SlideShare a Scribd company logo
Know Your Customers
August 17 2019
Confidential. Do not distribute.
Fiona Li
Manager, Analytics
Implement a Winning Strategy Driven by Customer Lifetime Value
2
Agenda
● What is Customer Lifetime Value (CLV) and why do we care?
● How CLV is used for a business decision
● Deep dive on BG/BB probabilistic models
● Demo on an offline model: Implement BG/BB models in Python
● Select the right CLV model for your business
Imagine you are opening a boutique movie theatre...let’s say we
named it Popcorn & Soda, located in the University Village...
3
“The average life span of my potential customers is 5 years.
And on average, people visit similar theatres spend about $20 per month. ”
What is my Customer Lifetime Value?
$20/month x 12 months/year x 5 years = $1,200
How much should I spend on acquiring a customer in the first year,
if I would like to maintain a 20% profit?
$20/month x 12 months/year x 20% = $48
4
Popcorn & Soda becomes a huge hit and you have earned
enough money for a MBA at Marshall...
m: the net cash flow per period (w/20% margin)
r: the retention rate (80% in this case)
d: the discount rate (assume 10%)
T: the time horizon for the calculation
What is my Customer Lifetime Value?
How much should I spend on acquiring a customer in the first year,
if I would like to maintain a 20% profit?
CLV (t=0) = $20/month x 12 months/year x 20% = $48
m = $20/month x 12 months/year x 5 years x 20% = $240
something = 2.91
CLV = $240 x 2.91 = $700
Reference: What’s Wrong With This CLV Formula? P. Fader, B.
Hardie
At the Marketing 101 Class:
Classifying Customer Bases
5
Contractual
customer “death” can be observed and often
modeled using survival-based approach
Non-contractual
customer “death” is unobserved and
lifetime distribution often modeled via
probabilistic models
Discrete
purchases
occur at fixed
period or
frequencies
● Streaming Services: Netflix, Hulu etc
● Magazine subscriptions
● Most insurance policy
● Movie, concert attendance
● Prescription refills
● Charity fund drives
Continuous
purchases can
happen at any
time
● Credit cards transactions
● Utilities
● Continuity programs: airline frequent flyer
discount program etc
● Retail purchasing, i.e.Starbucks
● Hotel stays
● Doctor visits
Typical Purchasing Journeys
Alice:
Brian:
Charlie:
6
Past Future
?
CLV Definition and Equation
7
Future Customer Lifetime Value =
Profit Margin % x Discounted Expected Residual Transactions x Predicted Order Value
Customer Lifetime Value is
“the total profit of the entire relationship with a customer throughout the customer journey, in the past and
future.”
Reference: Customer-Base Analysis in a Discrete-Time Noncontractual Setting P. Fader, B. Hardie, J.
Shang
“By sending a second email to our weekly newsletter subscribers, we observed a
80% lift in conversions! However, we also experienced a 20% increase of
unsubscription rate.
So should we continue sending out the second email or not? Given the numbers,
can we have a 1-year projection on the net impact?”
8
Popcorn & Soda is very successful, expanded to 20+ college
towns...now we are not only have a Marketing team but also
has a Data Science team...
In a recent memo from the Head of Marketing to the Head of Data Science:
“Two Coins” of BG/BB Model
a.k.a. Buy Till You Die model
9
Purchase Coin
Make a transaction!
Yay! I’m going to see a movie!
🤘🤘
No transaction!
I hate my coin!
😞
Death Coin
You “die”! No escape!
OMG! The Death Lord’s here!
☠️
You stay “alive”!
Angel to the rescue!
😇
Assume each customer has two uneven coins
Model Development
The dying (BG) and purchasing (BB) processes
10
“Oh, End Game is on tonight 8pm at Popcorn & Soda.
Should I go see it?”
“Hmm, I just moved to Hollywood. I do like the theatre, but
should I drive 10 miles to see a movie at Popcorn & Soda?”
“Oh, I received an email from P&S and saying that there are lots of previous movies are on at P&S with lower rates. So…”
A customer makes a transaction at any given
transaction opportunity with probability p. This is
known as a binomial distribution:
A “living” customer “dies” at the beginning of a
transaction opportunity with probability θ. This is
known as a geometric distribution:
Brian:
11
● The order of given number of transactions prior to the most recent transaction doesn’t matter.But what matters
is when the customer is “dead” θ (Recency) and how many transactions p (Frequency) he/she has.
● We compute the probability of the transaction on each customer and then multiply it by the probability with
other customers’.
Brian’s transaction history - 100100
Model Development
An Example
12
Beta Geometric Beta Binomial
BGBB Model
Model Development
Heterogeneity is captured in Beta Distribution and and the log-likelihood function
Demo in Python
13
Select a CLV Model for your business model
14
Deploy a Production ModelBusiness Context Select an Offline Model
Select a CLV model - suggestions and caveats
15
● Depends on the nature of
customer base, the data
size can be huge.
● To decide if it’s a
streaming, online or AaaS
deployment.
● To deploy in a container i.e.
Docker to control package
versions.
● To write engineering
production code with
adequate test cases.
● For contractuals, we can use
survival based models, while in non-
contractual settings, we can
consider probabilistic models, i.e.
hierarchical Bayesian such as
Pareto/NBD, Pareto/GGG.
● Tools/Packages can be:
○ Python: lifetime, PyMC
○ R: BTYD, BTYDPlus
● Metrics to evaluate performance:
MAPE, tradeoff between granularity
and run-time if running MCMC.
● What are the values added if
we have CLV?
● To understand the business
context and business model.
○ Is the business model on a
contractual or non-
contractual basis?
○ Are the customer
purchases continuous or
discrete?
● How far we would like to
have CLV predicted?
16
Popcorn & Soda successfully chose, implemented CLV models...
The Head of Data Science sends out a memo to the Executives and the Marketing Team:
“The estimated incremental revenue gained from the 80% conversion is about $3M. Based on the
10% increased number of unsubscribers we projected CLV for a year is about $1M. The Net Impact
from the calculation above is about $2M per year. Based on revenue impact calculations and cost
estimations, we believe that we will be better off if launching this product feature.”
Thank you
Confidential. Do not distribute.
Why do we care about CLV?
18
From a strategic point of view:
● Company valuation based on customer equity
● Annual budgeting based on CLV
● Customer segmentation
Operationally:
● Identify most profitable customers, then
● Identify traits and features of valuable customers for engagement and growth
● Dictates allocation of marketing resources, i.e. acquisition cost, SEM spend at a visitor level
● To track sales forecasts
APPENDIX

