Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
WIFI SSID:Spark+AISummit | Password: UnifiedDataAnalytics
Creating an Omnichannel Banking
Experience with Machine Learning
on Azure Databricks
Petr Pluhacek (Ceska Sporitelna)
Jaku...
Who is presenting
today
Petr Pluháček, Česká spořitelna
Product owner:
• Digital engagement and acquisition squad
Responsi...
Česká Spořitelna
About us:
• Almost 200 years history
• 4,6 millions customers
• 10 000 of employees
• Part of ERSTE group...
Who is presenting
today Jakub Stech, DataSentics
Data Science architect in:
• DataSentics and Digi data team in CSAS
Respo...
DataSentics – European Data Science Center
of Excellence based in Prague
• Machine learning and cloud data
engineering bou...
Digital marketing spend optimization
Case study #1
ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER
SOURCES
BRAND AWARENESS
SALES (ONLINE AND OFFLINE)
CARE
Digital marketing ...
How to improve digital marketing spend
effectiveness
Business challenge:
10
ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER
SOURCES
BRAND AWARENESS
SALES (ONLINE AND OFFLINE)
CARE
Digital marketing ...
Adform is one of the world's largest private and independent
advertising technology companies and is best known for its
se...
A viewable impression is a standard measure of ad
viewability defined by the International Advertising
Bureau (IAB) to be ...
Decrease costs for visible seconds
Better specification of business challenge
WHEN
WHO
WHAT
WHERE
DEVICE
GEO
INTERACTIONS
DURATION
INT. COUNT
HOW MUCH
Adform Master Data…
Robot score (Adform definition)
Soft fraud score (own definition)
Multiple impressions
CTR (too low/high)
Visibility %
Vis...
• API
• BigQuery
• SFTP
• CSV
• Packages
• JSON
• Database dump!
• Web Pages
• …
Daily download, transformation and scorin...
Automated
download
MS Azure
CSAS Storing the data in
Data Lake
Results stored
into Data Lake
Scoring new
domains
Whitelist...
23% Desktop
28% Mobile
Increased effectivity of 1 EUR:
Omnichannel Banking Experience
Case study #2
23
Leading every client to prosperity
=
Data-driven advisory based on
clients needs and real-time situations
24
… is not easy
in bank
Customer-
centricity….
Low frequency of
interactions between a
client in offline channels
100 Things,
We touch our
phones 2,617
times a day,
says study
Around 100
sessions every
day…
Offline vs. Online
Typical CRM data
• Age/sex/address, policy history, policy
configuration, claim history, sales channel,...
1
Ad Interactions
(what the person is
interested in across
the internet)
Own website
interactions
Emailing /
SMS / Push
Si...
1
Ad Interactions
(what the person is
interested in across
the internet)
Own website
interactions
Emailing /
SMS / Push
No...
CASH LOAN MORTGAGESAVING ACCOUNT
Non-client & client behavior
Classic client profiles
Connecting the data on
individual cu...
Digital data based sales signals
Mortgage / Loan / Saving account / Investments …
What product?
OK, Loan… but what message?
Digital data based sales signals
Digital data based sales signals
MACHINE LEARNING
Digital data based sales signals
Call script: “Client is interested in taking loan. He
is planning wedding according to his activity on
internet.”
Digital ...
Standard classification models
(GLMs as benchmark, GBTs in production)
ML(lib) part
TEXT REPRESENTATIONS
CUSTOMER PATHS
NU...
Loans predicted by Adform data
4x Higher probability
Words with predictive power: loan, car, moto, wedding, …
ADFORM
WEB
DMP
CRM
CALL CENTRES
…
ADFORM
GA
CRM
Automated
downloadMS Azure
ČS
Storing the data in
Data Lake
Results stored...
Current architecture
50% Improvement
Benchmark:
Offline sales signals for mortages sales: call centre 10% sucess rate
Current:
Digi data enhanc...
Take aways
• No extra tech, just extend platforms with AI
models
• Connecting the data on individual customer
level is cru...
Thank you
DON’T FORGET TO RATE
AND REVIEW THE SESSIONS
SEARCH SPARK + AI SUMMIT
Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks
Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks
Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks
Upcoming SlideShare
Loading in …5
×

of

Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 1 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 2 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 3 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 4 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 5 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 6 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 7 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 8 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 9 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 10 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 11 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 12 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 13 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 14 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 15 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 16 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 17 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 18 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 19 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 20 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 21 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 22 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 23 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 24 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 25 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 26 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 27 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 28 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 29 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 30 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 31 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 32 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 33 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 34 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 35 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 36 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 37 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 38 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 39 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 40 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 41 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 42 Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Slide 43
Upcoming SlideShare
What to Upload to SlideShare
Next
Download to read offline and view in fullscreen.

