Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks

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

展开查看详情

1.WIFI SSID:Spark+AISummit | Password: UnifiedDataAnalytics

2.Creating an Omnichannel Banking Experience with Machine Learning on Azure Databricks Petr Pluhacek (Ceska Sporitelna) Jakub Stech (DataSentics)

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.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.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.DataSentics – European Data Science Center of Excellence based in Prague • Machine learning and cloud data • Make data science and machine learning engineering boutique have a real impact on organizations across • 50 data specialists (data science, the world data/software engineering) • Bring to life transparent production-level • Helping customers build end-to-end data data science. solutions in cloud • Incubator of ML-based products • Partner of Databricks & Microsoft

7. Case study #1 Digital marketing spend optimization

8.Digital marketing interactions ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER SOURCES BRAND AWARENESS SALES (ONLINE AND OFFLINE) CARE

9. Business challenge: How to improve digital marketing spend effectiveness

10.10

11.Digital marketing interactions ADVERTISMENT WWW.CSAS.CZ INTERNET BANKING OTHER SOURCES BRAND AWARENESS SALES (ONLINE AND OFFLINE) CARE

12.ADVERTISMENT 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.

13.

14.

15.

16.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.

17. Better specification of business challenge Decrease costs for visible seconds

18.Adform Master Data… WHEN WHAT DEVICE WHO WHERE DURATION GEO INTERACTIONS HOW MUCH INT. COUNT

19.Metrics of domain model Visibility time Visibility % SEZNAM.CZ; BID 0.2 ; CTR (too low/high) NOVINKY.CZ; IDNES.CZ ; BID 0.6 ; BID 0.1 ; BLESK.CZ ; BID 0.4 ; Multiple impressions … Soft fraud score (own definition) Robot score (Adform definition)

20. Data sources • API Daily download, transformation and scoring jobs • BigQuery • SFTP • CSV • Packages • JSON • Database dump! • Web Pages 4+ BLNs rows in 12months! • …

21. Fully automated pipeline Automated Automated update download MS Azure Whitelists, bid SEZNAM.CZ; BID 0.2 ; CSAS multipliers,BID NOVINKY.CZ; 0.6 ; Storing the data in Scoring new Results stored cookies IDNES.CZ ; lists,BID 0.1 ; Data Lake domains into Data Lake blaclicks, BLESK.CZ ; …BID 0.4 ; …

22.Increased effectivity of 1 EUR: 23% Desktop 28% Mobile

23. Case study #2 Omnichannel Banking Experience 23

24. Leading every client to prosperity = Data-driven advisory based on clients needs and real-time situations 24

25. Customer- centricity…. … is not easy Low frequency of in bank interactions between a client in offline channels

26. We touch our phones 2,617 times a day, Around 100 says study sessions every day… 100 Things,

27.Offline vs. Online Typical CRM data Digital „footprints“ • Age/sex/address, policy history, policy • Ad interactions (wider internet behaviour), configuration, claim history, sales channel, web interaction (own sites), mobile apps, … external/partner data, … • Static, mostly long-term behaviour • Dynamically changing, reflecting short and • Facts and transactions long-term needs • Well structured, easy to process with • Uncertainty, fragments about interests, traditional tech behaviour, lifestyle • Enormous data (B+ ads, M+ visits of website…), messy, unstructured, changing interfaces

28.1 Offline vs. Online Own website Ad Interactions Mobile app Client portal / 3rd party data, Digital interactions (what the person is interactions Internetbanking voice, text, Digital campaign interested in across interactions image, geo engagement the internet) data, etc. management tools (3rd party) 2) Missing environment for data analytics and machine Siloed customer behaviour data learning 4) Limited customer 1) Missing connection experience between digital and Classic client profiles CRM 3) On-premise Classic environment is lacking CRM / data customer data from processes Branches Transactional Emailing / Callcentrum digital Classic campaign & sales networks data / product SMS / Push data / call management tools data logs

29.1 AI-augmented Customer Engagement 360° Own website Ad Interactions Mobile app Client portal / 3rd party data, Digital interactions (what the person is interactions Internetbanking voice, text, Digital campaign interested in across interactions image, geo engagement the internet) data, etc. management tools (3rd party) Machine Automatic learning optimization, personalization of Your Customer Non-client & client behavior customer journeys Engagement Connecting the data on 360° Platform New opportunities individual customer level Automatic signals (CSAS) Classic client profiles for classic channels Classic Higher efficiency CRM / data processes Branches Transactional Emailing / Callcentrum Classic campaign & sales networks data / product SMS / Push data / call management tools data logs