More Related Content

What's hot

Predicting Credit Card Defaults using Machine Learning Algorithms
Predicting Credit Card Defaults using Machine Learning AlgorithmsPredicting Credit Card Defaults using Machine Learning Algorithms
Predicting Credit Card Defaults using Machine Learning Algorithms
Sagar Tupkar
 
Churn Prediction in Practice
Churn Prediction in PracticeChurn Prediction in Practice
Churn Prediction in Practice
BigData Republic
 
Customer churn prediction in banking
Customer churn prediction in bankingCustomer churn prediction in banking
Customer churn prediction in banking
BU - PG Master Computing Conference
 
Decision Tree and Random forest
Decision Tree and Random forestDecision Tree and Random forest
Decision Tree and Random forest
Vishal Gupta
 
Machine Learning At Tubi
Machine Learning At TubiMachine Learning At Tubi
Machine Learning At Tubi
Jaya Kawale
 
Netflix talk at ML Platform meetup Sep 2019
Netflix talk at ML Platform meetup Sep 2019Netflix talk at ML Platform meetup Sep 2019
Netflix talk at ML Platform meetup Sep 2019
Faisal Siddiqi
 
Churn customer analysis
Churn customer analysisChurn customer analysis
Churn customer analysis
Dr.Bechoo Lal
 
Customer churn prediction for telecom data set.
Customer churn prediction for telecom data set.Customer churn prediction for telecom data set.
Customer churn prediction for telecom data set.
Kuldeep Mahani
 
Recent Trends in Personalization: A Netflix Perspective
Recent Trends in Personalization: A Netflix PerspectiveRecent Trends in Personalization: A Netflix Perspective
Recent Trends in Personalization: A Netflix Perspective
Justin Basilico
 
Ways to Reduce the Customer Churn Rate
Ways to Reduce the Customer Churn RateWays to Reduce the Customer Churn Rate
Ways to Reduce the Customer Churn Rate
FORMCEPT
 
Customer Churn Analysis and Prediction
Customer Churn Analysis and PredictionCustomer Churn Analysis and Prediction
Customer Churn Analysis and Prediction
SOUMIT KAR
 
Money 3.0
Money 3.0Money 3.0
Money 3.0
ChrisJCook
 
Deep neural networks for Youtube recommendations
Deep neural networks for Youtube recommendationsDeep neural networks for Youtube recommendations
Deep neural networks for Youtube recommendations
Aryan Khandal
 
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
Bartlomiej Twardowski
 
Telco churn presentation
Telco churn presentationTelco churn presentation
Telco churn presentation
Aditya Bahl
 
Why start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaignsWhy start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaigns
Data Con LA
 

What's hot (16)

Predicting Credit Card Defaults using Machine Learning Algorithms
Predicting Credit Card Defaults using Machine Learning AlgorithmsPredicting Credit Card Defaults using Machine Learning Algorithms
Predicting Credit Card Defaults using Machine Learning Algorithms
 
Churn Prediction in Practice
Churn Prediction in PracticeChurn Prediction in Practice
Churn Prediction in Practice
 