1 Like

Share

Download to read offline

Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks

Download to read offline

Ceska sporitelna is one of the largest banks in Central Europe and one it’s main goals is to improve the customer experience by weaving together the digital and traditional banking approach. The talk will focus on the story of how in order to reach this goal Ceska Sporitelna created a new team focused on building use cases on top of a combined digital and offline customer engagement 360 powered by a Spark and Databricks-centric agile advanced analytics platform in the Azure cloud combined with a on-prem data lake. This talk will cover:

The customer engagement 360 vision powered by machine learning and the cloud
Deep dive into the use case of optimizing and personalizing programmatic ad buying on the individual user and ad placement level thanks to Spark MLLib and NLP on top of hundreds of millions of ad interaction data
Deep dive into the use case of supporting the seamless transition of the customer journey from digital to traditional offline channels
The approach to building the agile analytics platform and experience of adopting the cloud in a EU-regulated financial institution

Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks

  1. 1. WIFI SSID:Spark+AISummit | Password: UnifiedDataAnalytics
  2. 2. Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Petr Pluhacek (Ceska Sporitelna) Jakub Stech (DataSentics)
  3. 3. Who is presenting today Petr Pluháček, Česká spořitelna Product owner: • Digital engagement and acquisition squad Responsibilities: • Digital sales • Web platforms and chats • Digital marketing data and analytics • Online customer experience Contacts: • ppluhacek@csas.cz • www.linkedin.com/in/pluhacek
  4. 4. Česká Spořitelna About us: • Almost 200 years history • 4,6 millions customers • 10 000 of employees • Part of ERSTE group • Driver of innovation in the group • Undergoing agile transformation ČS Mission: “We are your lifelong guide on the path to prosperity, and in this way we contribute together to the prosperity of the whole country. When someone believes in you, you achieve more."
  5. 5. Who is presenting today Jakub Stech, DataSentics Data Science architect in: • DataSentics and Digi data team in CSAS Responsibilities: • Translate business problems for data science team • Personalizing user experience using data and machine learning approaches • Building and employing the analytical platform in cloud Contacts: • jakub.stech@datasentics.com • www.linkedin.com/in/jakubstech
  6. 6. DataSentics – European Data Science Center of Excellence based in Prague • Machine learning and cloud data engineering boutique • 50 data specialists (data science, data/software engineering) • Helping customers build end-to-end data solutions in cloud • Incubator of ML-based products • Partner of Databricks & Microsoft • Make data science and machine learning have a real impact on organizations across the world • Bring to life transparent production-level data science.
  7. 7. Digital marketing spend optimization Case study #1
  8. 8. ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER SOURCES BRAND AWARENESS SALES (ONLINE AND OFFLINE) CARE Digital marketing interactions
  9. 9. How to improve digital marketing spend effectiveness Business challenge:
  10. 10. 10
  11. 11. ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER SOURCES BRAND AWARENESS SALES (ONLINE AND OFFLINE) CARE Digital marketing interactions
  12. 12. Adform is one of the world's largest private and independent advertising technology companies and is best known for its seamlessly integrated DSP, DMP, and Ad Server. ADVERTISMENT
  13. 13. A viewable impression is a standard measure of ad viewability defined by the International Advertising Bureau (IAB) to be an ad which appears at least 50% on screen for more than one second.
  14. 14. Decrease costs for visible seconds Better specification of business challenge
  15. 15. WHEN WHO WHAT WHERE DEVICE GEO INTERACTIONS DURATION INT. COUNT HOW MUCH Adform Master Data…
  16. 16. Robot score (Adform definition) Soft fraud score (own definition) Multiple impressions CTR (too low/high) Visibility % Visibility time SEZNAM.CZ; BID 0.2 ; NOVINKY.CZ; BID 0.6 ; IDNES.CZ ; BID 0.1 ; BLESK.CZ ; BID 0.4 ; … Metrics of domain model
  17. 17. • API • BigQuery • SFTP • CSV • Packages • JSON • Database dump! • Web Pages • … Daily download, transformation and scoring jobs 4+ BLNs rows in 12months! Data sources
  18. 18. Automated download MS Azure CSAS Storing the data in Data Lake Results stored into Data Lake Scoring new domains Whitelists, bid multipliers, cookies lists, blaclicks, … Automated update SEZNAM.CZ; BID 0.2 ; NOVINKY.CZ; BID 0.6 ; IDNES.CZ ; BID 0.1 ; BLESK.CZ ; BID 0.4 ; … Fully automated pipeline
  19. 19. 23% Desktop 28% Mobile Increased effectivity of 1 EUR:
  20. 20. Omnichannel Banking Experience Case study #2 23
  21. 21. Leading every client to prosperity = Data-driven advisory based on clients needs and real-time situations 24
  22. 22. … is not easy in bank Customer- centricity…. Low frequency of interactions between a client in offline channels
