session 4.App Promotion Trend in Japan
Adways Session 4.App Promotion Trend in Japan
ADWAYS KOREA
 
App promotion trend in Japan
App promotion trend in JapanApp promotion trend in Japan
App promotion trend in Japan
Daisuke Maeda
 
The Developer Tour - Hanoi
The Developer Tour - HanoiThe Developer Tour - Hanoi
The Developer Tour - Hanoi
ironSource
 
Shape.io - PPC Budget Management Suite
Shape.io - PPC Budget Management SuiteShape.io - PPC Budget Management Suite
Shape.io - PPC Budget Management Suite
Shape Integrated Software
 
Drive Customer Engagement with Gamification
Drive Customer Engagement with GamificationDrive Customer Engagement with Gamification
Drive Customer Engagement with Gamification
Badgeville, Inc.
 
Konstantin Dieterle, AppLift
Konstantin Dieterle, AppLiftKonstantin Dieterle, AppLift
Konstantin Dieterle, AppLift
White Nights Conference
 
Epam BI - Near Realtime Marketing Support System
Epam BI - Near Realtime Marketing Support SystemEpam BI - Near Realtime Marketing Support System
Epam BI - Near Realtime Marketing Support System
Dmitry Tolpeko
 
Parking frenzy
Parking frenzyParking frenzy
Parking frenzy
Tapas Dan
 
Digital Marketing ROI Workshop
Digital Marketing ROI WorkshopDigital Marketing ROI Workshop
Digital Marketing ROI Workshop
Arman Rousta
 
3*3 Developer Tour
3*3 Developer Tour3*3 Developer Tour
3*3 Developer Tour
ironSource
 
AppAgent ROY mobile marketing analytics
AppAgent ROY mobile marketing analyticsAppAgent ROY mobile marketing analytics
AppAgent ROY mobile marketing analytics
AppAgent / Strategic & Creative Mobile Marketing Agency
 
Mime_Analytics_Mii_brochure
Mime_Analytics_Mii_brochureMime_Analytics_Mii_brochure
Mime_Analytics_Mii_brochure
Monica Pearce
 
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
Octopus Events
 
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
AIST
 
Mime_Analytics_Mii_brochure_NZ
Mime_Analytics_Mii_brochure_NZMime_Analytics_Mii_brochure_NZ
Mime_Analytics_Mii_brochure_NZ
Simon Gentry
 

Semelhante a Optimising user acquisition through LTV prediction (20)

Mobile marketing 101 - January 2016
Mobile marketing 101 - January 2016Mobile marketing 101 - January 2016
Mobile marketing 101 - January 2016
 
Mobile marketing from analysis to launching a project
Mobile marketing from analysis to launching a projectMobile marketing from analysis to launching a project
Mobile marketing from analysis to launching a project
 
Market Research for Game Developers
Market Research for Game DevelopersMarket Research for Game Developers
Market Research for Game Developers
 
iSPORTiSTiCS investment opportunity 2020
iSPORTiSTiCS investment opportunity 2020iSPORTiSTiCS investment opportunity 2020
iSPORTiSTiCS investment opportunity 2020
 
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
Predicting Banking Customer Needs with an Agile Approach to Analytics in the ...
 
Adways Session 4.App Promotion Trend in Japan
Adways Session 4.App Promotion Trend in JapanAdways Session 4.App Promotion Trend in Japan
Adways Session 4.App Promotion Trend in Japan
 
App promotion trend in Japan
App promotion trend in JapanApp promotion trend in Japan
App promotion trend in Japan
 
The Developer Tour - Hanoi
The Developer Tour - HanoiThe Developer Tour - Hanoi
The Developer Tour - Hanoi
 
Shape.io - PPC Budget Management Suite
Shape.io - PPC Budget Management SuiteShape.io - PPC Budget Management Suite
Shape.io - PPC Budget Management Suite
 
Drive Customer Engagement with Gamification
Drive Customer Engagement with GamificationDrive Customer Engagement with Gamification
Drive Customer Engagement with Gamification
 
Konstantin Dieterle, AppLift
Konstantin Dieterle, AppLiftKonstantin Dieterle, AppLift
Konstantin Dieterle, AppLift
 