Customer churn prediction in banking
Customer churn prediction in bankingCustomer churn prediction in banking
Customer churn prediction in banking
 
Decision Tree and Random forest
Decision Tree and Random forestDecision Tree and Random forest
Decision Tree and Random forest
 
Machine Learning At Tubi
Machine Learning At TubiMachine Learning At Tubi
Machine Learning At Tubi
 
Netflix talk at ML Platform meetup Sep 2019
Netflix talk at ML Platform meetup Sep 2019Netflix talk at ML Platform meetup Sep 2019
Netflix talk at ML Platform meetup Sep 2019
 
Churn customer analysis
Churn customer analysisChurn customer analysis
Churn customer analysis
 
Customer churn prediction for telecom data set.
Customer churn prediction for telecom data set.Customer churn prediction for telecom data set.
Customer churn prediction for telecom data set.
 
Recent Trends in Personalization: A Netflix Perspective
Recent Trends in Personalization: A Netflix PerspectiveRecent Trends in Personalization: A Netflix Perspective
Recent Trends in Personalization: A Netflix Perspective
 
Ways to Reduce the Customer Churn Rate
Ways to Reduce the Customer Churn RateWays to Reduce the Customer Churn Rate
Ways to Reduce the Customer Churn Rate
 
Customer Churn Analysis and Prediction
Customer Churn Analysis and PredictionCustomer Churn Analysis and Prediction
Customer Churn Analysis and Prediction
 
Money 3.0
Money 3.0Money 3.0
Money 3.0
 
Deep neural networks for Youtube recommendations
Deep neural networks for Youtube recommendationsDeep neural networks for Youtube recommendations
Deep neural networks for Youtube recommendations
 
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
Systemy rekomendacji, Algorytmy rankingu Top-N rekomendacji bazujące na nieja...
 
Telco churn presentation
Telco churn presentationTelco churn presentation
Telco churn presentation
 
Why start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaignsWhy start using uplift models for more efficient marketing campaigns
Why start using uplift models for more efficient marketing campaigns
 

Similar to Data Con LA 2019 - Know Your Customer by Fiona Li

Customer lifetime value (1)
Customer lifetime value (1)Customer lifetime value (1)
Customer lifetime value (1)
KRISTINA FISCHEROVA
 
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture SeriesCalculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
Luciano Pesci, PhD
 
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on AdvertisingConvincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Kevin Dieny
 
[Webinar] How To Compete Against The Big Fish
[Webinar] How To Compete Against The Big Fish[Webinar] How To Compete Against The Big Fish
[Webinar] How To Compete Against The Big Fish
VWO
 
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
Tejari
 
(10-15-14) Lehigh Marketing 111 Lecture Materials
(10-15-14) Lehigh Marketing 111 Lecture Materials(10-15-14) Lehigh Marketing 111 Lecture Materials
(10-15-14) Lehigh Marketing 111 Lecture Materials
Brent Chudoba
 
1530 track 1 fader_using our laptop
1530 track 1 fader_using our laptop1530 track 1 fader_using our laptop
1530 track 1 fader_using our laptop
Rising Media, Inc.
 
Playing the Marketing Long Game
Playing the Marketing Long GamePlaying the Marketing Long Game
Playing the Marketing Long Game
Work-Bench
 
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on AdvertisingConvincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Hanapin Marketing
 
Target selling the breakthrough approach of today
Target selling the breakthrough approach of todayTarget selling the breakthrough approach of today
Target selling the breakthrough approach of today
Dr Wilfred Monteiro
 
Sun Emea Summit Slides 9.10.2006
Sun Emea Summit Slides   9.10.2006Sun Emea Summit Slides   9.10.2006
Sun Emea Summit Slides 9.10.2006
Alan Mather
 
Customer Lifetime Value: The Core Metric in Marketing
Customer Lifetime Value: The Core Metric in MarketingCustomer Lifetime Value: The Core Metric in Marketing
Customer Lifetime Value: The Core Metric in Marketing
Adknowledge
 
100 Days as a CEO
100 Days as a CEO100 Days as a CEO
100 Days as a CEO
Richard Brooks
 
LinkedIn for Business Development SourceIn London, James Osborne
LinkedIn for Business Development SourceIn London, James OsborneLinkedIn for Business Development SourceIn London, James Osborne
LinkedIn for Business Development SourceIn London, James Osborne
LinkedIn For Search and Recruitment Firms
 
Using Lifetime Value to Optimize Your Digital Marketing Investments
Using Lifetime Value to Optimize Your Digital Marketing InvestmentsUsing Lifetime Value to Optimize Your Digital Marketing Investments
Using Lifetime Value to Optimize Your Digital Marketing Investments
Adknowledge
 