  23. 23. 100 Things, We touch our phones 2,617 times a day, says study Around 100 sessions every day…
  24. 24. Offline vs. Online Typical CRM data • Age/sex/address, policy history, policy configuration, claim history, sales channel, … • Static, mostly long-term behaviour • Facts and transactions • Well structured, easy to process with traditional tech Digital „footprints“ • Ad interactions (wider internet behaviour), web interaction (own sites), mobile apps, external/partner data, … • Dynamically changing, reflecting short and long-term needs • Uncertainty, fragments about interests, behaviour, lifestyle • Enormous data (B+ ads, M+ visits of website…), messy, unstructured, changing interfaces
  25. 25. 1 Ad Interactions (what the person is interested in across the internet) Own website interactions Emailing / SMS / Push Siloed customer behaviour data Classic client profiles 4) Limited customer experience 2) Missing environment for data analytics and machine learning Classic CRM / data processes Branches & sales networks Transactional data / product data Callcentrum data / call logs Digital campaign management tools Classic campaign management tools 1) Missing connection between digital and CRM 3) On-premise environment is lacking customer data from digital Mobile app interactions Client portal / Internetbanking interactions 3rd party data, voice, text, image, geo data, etc. Digital engagement (3rd party) Offline vs. Online
  26. 26. 1 Ad Interactions (what the person is interested in across the internet) Own website interactions Emailing / SMS / Push Non-client & client behavior Classic client profiles Automatic optimization, personalization of customer journeys Machine learning Your Customer Engagement 360° Platform (CSAS) Classic CRM / data processes Automatic signals for classic channels Branches & sales networks Transactional data / product data Callcentrum data / call logs Connecting the data on individual customer level Digital campaign management tools Classic campaign management tools New opportunities Higher efficiency Mobile app interactions Client portal / Internetbanking interactions 3rd party data, voice, text, image, geo data, etc. Digital engagement (3rd party) AI-augmented Customer Engagement 360°
  27. 27. CASH LOAN MORTGAGESAVING ACCOUNT Non-client & client behavior Classic client profiles Connecting the data on individual customer level
  28. 28. Digital data based sales signals Mortgage / Loan / Saving account / Investments … What product?
  29. 29. OK, Loan… but what message? Digital data based sales signals
  30. 30. Digital data based sales signals
  31. 31. MACHINE LEARNING Digital data based sales signals
  32. 32. Call script: “Client is interested in taking loan. He is planning wedding according to his activity on internet.” Digital data based sales signals
  33. 33. Standard classification models (GLMs as benchmark, GBTs in production) ML(lib) part TEXT REPRESENTATIONS CUSTOMER PATHS NUMERIC REPRESENTATIONS
  34. 34. Loans predicted by Adform data 4x Higher probability Words with predictive power: loan, car, moto, wedding, …
  35. 35. ADFORM WEB DMP CRM CALL CENTRES … ADFORM GA CRM Automated downloadMS Azure ČS Storing the data in Data Lake Results stored into Data Lake AI monitoring dashboard Model (AI) re-training Scoring new data using existing model (AI) Whitelists, bid multipliers, cookies lists, blaclicks, … Automated update Transforma tion Results (bids, audiences, …) Fully automated pipeline
  36. 36. Current architecture
  37. 37. 50% Improvement Benchmark: Offline sales signals for mortages sales: call centre 10% sucess rate Current: Digi data enhanced signals for mortages sales: call centre 15% sucess rate
  38. 38. Take aways • No extra tech, just extend platforms with AI models • Connecting the data on individual customer level is crucial • Clear business specifications and convincing results are essential
  39. 39. Thank you
  40. 40. DON’T FORGET TO RATE AND REVIEW THE SESSIONS SEARCH SPARK + AI SUMMIT
  • tushar_kale

    Nov. 3, 2019

Ceska sporitelna is one of the largest banks in Central Europe and one it’s main goals is to improve the customer experience by weaving together the digital and traditional banking approach. The talk will focus on the story of how in order to reach this goal Ceska Sporitelna created a new team focused on building use cases on top of a combined digital and offline customer engagement 360 powered by a Spark and Databricks-centric agile advanced analytics platform in the Azure cloud combined with a on-prem data lake. This talk will cover: The customer engagement 360 vision powered by machine learning and the cloud Deep dive into the use case of optimizing and personalizing programmatic ad buying on the individual user and ad placement level thanks to Spark MLLib and NLP on top of hundreds of millions of ad interaction data Deep dive into the use case of supporting the seamless transition of the customer journey from digital to traditional offline channels The approach to building the agile analytics platform and experience of adopting the cloud in a EU-regulated financial institution

Views

Total views

724

On Slideshare

0

From embeds

0

Number of embeds

0

Actions

Downloads

29

Shares

0

Comments

0

Likes

1

×