Epam BI - Near Realtime Marketing Support System
Epam BI - Near Realtime Marketing Support SystemEpam BI - Near Realtime Marketing Support System
Epam BI - Near Realtime Marketing Support System
 
Parking frenzy
Parking frenzyParking frenzy
Parking frenzy
 
Digital Marketing ROI Workshop
Digital Marketing ROI WorkshopDigital Marketing ROI Workshop
Digital Marketing ROI Workshop
 
3*3 Developer Tour
3*3 Developer Tour3*3 Developer Tour
3*3 Developer Tour
 
AppAgent ROY mobile marketing analytics
AppAgent ROY mobile marketing analyticsAppAgent ROY mobile marketing analytics
AppAgent ROY mobile marketing analytics
 
Mime_Analytics_Mii_brochure
Mime_Analytics_Mii_brochureMime_Analytics_Mii_brochure
Mime_Analytics_Mii_brochure
 
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
Andrey Sukhovoy. Marketing Mix Modeling: Define the Optimal Budget Allocation...
 
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
Евгений Цымбалов, Webgames - Методы машинного обучения для задач игровой анал...
 
Mime_Analytics_Mii_brochure_NZ
Mime_Analytics_Mii_brochure_NZMime_Analytics_Mii_brochure_NZ
Mime_Analytics_Mii_brochure_NZ
 

Mais de GameCamp

Practical approach to creative testing and creative optimisation at Google UA...
Practical approach to creative testing and creative optimisation at Google UA...Practical approach to creative testing and creative optimisation at Google UA...
Practical approach to creative testing and creative optimisation at Google UA...
GameCamp
 
Creativity and science behind creative testing. creative testing framework an...
Creativity and science behind creative testing. creative testing framework an...Creativity and science behind creative testing. creative testing framework an...
Creativity and science behind creative testing. creative testing framework an...
GameCamp
 
How to boost your ASO with data analytics?
How to boost your ASO with data analytics?How to boost your ASO with data analytics?
How to boost your ASO with data analytics?
GameCamp
 
8 Types of mobile game and app creatives you should try
8 Types of mobile game and app creatives you should try8 Types of mobile game and app creatives you should try
8 Types of mobile game and app creatives you should try
GameCamp
 
Building the BI system and analytics capabilities at the company based on Rea...
Building the BI system and analytics capabilities at the company based on Rea...Building the BI system and analytics capabilities at the company based on Rea...
Building the BI system and analytics capabilities at the company based on Rea...
GameCamp
 
Ad-hoc tasks in the Data Scientist team based on Outfit7 example
Ad-hoc tasks in the Data Scientist team based on Outfit7 exampleAd-hoc tasks in the Data Scientist team based on Outfit7 example
Ad-hoc tasks in the Data Scientist team based on Outfit7 example
GameCamp
 
Driving profitability of Google App Campaigns in scale. What is easy, what is...
Driving profitability of Google App Campaigns in scale. What is easy, what is...Driving profitability of Google App Campaigns in scale. What is easy, what is...
Driving profitability of Google App Campaigns in scale. What is easy, what is...
GameCamp
 
Scaling UA activity - the challenges of growth
 Scaling UA activity - the challenges of growth Scaling UA activity - the challenges of growth
Scaling UA activity - the challenges of growth
GameCamp
 
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
GameCamp
 
Growth trends in mobile gaming based on the data
Growth trends in mobile gaming based on the dataGrowth trends in mobile gaming based on the data
Growth trends in mobile gaming based on the data
GameCamp
 
Facebook's and Social Creative Best Practice that worked for Huuuge
Facebook's and Social Creative Best Practice that worked for HuuugeFacebook's and Social Creative Best Practice that worked for Huuuge
Facebook's and Social Creative Best Practice that worked for Huuuge
GameCamp
 
Using Data Science for Behavioural Game Design
Using Data Science for Behavioural Game DesignUsing Data Science for Behavioural Game Design
Using Data Science for Behavioural Game Design
GameCamp
 
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
GameCamp
 
In-app offers monetization from basics to advanced technics based on Crazy Pa...
In-app offers monetization from basics to advanced technics based on Crazy Pa...In-app offers monetization from basics to advanced technics based on Crazy Pa...
In-app offers monetization from basics to advanced technics based on Crazy Pa...
GameCamp
 
