Facts up! Teach your team to use BI
|Michael Lewis decided to write the book “Moneyball: The Art of Winning an Unfair Game” (W.W. Norton & Co, 2003) to understand how the Oakland A’s -- one of the “poorest” teams in Major League Baseball -- could win so many games and continually make the playoffs.|
He found that the Oakland A’s, led by general manager Billy Beane, trust only certain types of statistics to find highly effective players at bargain-basement prices -- players other teams overlook because they are too slow, can’t field or don’t have the “look.” Guided by his Ivy League “sabermetricians,” Beane recruits mainly college players with a record for “getting on base” -- players who draw a high percentage of walks.
The book is a morality tale for modern-day business. Like Major League Baseball, many firms today cling to intuition and tradition as guideposts for making strategic and tactical decisions. Ten or 15 years ago, executives did not have much data to guide them, so going with their “gut feel” was making the best of a sorry situation.
Today, however, the majority of large- and mid-sized firms have implemented business intelligence (BI) systems. Most executives and managers can now access integrated data in a timely fashion to help them make key decisions. The question now, however, is: Have they made the cultural leap from intuition to data as a guide for making decisions?
Best practices for managing change
Many BI project managers discover the hard way that implementing a business intelligence system is less about technology than change management. BI is often a revolutionary agent that threatens to change the way people work, think and interact with their bosses and peers.
The Scotts Company, a maker of lawn and garden products, is on a major quest to fully embrace “fact-based decision-making,” according to Deb Masdea, director of business information and analysis. Prior to the implementation of an SAP BW data warehouse, “people were making decisions on gut feel,” said Masdea. “Our executives now say, ‘Where’s the data to back up this decision?’”
But the transition to a fact-based decision-making culture didn’t happen overnight. Many users -- particularly middle management -- resisted the change. Masdea had to address a host of concerns that most firms face when implementing BI on an enterprise scale. Among the concerns: users don’t trust the data; users worry that the data will threaten their position in the company; users fear they might not be good at analyzing data; users are technically illiterate; and users are too busy.
To ameliorate the resistance to change, Masdea and her team employed several strategies that TDWI considers best practices for managing cultural change wrought by BI:
1. Executive-level support. Scotts’ entire SAP implementation, which included BW, was driven by its CEO.
2. Executive-level training. Masdea trained every executive personally, even though she knew many wouldn’t use the system.
3. Executive-level usage. The key to getting mid-management support was getting executives or their surrogates to use the system. Once executives began referring to BW data and reports, mid-management followed.
4. Super users. Scotts created a network of 100 “super users,” business users in various departments who train and support their peers on the system. These super users know user needs well, and understand the data and how to analyze it.
5. Focused training. BI training emphasized how users should “think and work differently,” said Masdea. “We walk them through the analytical process online, which is different from [looking] at data on paper.”
6. Go cold turkey. As Scotts provided reports or content via BW, it eliminated other sources of data, including legacy reports and operational systems.
7. Data quality expectations. Besides reconciling data with source systems, Scotts spent a lot of time educating users about the “new data” and why it looks different from their existing models.
8. Evangelization. Masdea’s full-time job is to evangelize “fact-based decision-making.” She and her team oversee the super-user network and meet continually with executives and managers to help them better leverage the data warehouse when making decisions.
Rather than rely on intuition, Scotts’ users now employ data and intuition synergistically. This aligns with a recent TDWI survey in which 46.5% of respondents said business users in their firms use “intuition, validated by data” to make decisions, and 44% use “data, supported by intuition.”
Just as Billy Beane of the Oakland A’s has fomented a revolution within Major League Baseball, project managers need to be cognizant of the turmoil they’re creating when implementing BI. They can’t afford to focus solely on technology issues. Even the most perfect BI system will flop unless project managers overcome individual and organizational resistance to change.
Wayne W. Eckerson is director of education and research for The Data Warehousing Institute, where he oversees TDWI's educational curriculum, member publications, and various research and consulting services. He has published and spoken extensively on data warehousing and business intelligence subjects since 1994.