StartupTalk #35 - How to Price your First SaaS Product
StartupTalk #35 - How to Price your First SaaS ProductStartupTalk #35 - How to Price your First SaaS Product
StartupTalk #35 - How to Price your First SaaS Product
PreSeed Ventures
 
Fundraising During a Pandemic with Point Nine Capital
Fundraising During a Pandemic with Point Nine CapitalFundraising During a Pandemic with Point Nine Capital
Fundraising During a Pandemic with Point Nine Capital
saastr
 
Lifetime Value - The Only Metric That Matters (DMC September 2018)
Lifetime Value - The Only Metric That Matters (DMC September 2018)Lifetime Value - The Only Metric That Matters (DMC September 2018)
Lifetime Value - The Only Metric That Matters (DMC September 2018)
Luciano Pesci, PhD
 
Lifetime Value (the only metric that matters) Utah DMC September 2018
Lifetime Value (the only metric that matters) Utah DMC September 2018Lifetime Value (the only metric that matters) Utah DMC September 2018
Lifetime Value (the only metric that matters) Utah DMC September 2018
Utah Digital Marketing Collective
 
Netflix's Customer Obsession by fmr Netflix VP of Product
Netflix's Customer Obsession by fmr Netflix VP of ProductNetflix's Customer Obsession by fmr Netflix VP of Product
Netflix's Customer Obsession by fmr Netflix VP of Product
Product School
 

Similar to Data Con LA 2019 - Know Your Customer by Fiona Li (20)

Customer lifetime value (1)
Customer lifetime value (1)Customer lifetime value (1)
Customer lifetime value (1)
 
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture SeriesCalculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
Calculating Your Customer Lifetime Value - Dawn of the Data Age Lecture Series
 
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on AdvertisingConvincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
 
[Webinar] How To Compete Against The Big Fish
[Webinar] How To Compete Against The Big Fish[Webinar] How To Compete Against The Big Fish
[Webinar] How To Compete Against The Big Fish
 
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
Two Case Studies that Unlocked Invisible Value Where Cost Price was a Lower L...
 
(10-15-14) Lehigh Marketing 111 Lecture Materials
(10-15-14) Lehigh Marketing 111 Lecture Materials(10-15-14) Lehigh Marketing 111 Lecture Materials
(10-15-14) Lehigh Marketing 111 Lecture Materials
 
1530 track 1 fader_using our laptop
1530 track 1 fader_using our laptop1530 track 1 fader_using our laptop
1530 track 1 fader_using our laptop
 
Playing the Marketing Long Game
Playing the Marketing Long GamePlaying the Marketing Long Game
Playing the Marketing Long Game
 
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on AdvertisingConvincing Your Boss(es) to Confidently Spend (more) on Advertising
Convincing Your Boss(es) to Confidently Spend (more) on Advertising
 
Target selling the breakthrough approach of today
Target selling the breakthrough approach of todayTarget selling the breakthrough approach of today
Target selling the breakthrough approach of today
 
Sun Emea Summit Slides 9.10.2006
Sun Emea Summit Slides   9.10.2006Sun Emea Summit Slides   9.10.2006
Sun Emea Summit Slides 9.10.2006
 
Customer Lifetime Value: The Core Metric in Marketing
Customer Lifetime Value: The Core Metric in MarketingCustomer Lifetime Value: The Core Metric in Marketing
Customer Lifetime Value: The Core Metric in Marketing
 
100 Days as a CEO
100 Days as a CEO100 Days as a CEO
100 Days as a CEO
 
LinkedIn for Business Development SourceIn London, James Osborne
LinkedIn for Business Development SourceIn London, James OsborneLinkedIn for Business Development SourceIn London, James Osborne
LinkedIn for Business Development SourceIn London, James Osborne
 
Using Lifetime Value to Optimize Your Digital Marketing Investments
Using Lifetime Value to Optimize Your Digital Marketing InvestmentsUsing Lifetime Value to Optimize Your Digital Marketing Investments
Using Lifetime Value to Optimize Your Digital Marketing Investments
 
StartupTalk #35 - How to Price your First SaaS Product
StartupTalk #35 - How to Price your First SaaS ProductStartupTalk #35 - How to Price your First SaaS Product
StartupTalk #35 - How to Price your First SaaS Product
 
Fundraising During a Pandemic with Point Nine Capital
Fundraising During a Pandemic with Point Nine CapitalFundraising During a Pandemic with Point Nine Capital
Fundraising During a Pandemic with Point Nine Capital
 
Lifetime Value - The Only Metric That Matters (DMC September 2018)
Lifetime Value - The Only Metric That Matters (DMC September 2018)Lifetime Value - The Only Metric That Matters (DMC September 2018)
Lifetime Value - The Only Metric That Matters (DMC September 2018)
 