Building cost-effective mobile product & marketing app analytics based on GCP...
Building cost-effective mobile product & marketing app analytics based on GCP...Building cost-effective mobile product & marketing app analytics based on GCP...
Building cost-effective mobile product & marketing app analytics based on GCP...
GameCamp
 
Enabling the growth of the game through performance marketing in 2020
Enabling the growth of the game through performance marketing in 2020Enabling the growth of the game through performance marketing in 2020
Enabling the growth of the game through performance marketing in 2020
GameCamp
 
Growth markets in mobile gaming
 Growth markets in mobile gaming Growth markets in mobile gaming
Growth markets in mobile gaming
GameCamp
 
LTV measurement and multi-touch attribution
LTV measurement and multi-touch attributionLTV measurement and multi-touch attribution
LTV measurement and multi-touch attribution
GameCamp
 
Big Problem, BigQuery: User Feature Engineering in Event-driven Analytics
Big Problem, BigQuery: User Feature Engineering in Event-driven AnalyticsBig Problem, BigQuery: User Feature Engineering in Event-driven Analytics
Big Problem, BigQuery: User Feature Engineering in Event-driven Analytics
GameCamp
 
Not so trivial lessons learned from launching mid-core games.
Not so trivial lessons learned from launching mid-core games.Not so trivial lessons learned from launching mid-core games.
Not so trivial lessons learned from launching mid-core games.
GameCamp
 

Mais de GameCamp (20)

Practical approach to creative testing and creative optimisation at Google UA...
Practical approach to creative testing and creative optimisation at Google UA...Practical approach to creative testing and creative optimisation at Google UA...
Practical approach to creative testing and creative optimisation at Google UA...
 
Creativity and science behind creative testing. creative testing framework an...
Creativity and science behind creative testing. creative testing framework an...Creativity and science behind creative testing. creative testing framework an...
Creativity and science behind creative testing. creative testing framework an...
 
How to boost your ASO with data analytics?
How to boost your ASO with data analytics?How to boost your ASO with data analytics?
How to boost your ASO with data analytics?
 
8 Types of mobile game and app creatives you should try
8 Types of mobile game and app creatives you should try8 Types of mobile game and app creatives you should try
8 Types of mobile game and app creatives you should try
 
Building the BI system and analytics capabilities at the company based on Rea...
Building the BI system and analytics capabilities at the company based on Rea...Building the BI system and analytics capabilities at the company based on Rea...
Building the BI system and analytics capabilities at the company based on Rea...
 
Ad-hoc tasks in the Data Scientist team based on Outfit7 example
Ad-hoc tasks in the Data Scientist team based on Outfit7 exampleAd-hoc tasks in the Data Scientist team based on Outfit7 example
Ad-hoc tasks in the Data Scientist team based on Outfit7 example
 
Driving profitability of Google App Campaigns in scale. What is easy, what is...
Driving profitability of Google App Campaigns in scale. What is easy, what is...Driving profitability of Google App Campaigns in scale. What is easy, what is...
Driving profitability of Google App Campaigns in scale. What is easy, what is...
 
Scaling UA activity - the challenges of growth
 Scaling UA activity - the challenges of growth Scaling UA activity - the challenges of growth
Scaling UA activity - the challenges of growth
 
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
Growth Strategies: Lower Your CPI by Implementing ASO into UA Workflow.
 
Growth trends in mobile gaming based on the data
Growth trends in mobile gaming based on the dataGrowth trends in mobile gaming based on the data
Growth trends in mobile gaming based on the data
 
Facebook's and Social Creative Best Practice that worked for Huuuge
Facebook's and Social Creative Best Practice that worked for HuuugeFacebook's and Social Creative Best Practice that worked for Huuuge
Facebook's and Social Creative Best Practice that worked for Huuuge
 
Using Data Science for Behavioural Game Design
Using Data Science for Behavioural Game DesignUsing Data Science for Behavioural Game Design
Using Data Science for Behavioural Game Design
 
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
Learnings and insights from Benchmarks+. Performance Insights for 1.2bn Month...
 
In-app offers monetization from basics to advanced technics based on Crazy Pa...
In-app offers monetization from basics to advanced technics based on Crazy Pa...In-app offers monetization from basics to advanced technics based on Crazy Pa...
In-app offers monetization from basics to advanced technics based on Crazy Pa...
 
