Jeanne Harris is executive research fellow and a senior executive at the Accenture Institute for High Performance in Chicago. She directs the global research agenda on information, technology, analytics and talent at the Institute. In the IKNS program, she is an instructor in the Business Analytics class.
[These are my notes from Columbia University’s 2013 summer residency program for its Masters of Science in Information and Knowledge Strategy. Since I’m publishing them as soon as possible after the end of a session, they may contain the occasional typographical or grammatical error. Please excuse those. To the extent I’ve made any editorial comments, I’ve shown those in brackets.]
- Lessons from Peter Drucker. Jeanne Harris had the privilege of working with Peter Drucker early in her career. Here are some of the data-related lessons she learned from him:
- Look for patterns. Analytics help reveal patterns, which in turn reveal what is happening below the surface of the organization.
- Don’t settle for just a single insight. Businesses are complicated animals so it is rare for a single insight to explain everything about an organization. Look for several insights that shed light on the business as a system.
- What gets measured gets done. Things don’t get implemented unless people and processes are measured and monitored.
- The Value of Analytics. Early studies indicated that companies that implemented ERP systems rarely received any value from the system. When Jeanne Harris and Tom Davenport dug deeper into the research, they discovered that reality was a bit more nuanced that the research indicated. In fact, while some companies found ERP systems disappointing, the companies that considered analytics to be a key part of their business strategy received great value from their ERP systems. The main differentiator was a company’s approach to and use of data.
- Competing on Analytics. People have been collecting data (and performing analysis) for centuries. What’s different is knowing how to use data analytics to (1) differentiate your company or product, (2) execute your business strategy, or (3) build data into your products. A perfect example of a company that competes on data analytics is Netflix. Data helps Netflix be a huge disruptor in its space. For example, Netflix initially used data via its Cinematch algorithm to help (1) customers choose films to watch; (2) Netflix manage its inventory; (3) Netflix negotiate its deals with movie studios (which are typically less data driven); (4) Netflix manage customers who tend to hold onto movies longer than average. (Netflix manages them by not sending them the first batch of new release movies.) In short, data analytics give Netflix a huge competitive edge.
- How do you Build Analytical Capability in Your Organization? It isn’t enough to use data to execute your strategy or achieve operational efficiency. In fact, the more effective use of analytics is to support better decision making. Ultimately, you need to use analytics to stay ahead of your competitors. This requires cultural change within the organization. Harris told us how Progressive Insurance Company used data to differentiate the safe motorcycle owners from the reckless ones. Next they offered preferential rates to the safe customers and pushed the bad risks to their competitors. In another data-driven effort, Progressive offered customers the opportunity through its “Snapshot” program to install a device on the vehicles that collected data on their driving habits. This data was then analyzed to determine each participating customer’s likelihood of having an accident. With this information, Progressive could offer a more appropriate insurance rate.
- Big Data. One definition of Big Data is “data too big to be analyzed by SQL.” Harris noted that we have only begun to scratch the surface of Big Data’s potential to realize major business outcomes. The technology is finally up to the task, but we don’t yet have enough people able to derive insights from that data. As a result, there is incredible demand for trained data analysts. The other problem is that many decision makers in organizations are essentially innumerate and, therefore, don’t understand how to use Big Data and the insights it can reveal. Harris quoted Tom Davenport who was concerned that “a lot of Big Data is low analytics” and, therefore, people are missing the great insights that could be attained by using more sophisticated analysis.
- What Skills are Necessary? Harris recommends the following skills:
- You need to be numerate. This means not only understand numbers, but also statistics.
- You must understand scientific methodologies (e.g., forming a hypothesis, A/B testing, the scientific method generally).
- You must understand what the data are really telling you.
- You must understand how the data relate to your strategy.
- You must know how to translate the data analysis into operations/execution and, ultimately, how to use the data to change the organization itself.
- Analytics Makes s Difference. Big Data isn’t only for Big Business. Even simple analytical applications can help small businesses and nonprofits. For example, Harris and her team helped a theater company use their sales data to materially improve their ticket subscriptions. For organizations large and small, there now is a drive to data monetization.