Lifetime Value (the only metric that matters) Utah DMC September 2018
Lifetime Value (the only metric that matters) Utah DMC September 2018Lifetime Value (the only metric that matters) Utah DMC September 2018
Lifetime Value (the only metric that matters) Utah DMC September 2018
 
Netflix's Customer Obsession by fmr Netflix VP of Product
Netflix's Customer Obsession by fmr Netflix VP of ProductNetflix's Customer Obsession by fmr Netflix VP of Product
Netflix's Customer Obsession by fmr Netflix VP of Product
 

More from Data Con LA

Data Con LA 2022 Keynotes
Data Con LA 2022 KeynotesData Con LA 2022 Keynotes
Data Con LA 2022 Keynotes
Data Con LA
 
Data Con LA 2022 Keynotes
Data Con LA 2022 KeynotesData Con LA 2022 Keynotes
Data Con LA 2022 Keynotes
Data Con LA
 
Data Con LA 2022 Keynote
Data Con LA 2022 KeynoteData Con LA 2022 Keynote
Data Con LA 2022 Keynote
Data Con LA
 
Data Con LA 2022 - Startup Showcase
Data Con LA 2022 - Startup ShowcaseData Con LA 2022 - Startup Showcase
Data Con LA 2022 - Startup Showcase
Data Con LA
 
Data Con LA 2022 Keynote
Data Con LA 2022 KeynoteData Con LA 2022 Keynote
Data Con LA 2022 Keynote
Data Con LA
 
Data Con LA 2022 - Using Google trends data to build product recommendations
Data Con LA 2022 - Using Google trends data to build product recommendationsData Con LA 2022 - Using Google trends data to build product recommendations
Data Con LA 2022 - Using Google trends data to build product recommendations
Data Con LA
 
Data Con LA 2022 - AI Ethics
Data Con LA 2022 - AI EthicsData Con LA 2022 - AI Ethics
Data Con LA 2022 - AI Ethics
Data Con LA
 
Data Con LA 2022 - Improving disaster response with machine learning
Data Con LA 2022 - Improving disaster response with machine learningData Con LA 2022 - Improving disaster response with machine learning
Data Con LA 2022 - Improving disaster response with machine learning
Data Con LA
 
Data Con LA 2022 - What's new with MongoDB 6.0 and Atlas
Data Con LA 2022 - What's new with MongoDB 6.0 and AtlasData Con LA 2022 - What's new with MongoDB 6.0 and Atlas
Data Con LA 2022 - What's new with MongoDB 6.0 and Atlas
Data Con LA
 
Data Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentationData Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentation
Data Con LA
 
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
Data Con LA
 
Data Con LA 2022 - Moving Data at Scale to AWS
Data Con LA 2022 - Moving Data at Scale to AWSData Con LA 2022 - Moving Data at Scale to AWS
Data Con LA 2022 - Moving Data at Scale to AWS
Data Con LA
 
Data Con LA 2022 - Collaborative Data Exploration using Conversational AI
Data Con LA 2022 - Collaborative Data Exploration using Conversational AIData Con LA 2022 - Collaborative Data Exploration using Conversational AI
Data Con LA 2022 - Collaborative Data Exploration using Conversational AI
Data Con LA
 
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
Data Con LA
 
Data Con LA 2022 - Intro to Data Science
Data Con LA 2022 - Intro to Data ScienceData Con LA 2022 - Intro to Data Science
Data Con LA 2022 - Intro to Data Science
Data Con LA
 
Data Con LA 2022 - How are NFTs and DeFi Changing Entertainment
Data Con LA 2022 - How are NFTs and DeFi Changing EntertainmentData Con LA 2022 - How are NFTs and DeFi Changing Entertainment
Data Con LA 2022 - How are NFTs and DeFi Changing Entertainment
Data Con LA
 
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
Data Con LA
 
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
Data Con LA
 
Data Con LA 2022- Embedding medical journeys with machine learning to improve...
Data Con LA 2022- Embedding medical journeys with machine learning to improve...Data Con LA 2022- Embedding medical journeys with machine learning to improve...
Data Con LA 2022- Embedding medical journeys with machine learning to improve...
Data Con LA
 
Data Con LA 2022 - Data Streaming with Kafka
Data Con LA 2022 - Data Streaming with KafkaData Con LA 2022 - Data Streaming with Kafka
Data Con LA 2022 - Data Streaming with Kafka
Data Con LA
 

More from Data Con LA (20)

Data Con LA 2022 Keynotes
Data Con LA 2022 KeynotesData Con LA 2022 Keynotes
Data Con LA 2022 Keynotes
 
Data Con LA 2022 Keynotes
Data Con LA 2022 KeynotesData Con LA 2022 Keynotes
Data Con LA 2022 Keynotes
 
Data Con LA 2022 Keynote
Data Con LA 2022 KeynoteData Con LA 2022 Keynote
Data Con LA 2022 Keynote
 