Building cost-effective mobile product & marketing app analytics based on GCP...
Building cost-effective mobile product & marketing app analytics based on GCP...Building cost-effective mobile product & marketing app analytics based on GCP...
Building cost-effective mobile product & marketing app analytics based on GCP...
 
Enabling the growth of the game through performance marketing in 2020
Enabling the growth of the game through performance marketing in 2020Enabling the growth of the game through performance marketing in 2020
Enabling the growth of the game through performance marketing in 2020
 
Growth markets in mobile gaming
 Growth markets in mobile gaming Growth markets in mobile gaming
Growth markets in mobile gaming
 
LTV measurement and multi-touch attribution
LTV measurement and multi-touch attributionLTV measurement and multi-touch attribution
LTV measurement and multi-touch attribution
 
Big Problem, BigQuery: User Feature Engineering in Event-driven Analytics
Big Problem, BigQuery: User Feature Engineering in Event-driven AnalyticsBig Problem, BigQuery: User Feature Engineering in Event-driven Analytics
Big Problem, BigQuery: User Feature Engineering in Event-driven Analytics
 
Not so trivial lessons learned from launching mid-core games.
Not so trivial lessons learned from launching mid-core games.Not so trivial lessons learned from launching mid-core games.
Not so trivial lessons learned from launching mid-core games.
 

