Big Data and MDM altogether: the winning association
4 likes2,906 views
The document discusses the integration of Master Data Management (MDM) with big data, emphasizing its importance for businesses to leverage data assets effectively. It outlines the challenges and opportunities presented by big data, including the need for improved data sharing, governance, and analytics for enhanced decision-making across various industries. Additionally, it highlights case studies demonstrating successful MDM and big data applications in sectors like retail, insurance, and hospitality.
Big Data and MDM altogether: the winning association
- 1. MDM and Big Data : the wining association Jean-Michel Franco Innovation Director Copyright 2007, Information Builders. Slide 1
- 2. Business & Decision : a global Consulting and System Integration company Founded in 1992 Revenue 2012 : 221,9 M€ 2 500 employees 16 countries 5 expertise recognized by global independent industry analysts BI / EPM CRM Digital Marketing BI & EPM Services Europe MarketScope Magic Quadrant for CRM Services Interactive Design Agency Overview, Europe, 2013 EIM MDM / BRMS / Search / ECM CONSULTING Consulting, Change Management , enterprise software design, training 2
- 3. Business & Decision and MDM MDM and Information Management specialist Proven iterative Design & implementation methodology Innovation, engagement , expertise Master Data Management dedicated practice 80 consultants globally Strong market research and partnership relationship management Closely linked to other group expertise (CRM, BI, e-bus) Proven Agile MDM approach, closely linked to the business with engagement on costs and time to deliver Versatile consultants with BI and IT skills 50 % delivered through a fixed priced approach 90% of the business targeting large accounts End to end engagement Deep knowledge of technology solutions Expertise across industries and MDM domains 3
- 4. Are you ready to get value from your data assets ? lessons learned from Amazon.com. Know your customer Source : Faber Novel Expand your service portfolio Value Your ecosystem
- 5. From data retention to data sharing New organizational and technological paradigm Knowledge is Power Knowledge is Profit • Retention and data silos • Heterogeneity and « best-of-breed » • Decentralization et autonomy • Vertical organization • Opaqueness • Communication and global sharing • Mutualization • Centralization/ federation and collaboration •Horizontal Organization • Transparency Some answers are in your data – if only you could take advantage of them 5
- 6. Business needs are becoming more and more precise and urgent Leveraging your data assets are a must, not an option, to tackle current Business challenges IT system convergence & consolidation Deinterleaving of IT systems for deregulated markets Governance, Risk and Compliance Time to market Customer Knowledge Information Hub Business Glossary Master data Repository Data Quality Integration Extended enterprise
- 7. From data integration to Information Governance From a siloed, IT driven model (Data Management)… IT …to a federated, and shared responsibility model (Information Governance) Lines of businesses Business define their need and use information within enterprise applications IT designs, implements, runs and manages Ongoing conflicts on data quality and relevancy, lack of autonomy, slow time to market... Métiers Line of businesses define their needs, administrate the information, document them, sometimes mashes them up and contribute to their relevancy and maintenance IT accompanies, controls, rolls out, delivers “as a service”, secure and manages
- 8. Master Data Management 101 Master data management (MDM) is a comprehensive method of enabling an enterprise to link all of its critical data to one file, called a master file, that provides a common point of reference. When properly done, MDM streamlines data sharing among personnel and departments. In addition, MDM can facilitate computing in multiple system architectures, platforms and applications. The Master Data Management aims to develop the processes, organization and tools to collect, reference, manage and share the Master Data and links between them across organizations, people, processes and systems The party/ persons The products Les The places Lieux The organizations The Shared assets Clients Citizen Products User Services Points de Countries vente Charts of accounts Entrepots Warehouse Stores Supplier 33% 44% 3% Business rules SLA Territories 21 % Subsidiaries Assets Configurations Employees Operational and legal organization Partners Rates Real estate Standard codifications 8
- 9. MDM isn’t self sufficient : disciplines of Enterprise Information Management Document, eliminate data redundancies Improve and certify data quality and relevance Transfer information, secure it, trace it Expose the information and make it accessible « as a service » Transfer information, secure it, trace it Master Data Management, Meta Data Management Data Quality Information Governance Enterprise Information Integration, Info Lifecycle Management, Data Loss Prevention Data Services SOA, enterprise search, BI, Enterprise portals
- 10. What are the data types to consider ? Describe Data types : Analytical data Meta Data Transactional data use Master and reference data Example: Transactional data Meta-data Customer_Id First name Name Product Supplier Date Amount 92584789 Ann B. TXF98 Dell 24/12/2013 650 $ 92584789 Ann B. AXC54 Maped 24/12/2013 2,44 $ 92584789 Ann B. TRE56 Playmobil 24/12/2013 …. Master Data Analytical data scoring RFM CLV 129,36 $
- 11. From data integration to information governance : Where to start? Design the platform Define the roles Engage the programs, domains per domain Product Customer Organization Sites Platform design -> Needs assessment workshop -> Proof of concept -> Roadmaps and budgets Orga. blueprint -> Information Governance competency center -> Data Stewardships -> Service Center Roll out « Fast delivery » -> Iterative modeling -> Data mappings -> Data quality maps