Data Con LA 2022 - Startup Showcase
Data Con LA 2022 - Startup ShowcaseData Con LA 2022 - Startup Showcase
Data Con LA 2022 - Startup Showcase
 
Data Con LA 2022 Keynote
Data Con LA 2022 KeynoteData Con LA 2022 Keynote
Data Con LA 2022 Keynote
 
Data Con LA 2022 - Using Google trends data to build product recommendations
Data Con LA 2022 - Using Google trends data to build product recommendationsData Con LA 2022 - Using Google trends data to build product recommendations
Data Con LA 2022 - Using Google trends data to build product recommendations
 
Data Con LA 2022 - AI Ethics
Data Con LA 2022 - AI EthicsData Con LA 2022 - AI Ethics
Data Con LA 2022 - AI Ethics
 
Data Con LA 2022 - Improving disaster response with machine learning
Data Con LA 2022 - Improving disaster response with machine learningData Con LA 2022 - Improving disaster response with machine learning
Data Con LA 2022 - Improving disaster response with machine learning
 
Data Con LA 2022 - What's new with MongoDB 6.0 and Atlas
Data Con LA 2022 - What's new with MongoDB 6.0 and AtlasData Con LA 2022 - What's new with MongoDB 6.0 and Atlas
Data Con LA 2022 - What's new with MongoDB 6.0 and Atlas
 
Data Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentationData Con LA 2022 - Real world consumer segmentation
Data Con LA 2022 - Real world consumer segmentation
 
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
Data Con LA 2022 - Modernizing Analytics & AI for today's needs: Intuit Turbo...
 
Data Con LA 2022 - Moving Data at Scale to AWS
Data Con LA 2022 - Moving Data at Scale to AWSData Con LA 2022 - Moving Data at Scale to AWS
Data Con LA 2022 - Moving Data at Scale to AWS
 
Data Con LA 2022 - Collaborative Data Exploration using Conversational AI
Data Con LA 2022 - Collaborative Data Exploration using Conversational AIData Con LA 2022 - Collaborative Data Exploration using Conversational AI
Data Con LA 2022 - Collaborative Data Exploration using Conversational AI
 
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
Data Con LA 2022 - Why Database Modernization Makes Your Data Decisions More ...
 
Data Con LA 2022 - Intro to Data Science
Data Con LA 2022 - Intro to Data ScienceData Con LA 2022 - Intro to Data Science
Data Con LA 2022 - Intro to Data Science
 
Data Con LA 2022 - How are NFTs and DeFi Changing Entertainment
Data Con LA 2022 - How are NFTs and DeFi Changing EntertainmentData Con LA 2022 - How are NFTs and DeFi Changing Entertainment
Data Con LA 2022 - How are NFTs and DeFi Changing Entertainment
 
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
Data Con LA 2022 - Why Data Quality vigilance requires an End-to-End, Automat...
 
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
Data Con LA 2022-Perfect Viral Ad prediction of Superbowl 2022 using Tease, T...
 
Data Con LA 2022- Embedding medical journeys with machine learning to improve...
Data Con LA 2022- Embedding medical journeys with machine learning to improve...Data Con LA 2022- Embedding medical journeys with machine learning to improve...
Data Con LA 2022- Embedding medical journeys with machine learning to improve...
 
Data Con LA 2022 - Data Streaming with Kafka
Data Con LA 2022 - Data Streaming with KafkaData Con LA 2022 - Data Streaming with Kafka
Data Con LA 2022 - Data Streaming with Kafka
 

Recently uploaded

Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
Safe Software
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
panagenda
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
Zilliz
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
LucaBarbaro3
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
DanBrown980551
 
Azure API Management to expose backend services securely
Azure API Management to expose backend services securelyAzure API Management to expose backend services securely
Azure API Management to expose backend services securely
Dinusha Kumarasiri
 
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStrDeep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
saastr
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
shyamraj55
 
Recommendation System using RAG Architecture
Recommendation System using RAG ArchitectureRecommendation System using RAG Architecture
Recommendation System using RAG Architecture
fredae14
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
Hiroshi SHIBATA
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
Octavian Nadolu
 
A Comprehensive Guide to DeFi Development Services in 2024
A Comprehensive Guide to DeFi Development Services in 2024A Comprehensive Guide to DeFi Development Services in 2024
A Comprehensive Guide to DeFi Development Services in 2024
Intelisync
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
Ivanti
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
akankshawande
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
Chart Kalyan
 
dbms calicut university B. sc Cs 4th sem.pdf
dbms  calicut university B. sc Cs 4th sem.pdfdbms  calicut university B. sc Cs 4th sem.pdf
dbms calicut university B. sc Cs 4th sem.pdf
Shinana2
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
Jakub Marek
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
saastr
 
Taking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdfTaking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdf
ssuserfac0301
 
