#1NLab16 - Fortifying Big Data & Making Insights Count
1 like538 views
The document discusses the evolution and significance of data-driven decision-making, particularly in sports and business contexts. It contrasts big data, which focuses on large, quantitative datasets, with thick data, which emphasizes qualitative insights and human-centered understanding. Through various case studies, it highlights the importance of integrating both types of data to optimize performance and meet organizational goals.
#1NLab16 - Fortifying Big Data & Making Insights Count
- 1. NOV 2-4, 2016 Fortifying Big Data and Making Insights Count
- 2. NOV 2-4, 2016 Ben Magnuson @1N_Ben
- 3. NOV 2-4, 2016 The Data Driven Organization
- 4. NOV 2-4, 2016 Decision Making is Changing § B2B seeing substantial growth in marketing technology and data § Companies more likely than ever to state their decision making is data-driven § Leaders in companies “substantially outperforming their competitors” are 166% more likely to make decisions based on data* *Source: IBM’s Analytics: A Blueprint for Value
- 5. There Is More to Data Than Numbers § Big Data – extremely large sets of data, often unstructured, used to find patterns § Thick Data – ethnographic and qualitative measurements that introduce stories of value
- 6. NOV 2-4, 2016 What is measurable isn’t the same as what is valuable. – Tricia Wang, Ethnography Matters “
- 7. The Data Takeover in Sports § The 2000s saw a data revolution take over baseball § Talent evaluation positions once held by former players were going to 20-something Harvard graduates § “The people that we’re hiring and other baseball teams are hiring, we’re competing with the Apples and the Googles of the world.” – Billy Beane, interview in WSJ on 9/21/15
- 8. Sabermetrics, A History of Data in Baseball 1977: Bill James publishes the first edition of The Bill James Baseball Abstract 1981: Stats, Inc founded to record detailed score sheets for every baseball game 2001: Bill James introduces Win Shares in his The New Historical Baseball Abstract 2003: Michael Lewis publishes Moneyball on the use of analytics by the Oakland As to find value
- 9. Sabermetrics, A History of Data in Baseball 2003: Bill James offered position in Boston front office 2004: Red Sox win world series 2009: Baseball- Reference adds Sean Smith’s version of WAR 2011: Moneyball the movie, starring Brad Pitt, debuts
- 10. NOV 2-4, 2016 Interesting Color Full Image
- 11. NOV 2-4, 2016 Mike Trout – 10.3 WAR Miguel Cabrera - Triple Crown Winner 2012 MVP Race
- 12. NOV 2-4, 2016 When Spreadsheets Aren’t Enough § The success of baseball analytics led to similar movements in other major sports § While this introduced great new insights into game strategies and team performances, there was a struggle to introduce similar player valuations like WAR
- 13. Two Advanced Approaches § Football Outsiders § DVOA – A method of evaluating teams, units or players. It takes every single play during the NFL season and compares each one to a league-average baseline based on situation. § When we say, "In 2014, Marshawn Lynch had a DVOA of 23.1%,” what we are really saying is “In 2014, Marshawn Lynch, playing in Darrell Bevell’s offensive system with the Seattle offensive line blocking for him and Russell Wilson selling the keeper when necessary, had a DVOA of 23.1%.“ § Pro Football Focus § Grades – PFF takes a different approach and tries to assess value by observing every individual play and assigning a grade from -2 to +2 § Tries to address the variability of different schemes and teammates by judging their execution in each individual play
- 14. NOV 2-4, 2016 PFF Grading Process Step 1: Every player in every play is given a grade from -2 to 2 Step 2: A second scout grades the same play Step 3: Reconcile the differences Step 4: We verify by sending tape to the Pro Coach Network
- 15. NOV 2-4, 2016 Providing Context
- 16. NOV 2-4, 2016 Comparing Big and Thick Data § Relies on machine learning § Reveals insights with a particular range of quantified data points § Loses resolution § Relies on human learning § Reveals the social context of connections between data points § Loses scale *Source – Tricia Wang, Ethnography Matters Big Data Thick Data
- 17. NOV 2-4, 2016 Measuring Value - Review § Identify goals § Establish key metrics § Measure results § Optimize performance
- 18. Complementing Big Data: Case Studies § Target has been at forefront of big data analysis and particularly predictive analytics § Earlier in 2016, the Minnesota Star-Tribune reported that Target CEO Brian Cornell was visiting customers homes “hoping to understand such things as consumers’ food choices, fashion trends and shopping habits”
- 19. Case Study: Grocer § Harvard Business Review published a story on the CEO of a high- end grocery chain in Europe that was seeing it’s sales decline § Customers were no longer shopping there as frequently or buying as much § Conducted a survey, where customers said they were willing to pay more for quality § If that was case, why weren’t they going to the high end store?
- 20. NOV 2-4, 2016 Case Study: Grocer § They ran an anthropological study where they embedded recorders with their customers for 2 months § Found that the idea of the family dinner was frayed § Customers were shopping less because often on the go, and shopping at smaller stores that offered quicker meals § Revamped stores to have “mini-stores” within. Sales improved.
- 21. NOV 2-4, 2016 Practical Applications of Thick Data
- 22. NOV 2-4, 2016 Classic Monochromatic Full Image
- 23. NOV 2-4, 2016 Takeaways § Digital data insights is an active experience § Find your KPIs and organize your analytics to measure the success of your goals § If the measurements are telling you the what, but not the why, use thick data techniques such as interviewing and usability testing
- 25. NOV 2-4, 2016 Fortifying Big Data and Making Insights Count