- 12. Big Data : définition Big Data is high volume, high-velocity and highvariety information assets that demand cost-effective innovative forms of information processing for enhanced insight and decision making From “the 451 Group” et Gartner Source : Wall Street Journal “The challenges include capture, curation, storage, search, sharing, analysis and visualization..” (wikipedia) Inspired by Wikipedia
- 13. Big Data is the long tail of information management “Today, we sold more books that didn’t sell at all yesterday than we sold today of all the books that did sell yesterday” (amazon.com, via Josh Petersen & Wikipedia) BI as we know it Popularity - Information sourced from internal IT Systems - information provisioned in batch mode - Static information modeling BI as we’d like it to be BI as we know + externally sourced data + « just in time » data + un-structured, semi-structured data + information as it comes (schema less) Available information
- 14. Data Warehousing on steroids for better pricing, planning and customer facing policies Retailers pioneered Enterprise Data Warehouses, especially for market basket analysis. But retailers are now pressured to get more value from their data assets, to deepen and sharpen analytical capabilities and make them « actionable » . Dynamic and micro-segmented pricing policies Personalization of the offers for loyalty programs Adjustment of offers to demand by locations Consistency across channels (e-commerce, stores , drive)
- 15. Transparency for supply chain efficiencies and superior customer services In France, 25% of the water that flows into the distribution networks is lost due to leaks or frauds ; This accounts for 2,4 billions Euros per year. (*) Digital channels and internet of objects open new opportunities to bring transparency into the supply chain, and deliver superior customer service (*) Source : SIA conseil Real time information on water flows and quality A value added service for consumers and institutions Detection of leaks as they occur along the network and at the end of the chain A common engagement between supplier and customers in terms of sustainability Automation of the collection process for billing
- 16. Innovation in insurance industry and agritech Innovate with data centric new offers A start-up to manage risks and insure farmers through online services that predicts how climate affects crop yield and personalized insurance offers. A predictive platform that combines hyper local weather data with agronomic yield data down to the field level, all undergirded by weather simulations. * acquired by Monsanto on October 2013 for 950 millions $ Trusted advizorship through online personalized services to help farmers better predict manage the climate conditions Services can be deployed globally without limits, allowing to tackle new markets Claims management processes fully automated from observation to payment Huge opportunities to transform best practices in agriculture and climate management
- 17. Innovating in the insurance industry : Fraud Management Apply the principle of Credit Scoring for claims management. Integrate unstructured and semi-structured data to highlight inconsistencies in claims declarations. Push the analytics on the field, close to the customer and when an where the claim is declared . Success rate of investigations : from 50 to 85% 25% of claims are closed immediately at the first step, against 4% before -> better service for honest customers Scoring drives the claims process and improves its efficiency (Actionable analytics)
- 18. Elevating the good old Data Warehouse with Big Data : Searching for your « long tail » Extend the founding principles of Data Warehousing and Information Management for more: On line transactional processing Immediacy Data Warehouse Big Data Precision Create new source insights through new data flows Sensors, Internet of things Business Intelligence a analytics Agility External data, shared data, open data Capture and share unstructured data Documents, rich content… Public data sourced from social networks and internet
- 19. Big Data by industry and by activity IBM : the real world use of Big Data
- 20. What do you need to manage your Big Data?
- 21. MDM and Big Data : The wining association 21
- 22. Why Big Data needs MDM? Example : Digitalizing the ordering process to Santa Claus Entity extraction Data Quality management Reconciliation with master data Data enrichment Customer _Id First name 92584789 Ann B. 92584789 Ann 92584789 Ann …. Name Product Supplier Date Amount TXF98 Dell 24/12/2013 650 $ B. AXC54 Maped 24/12/2013 2,44 $ B. TRE56 Playmobil 24/12/2013 129,36 $ 22
- 23. Why MDM needs Big Data ? Ex.: From Customer Data Integration to an active and real time 360° Customer View Master Data Contact Center Transactional Data Face to face Interactions SMS/Mail/Chat… Mobile Applications 3rd party Data Platform Customer Journey Data Customer Data Platform ROI Analytical data (stores, agencies…) Web Site *Source : H Sorensen 23
- 24. Innovation in the hospitality industry: real time recommendations • From attention to intention economy • Test offers and challenge their efficiency on an ongoing basis • Provide a consistent quality of service across channels • Better manage recommendations across the brands, together with interactions with customers, promoters, detractors… Improve click rate (+43%) and transformation rate Ability to test new offers, and to stop or improve them as soon as needed Ability to listen and react to promoters and detractors in social networks Personalized offers and personalized interactions Federation of customer knowledge across brands to adapt to organizational changes
- 25. Innovation in the banking industry: Multi channel customer journeys • Acquire/Enrich Customer knowledge • Recommend the next best offer according to the context • Manage end to end purchasing journey from intent to payment • Monitor real time the relevance and success of offers Personalized interactions with unknown internauts based on their click stream Personalized interactions with known customers based on their profile and current /past on line and offline journey Ability to track the timeline of customer interactions both offline and online Definition of new customer segments based on analytics around customer journeys
- 26. Next step: Towards « predictive/prescriptive apps » : Next generation of apps that can anticipate user need and recommend 26