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptxOcean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
SitimaJohn
 

Recently uploaded (20)

Driving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success StoryDriving Business Innovation: Latest Generative AI Advancements & Success Story
Driving Business Innovation: Latest Generative AI Advancements & Success Story
 
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAUHCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
HCL Notes und Domino Lizenzkostenreduzierung in der Welt von DLAU
 
Generating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and MilvusGenerating privacy-protected synthetic data using Secludy and Milvus
Generating privacy-protected synthetic data using Secludy and Milvus
 
Trusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process MiningTrusted Execution Environment for Decentralized Process Mining
Trusted Execution Environment for Decentralized Process Mining
 
5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides5th LF Energy Power Grid Model Meet-up Slides
5th LF Energy Power Grid Model Meet-up Slides
 
Azure API Management to expose backend services securely
Azure API Management to expose backend services securelyAzure API Management to expose backend services securely
Azure API Management to expose backend services securely
 
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStrDeep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
Deep Dive: Getting Funded with Jason Jason Lemkin Founder & CEO @ SaaStr
 
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with SlackLet's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
Let's Integrate MuleSoft RPA, COMPOSER, APM with AWS IDP along with Slack
 
Recommendation System using RAG Architecture
Recommendation System using RAG ArchitectureRecommendation System using RAG Architecture
Recommendation System using RAG Architecture
 
Introduction of Cybersecurity with OSS at Code Europe 2024
Introduction of Cybersecurity with OSS  at Code Europe 2024Introduction of Cybersecurity with OSS  at Code Europe 2024
Introduction of Cybersecurity with OSS at Code Europe 2024
 
Artificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopmentArtificial Intelligence for XMLDevelopment
Artificial Intelligence for XMLDevelopment
 
A Comprehensive Guide to DeFi Development Services in 2024
A Comprehensive Guide to DeFi Development Services in 2024A Comprehensive Guide to DeFi Development Services in 2024
A Comprehensive Guide to DeFi Development Services in 2024
 
June Patch Tuesday
June Patch TuesdayJune Patch Tuesday
June Patch Tuesday
 
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development ProvidersYour One-Stop Shop for Python Success: Top 10 US Python Development Providers
Your One-Stop Shop for Python Success: Top 10 US Python Development Providers
 
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdfHow to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
How to Interpret Trends in the Kalyan Rajdhani Mix Chart.pdf
 
dbms calicut university B. sc Cs 4th sem.pdf
dbms  calicut university B. sc Cs 4th sem.pdfdbms  calicut university B. sc Cs 4th sem.pdf
dbms calicut university B. sc Cs 4th sem.pdf
 
Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)Main news related to the CCS TSI 2023 (2023/1695)
Main news related to the CCS TSI 2023 (2023/1695)
 
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
Overcoming the PLG Trap: Lessons from Canva's Head of Sales & Head of EMEA Da...
 
Taking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdfTaking AI to the Next Level in Manufacturing.pdf
Taking AI to the Next Level in Manufacturing.pdf
 
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptxOcean lotus Threat actors project by John Sitima 2024 (1).pptx
Ocean lotus Threat actors project by John Sitima 2024 (1).pptx
 

