Five Attributes to a Successful Big Data Strategy
2 likes1,600 views
The document outlines five key attributes for a successful big data strategy, emphasizing the importance of a structured approach that includes business case development, architecture planning, skill development, governance, and proof of concept (POC) initiatives. It highlights common challenges, lessons learned from case studies, and the high failure rates of big data projects, stressing the need for a clear business value and comprehensive planning. It further encourages organizations to pursue strategic analytics to leverage big data effectively for improved performance and cost savings.
Five Attributes to a Successful Big Data Strategy
- 1. Five Attributes to a Successful Big Data Strategy Bill Busch SSA | Enterprise Information Solutions CWP Twitter: @agilebibill
- 2. Perficient is a leading information technology consulting firm serving clients throughout North America. We help clients implement business-driven technology solutions that integrate business processes, improve worker productivity, increase customer loyalty and create a more agile enterprise to better respond to new business opportunities. About Perficient
- 3. • Founded in 1997 • Public, NASDAQ: PRFT • 2013 revenue $373 million • Major market locations throughout North America • Atlanta, Boston, Charlotte, Chicago, Cincinnati, Columbus, Dallas, Denver, Detroit, Fairfax, Houston, Indianapolis, Los Angeles, Minneapolis, New Orleans, New York City, Northern California, Philadelphia, Southern California, St. Louis, Toronto and Washington, D.C. • Global delivery centers in China, Europe and India • >2,100 colleagues • Dedicated solution practices • ~85% repeat business rate • Alliance partnerships with major technology vendors • Multiple vendor/industry technology and growth awards Perficient Profile
- 4. BUSINESS SOLUTIONS Business Intelligence Business Process Management Customer Experience and CRM Enterprise Performance Management Enterprise Resource Planning Experience Design (XD) Management Consulting TECHNOLOGY SOLUTIONS Business Integration/SOA Cloud Services Commerce Content Management Custom Application Development Education Information Management Mobile Platforms Platform Integration Portal & Social Our Solutions Expertise
- 5. Bill Busch SSA | Enterprise Information Solutions CWP • Bill leads Perficient's enterprise data practice and specializes in business-enabling BI solutions. • Responsibilities: • Executive data strategy • Roadmap development • Delivery of high-impact solutions that enable organizations to leverage enterprise data • Bill has spent the last 15 years in executive leadership roles in business intelligence, data warehousing, information/data architecture and analytics. His most recent achievement is as visionary and leader of Perficient’s Big Data Lab, an environment that enables Perficient to conduct state-of-the art Big Data research and development. Speaker
- 6. Agenda • Challenges with Big Data • Big Data Strategy • 5 Attributes of a Big Data Strategy – Business Case – Architecture – Skill Development – Governance – Big Data POC • Questions and Answers
- 7. 69% Higher revenue per employee 20% Companies realize cost savings from tool rationalization Why Approach Big Data Strategically? A Strategic Approach Will: • Align the company stakeholders • Communicate value creation • Get IT to stop playing and start creating business value with Big Data technologies • Establish a complete people, process, and technology aligned plan • Prioritize business cases to those that attainable and create real business value • Drive changes to delivery and governance that typically limit Big Data value • Define Big Data’s role within an enterprise data architecture BUT…….BUT……. 95% Failure rate of Big Data projects 77% High performing companies will strategically leverage analytics vs. only 33% of low performing companies
- 8. Big Data Business Cases • Business Focused Benefits – Optimization – Prediction • IT Business Case – Benefits • Cost savings /avoidance • Additional capability – Analytics and Data Discovery – Data Warehouse Augmentation – Data Hub/Data Lake • Consider using a layered business case • Do not use a business case that can easily solved with an existing DW Case Study Situation Role of big data was not defined within the organization. Financial transaction processing company chose a parameterized reporting that was solved using traditional EDW at minimal cost Results Role of big data was not defined within the organization was delayed because the business case Lessons Learned • Choose a use case that cant be easily solved with a traditional system • Established industry use cases are easiest to support • Do not put all your Big Data eggs in one business case
- 9. Business Case: Plan For Benefits Analysis • Benefits analysis is a process by which business benefits are quantified (usually in $) • Upfront ROI on big data cases is difficult to specify • Benefits analysis can be the key to continued funding • Specify a process and responsibility for Benefits Analysis in your strategy