Optimising user acquisition through LTV prediction

  • 1. Optimising user acquisition through LTV prediction Róbert Magyar Warsaw 14 -15 October 2019
  • 2. Warsaw | 14-15 October 2019 Róbert Magyar Data Science Lead robert.magyar@superscale.com
  • 3. “We’re forming growth partnerships with world’s top developers to scale their games to maximum potential.”
  • 4. Warsaw | 14-15 October 2019 ● 70+ World Class Experts In-House ○ (Self-)Publishing Infrastructure i. Business Intelligence ii. Analytics & Data Science ○ Monetization i. LiveOps Optimization ii. Game & Monetization Design ○ User Acquisition i. Creatives ii. UA Campaign Management iii. ASO ● Founded in 2016 ● Bratislava, London, Berlin, Helsinki & Prague offices First Light Games Our partners Games
  • 5. Warsaw | 14-15 October 2019 UA Optimization Motivation
  • 6. Warsaw | 14-15 October 2019 How can we improve User Acquisition? Huge topic - lots of angles ● Lookalike redesign ● Different UA channels ● New creatives ● Different spending strategy ● Passively through game related efforts e.g. Economy Balancing ... ● Optimization of UA investments through LTV prediction - focus of this talk
  • 7. Warsaw | 14-15 October 2019 LTV predictions as an actionable insight for UA team Comes early in the campaign lifecycle Helps to identify opportunities Accuracy is stable over time Supports decision making Defines strategy of UA investment optimization
  • 8. Warsaw | 14-15 October 2019 Waiting for data is not fun Usual approach for building LTV prediction model: ● wait several weeks or months to gather data about the campaigns ● create LTV model ● use the model to estimate UA performance .. but market changes over time, new competitors arise etc
  • 9. Warsaw | 14-15 October 2019 The Question How can we improve UA efforts with LTV predictions ● early in campaigns lifecycle? ● without need of waiting several months to understand performance trend? => by understanding monetization of the game and building predictions from bottom-up How to actually do it? ● utilizing cloud machine learning
  • 10. Warsaw | 14-15 October 2019 Some games have steep early monetization which creates an advantage in modeling. Game monetization defines data needed for predictions = Game monetization => defines how many days of data we need and helps to understand the payback of UA 50% of revenue in first couple of days = enough data to work with
  • 11. Warsaw | 14-15 October 2019 Variation in campaign cohorts performance = accuracy loss => Accuracy of models is impacted by daily market changes, this can be seen on significant movements in campaign’s cohorts performances over time
  • 12. Warsaw | 14-15 October 2019 Building accurate predictions Focusing on daily cohorts => cohort’s distribution of payers does not change over time which improves accuracy while building up the predictions: 1. Predictions based on cohort level not on campaign level 2. Aggregate predictions for each campaign Idea is to go hierarchically to the lowest level possible (cohort/ad/adset etc) while: - Keeping enough players in cohorts = Daily cohorts => logarithmic growth can be leveraged => no need to wait several months for data
  • 13. Warsaw | 14-15 October 2019 Keeping enough players in cohorts Number of players needed is based on: - Conversion and number of payers Unusual growth of revenue in the cohort can be the clue of not having enough players to work with. Looks like step function - reason could be: 1. Game relying mainly on Liveops offers (not so good design) 2.Conversion for the cohort is not high enough
  • 14. Warsaw | 14-15 October 2019 UA Optimization LTV prediction in Action
  • 15. Warsaw | 14-15 October 2019 Information we are able to use: - Monetary : Conversion, ARPU, ARPPU, # of purchases, revenue, payers distribution, probability of 2nd purchase etc - Absolute value - Relative change/growth over first couple of days - per cohort or campaign and aggregated - Behavioral (Engagement metrics) : playtime, retention, first day drop-off, percentage of one-shots etc Leveraging all the data
  • 16. Warsaw | 14-15 October 2019 Business optimization goal
  • 17. SuperScale Analytics + UA Stack Data Sources (downloaded daily) Game Data (Appsflyer, Google Firebase, ...) Attribution Platform (Appsflyer, ...) UA Channel Data (Facebook, Google UAC, Ad-n...) Store Revenue Data (G Play, iTunes)
  • 18. SuperScale Analytics + UA Stack Data Sources (downloaded daily) Data Storage Materialized Views Google BigQuery UA Campaign Evaluation MVs Game Data (Appsflyer, Google Firebase, ...) Attribution Platform (Appsflyer, ...) UA Channel Data (Facebook, Google UAC, Ad-n...) Store Revenue Data (G Play, iTunes) Google Cloud Platform Machine Learning (stochastic algorithms, random forest, k-means clustering, …) SBDW ETL Processing (User States, Materialized Views)
  • 19. SuperScale Analytics + UA Stack Data Sources (downloaded daily) Data Storage Materialized Views Application Google BigQuery Data Validator (store revenue vs app revenue, AF revenue vs store) ROI/ROAS Spreadsheets UA Campaign Evaluation MVs Game Data (Appsflyer, Google Firebase, ...) Attribution Platform (Appsflyer, ...) UA Channel Data (Facebook, Google UAC, Ad-n...) Store Revenue Data (G Play, iTunes) FB Campaign Optimization Data Visualisation Campaign LTV/ROI prediction Google Cloud Platform Machine Learning (stochastic algorithms, random forest, k-means clustering, …) SBDW ETL Processing (User States, Materialized Views)
  • 20. Warsaw | 14-15 October 2019 Building predictions from a bottom-up Daily cohort 1 Daily cohort 2 Daily cohort 3 .. UA DATA PREDICTIONS for each cohort AGGREGATION of cohorts predictions on a higher level .. Weekly Monthly Quarterly Campaign Overall CPI LTV Conversion, ARPU, ARPPU, # purchases, revenue, payers distribution (absolute values, relative values) Campaign success prediction Weighting of data points based on recency
  • 21. Warsaw | 14-15 October 2019 Cloud pipeline takes care of automatic scaling of projects. Cloud prediction pipeline Data warehouse Batch preprocessing Machine learning BigQuery Creating models for: Daily, weekly, monthly cohorts and campaigns Additional analysis: Geo analysis Break-even analysis Business Intelligence Data Studio Scheduler Cloud Functions Models storage Google Cloud Storage Google Dataflow ML Engine on AI platform Raw GAME Data + Campaign data + LKLs data Periscope
  • 22. Warsaw | 14-15 October 2019 Supporting decision making - examples Weekly cohort estimate d LTV IAP D7 ROAS performance per country Weekly cohort LTV performance estimation Campaigns breakeven analysis Top country to target for LKLs Which strategy was the best When can we expect payback
  • 23. Warsaw | 14-15 October 2019 - LTV/ROAS predictions can help to identify opportunities but timing is the factor that makes or breaks any modelling - Aggregating weekly and monthly predictions from daily ones can improve accuracy significantly - Weighting data points based on recency can improve accuracy of estimations Takeaways
  • 24. Warsaw | 14-15 October 2019 Thank you for attention. Questions?