Data Con LA 2019 - Know Your Customer by Fiona Li

  • 1. Know Your Customers August 17 2019 Confidential. Do not distribute. Fiona Li Manager, Analytics Implement a Winning Strategy Driven by Customer Lifetime Value
  • 2. 2 Agenda ● What is Customer Lifetime Value (CLV) and why do we care? ● How CLV is used for a business decision ● Deep dive on BG/BB probabilistic models ● Demo on an offline model: Implement BG/BB models in Python ● Select the right CLV model for your business
  • 3. Imagine you are opening a boutique movie theatre...let’s say we named it Popcorn & Soda, located in the University Village... 3 “The average life span of my potential customers is 5 years. And on average, people visit similar theatres spend about $20 per month. ” What is my Customer Lifetime Value? $20/month x 12 months/year x 5 years = $1,200 How much should I spend on acquiring a customer in the first year, if I would like to maintain a 20% profit? $20/month x 12 months/year x 20% = $48
  • 4. 4 Popcorn & Soda becomes a huge hit and you have earned enough money for a MBA at Marshall... m: the net cash flow per period (w/20% margin) r: the retention rate (80% in this case) d: the discount rate (assume 10%) T: the time horizon for the calculation What is my Customer Lifetime Value? How much should I spend on acquiring a customer in the first year, if I would like to maintain a 20% profit? CLV (t=0) = $20/month x 12 months/year x 20% = $48 m = $20/month x 12 months/year x 5 years x 20% = $240 something = 2.91 CLV = $240 x 2.91 = $700 Reference: What’s Wrong With This CLV Formula? P. Fader, B. Hardie At the Marketing 101 Class:
  • 5. Classifying Customer Bases 5 Contractual customer “death” can be observed and often modeled using survival-based approach Non-contractual customer “death” is unobserved and lifetime distribution often modeled via probabilistic models Discrete purchases occur at fixed period or frequencies ● Streaming Services: Netflix, Hulu etc ● Magazine subscriptions ● Most insurance policy ● Movie, concert attendance ● Prescription refills ● Charity fund drives Continuous purchases can happen at any time ● Credit cards transactions ● Utilities ● Continuity programs: airline frequent flyer discount program etc ● Retail purchasing, i.e.Starbucks ● Hotel stays ● Doctor visits
  • 7. CLV Definition and Equation 7 Future Customer Lifetime Value = Profit Margin % x Discounted Expected Residual Transactions x Predicted Order Value Customer Lifetime Value is “the total profit of the entire relationship with a customer throughout the customer journey, in the past and future.” Reference: Customer-Base Analysis in a Discrete-Time Noncontractual Setting P. Fader, B. Hardie, J. Shang
  • 8. “By sending a second email to our weekly newsletter subscribers, we observed a 80% lift in conversions! However, we also experienced a 20% increase of unsubscription rate. So should we continue sending out the second email or not? Given the numbers, can we have a 1-year projection on the net impact?” 8 Popcorn & Soda is very successful, expanded to 20+ college towns...now we are not only have a Marketing team but also has a Data Science team... In a recent memo from the Head of Marketing to the Head of Data Science:
  • 9. “Two Coins” of BG/BB Model a.k.a. Buy Till You Die model 9 Purchase Coin Make a transaction! Yay! I’m going to see a movie! 🤘🤘 No transaction! I hate my coin! 😞 Death Coin You “die”! No escape! OMG! The Death Lord’s here! ☠️ You stay “alive”! Angel to the rescue! 😇 Assume each customer has two uneven coins
  • 10. Model Development The dying (BG) and purchasing (BB) processes 10 “Oh, End Game is on tonight 8pm at Popcorn & Soda. Should I go see it?” “Hmm, I just moved to Hollywood. I do like the theatre, but should I drive 10 miles to see a movie at Popcorn & Soda?” “Oh, I received an email from P&S and saying that there are lots of previous movies are on at P&S with lower rates. So…” A customer makes a transaction at any given transaction opportunity with probability p. This is known as a binomial distribution: A “living” customer “dies” at the beginning of a transaction opportunity with probability θ. This is known as a geometric distribution:
  • 11. Brian: 11 ● The order of given number of transactions prior to the most recent transaction doesn’t matter.But what matters is when the customer is “dead” θ (Recency) and how many transactions p (Frequency) he/she has. ● We compute the probability of the transaction on each customer and then multiply it by the probability with other customers’. Brian’s transaction history - 100100 Model Development An Example
  • 12. 12 Beta Geometric Beta Binomial BGBB Model Model Development Heterogeneity is captured in Beta Distribution and and the log-likelihood function
  • 14. Select a CLV Model for your business model 14
  • 15. Deploy a Production ModelBusiness Context Select an Offline Model Select a CLV model - suggestions and caveats 15 ● Depends on the nature of customer base, the data size can be huge. ● To decide if it’s a streaming, online or AaaS deployment. ● To deploy in a container i.e. Docker to control package versions. ● To write engineering production code with adequate test cases. ● For contractuals, we can use survival based models, while in non- contractual settings, we can consider probabilistic models, i.e. hierarchical Bayesian such as Pareto/NBD, Pareto/GGG. ● Tools/Packages can be: ○ Python: lifetime, PyMC ○ R: BTYD, BTYDPlus ● Metrics to evaluate performance: MAPE, tradeoff between granularity and run-time if running MCMC. ● What are the values added if we have CLV? ● To understand the business context and business model. ○ Is the business model on a contractual or non- contractual basis? ○ Are the customer purchases continuous or discrete? ● How far we would like to have CLV predicted?
  • 16. 16 Popcorn & Soda successfully chose, implemented CLV models... The Head of Data Science sends out a memo to the Executives and the Marketing Team: “The estimated incremental revenue gained from the 80% conversion is about $3M. Based on the 10% increased number of unsubscribers we projected CLV for a year is about $1M. The Net Impact from the calculation above is about $2M per year. Based on revenue impact calculations and cost estimations, we believe that we will be better off if launching this product feature.”
  • 17. Thank you Confidential. Do not distribute.
  • 18. Why do we care about CLV? 18 From a strategic point of view: ● Company valuation based on customer equity ● Annual budgeting based on CLV ● Customer segmentation Operationally: ● Identify most profitable customers, then ● Identify traits and features of valuable customers for engagement and growth ● Dictates allocation of marketing resources, i.e. acquisition cost, SEM spend at a visitor level ● To track sales forecasts APPENDIX