- 10. Setting Expectations Case Study Situation Google analyzed over 500 million web searches a day and correlated this to disease data for flu. Results Google’s overestimated the number of flu occurrences for the between 2011-2013 by a factor of nearly two. Lessons Learned • Predictive modeling is applied science and is difficult • Many times, you will need more data • Understand changes in source data • Cost savings tend to come from larger implementations • Business cases built on analytics must realize the scientific research component • Studies build on each other • Understanding why a model has failed can have value • Test & learn cultures lend themselves to big data analytics • Providing a capability that is leveraged by people • Focus the organization on delivering a tool/capability vs a business process delivering ROI
- 11. Skill Development “It's all to do with the training: you can do a lot if you're properly trained.” Queen Elizabeth II • Strategy should realistically access the skills of the organization to leverage the Big Data environment • More than tool based training – do you have the data scientists and statisticians in-house • Consider establishing analytical user- groups to drive organizational learning • Plan to develop IT’s delivery and support skills – Includes training on new delivery processes
- 12. Architecture “The mother art is architecture. Without an architecture of our own we have no soul of our own civilization.” Frank Lloyd Wright Specify the complete architecture Ingestion/Extraction/Job Control Data Storage Areas Refinery & Data Preparation Security Metadata Analytical, Data Discovery, BI, Model Execution Tools HW Platform (Best of Breed vs. Appliance) Hadoop Distribution /Targeted Release
- 13. Architecture Data Ingestion Case Study Situation Large financial services company wanted to time to detect fraud. It was taking weeks and sometimes months to source new data. Results Developed a custom, metadata driven solution that allowed new data feeds to be added by just modifying metadata. This reduced time to deliver data feeds to less than a week. Lessons Learned • The light transformation requirements of Big Data ELT allow for metadata configured ELT. • Significant opportunity to reduce costs & quickly create business value. Perficient has seen a pattern of companies not addressing: – Hand-coding point to point data integrations of Sqoop, Flume, Pig, Map Reduce, Java, etc. is repeating the sins of the past – Metadata configured ingestion is not that expensive and quick to develop – Comprehensive view of data integration • CDC of source systems • Transformations to standardize data format • Supportability of the final system • Integration with current batch – Do not forget network infrastructure
- 14. Architecture Data Storage Options Plan for the Big Data environment to consist of many different data storage areas Analytics ExtractsAnalytics ExtractsAnalytics Extracts Consolidated Data Delta Data Discovery and Analytics Sandbox Analytics Writeback Standardized Reference Data Scrubbed Data Receiving Zone Processed Data (Future) Refinery Jobs Data Publishing Message /HL7 Store HL7 Scraping Analytics and Data Discovery Data Warehouse Data Lake
- 15. Governance • Governance must be addressed at the onset of a Big Data project • Delivery and support processes must change to enable • Security -- Need to know vs. need not to know • Data governance must be exception based • User classification (tools and data access) • Create save swimming pool for data scientists • Involve business! “Those who expect to reap the blessings of freedom must, like men, undergo the fatigue of supporting it.” Thomas Paine
- 16. POC Imperative Case Study Situation A Fortune 100 company conducted a Big Data POC. The major work effort was to load over 100+ tables chosen by IT. Results • Project ran behind when data quality issues were not considered of timelines and resources. • Prioritized business cases were not identified due to the pure IT focus of the project Lessons Learned • Set up POC to drive architecture standards & business case prioritization • Focus scope of POC to predefined use cases Consider a POC as a part of the strategy: – Work through architectural details/challenges – Provide a plan based on real-world experience – Test BI/Data Discovery Tools – Provide sizing information – Business use-case validation/prioritization
- 17. Conclusion • Big Data is a significant investment • A comprehensive plan will go a long way to assuring success
- 18. As a reminder, please submit your questions in the chat box. We will get to as many as possible.
- 19. Daily unique content about content management, user experience, portals and other enterprise information technology solutions across a variety of industries. Perficient.com/SocialMedia Facebook.com/Perficient Twitter.com/Perficient
- 20. Thank you for your participation today. Please fill out the survey at the close of this session.