“Every strike brings me closer to the next home run” – Babe Ruth

I just got through the April 2011 Harvard Business Review that'd been sitting in my to-read basket for months. The issue revolves on failure – how to understand it, learn from it, and recover from it. The articles made for informative reads with a few particularly pertinent for BI.

I like the "spectrum for reasons of failure" in Amy Edmondson's “Strategies for Learning From Failure.” At the blameworthy end are deviance, inattention and lack of ability. Praiseworthy failures, on the other hand, follow from uncertainty surrounding future events, systematic hypothesis testing and experimentation. The more complex and unknown the optimal course of action, the more suitable are scientific efforts to learn that often lead to failure. Edmondson espouses building a learning culture that detects failures early, deploys evidence-based failure analysis, and promotes experimentation with high failure endeavers. "Those that catch, correct and learn from failure before others do will succeed. Those that wallow in the blame game will not."

Rita Gunther McGrath takes Edmondson's arguments a step further in “Failing By Design.” She argues for a corporate culture that fosters "intelligent failure" to promote organizational learning. IF can be particularly fruitful in high failure environments such as faced by venture capital firms, where a 20 percent hit rate with investments is considered a success. Noting the ascendance of Google's add-based search model, McGrath opines "Without all that trial and error, it's highly unlikely that Google would have built the algorithm-based juggernaut so familiar today."

McGrath outlines seven principles for "Putting Intelligent Failure to Work". Among the most prominent are conversion of assumptions into knowledge as early in an initiative as possible. "... make sure you and your team are open to revising them as new information comes in. The risk is that we have a tendency to gravitate towards information that confirms what we already believe ..."

Two other principles are adaptations of the well-worn sales bromide – lose early. McGrath espouses failing both quickly and cheaply. Her last, and perhaps most important principle, involves codifying and sharing lessons learned through failure with the organization. Among the techniques she's seen as successful "are mini postmortems as a project proceeds, checkpoint reviews as key thresholds are reached, and after-action review meetings at the project's conclusion."

Even with all the organizational wisdom to be derived from the failure articles, I still found “Why Don't Learders Learn From Success,” the most thought provoking. Authors Francesca Gino and Gary Pisano argue that problem with success is that it often ultimately leads to failure by "hindering learning at both the individual and organizational level."

The authors note three cognitive impediments derived from research in behavioral decision-making that militate against learning from success. The first involves fundamental attribution errors that link success to skill and talents, while relegating failure to environmental factors and random events. In a business setting, this translates responsibility for the good times to people, the business model and strategy. The bad times, in contrast, are generally seen as due to a non-cooperative business climate or simply bad luck. I wish a had a dollar for every technology colleague who touted his company's success in the 90s, while explaining away lesser performance over the last 10 years to burst bubbles and an unfriendly market!

The second bias that can trip up persistent success is overconfidence, which often inspires a false sense of assurance that the current way is the right one. The third impediment is failure-to-ask-why. Just as companies need to do post-mortems on their failures, they should also systematically investigate the causes of success. "Yet it is all too common for executives to attribute the success of their organizations to their own insights and managerial skills and ignore or downplay random events or external factors outside their control."

BI can play an important role in supporting the unbiased attribution of success and failure in business that leads to a healthy learning culture. As discussed in recent blogs, BI should borrow from the performance measurement discipline of investment science that uses designs/analytics to differentiate beta from alpha performance. Beta has to do with factors, such as economic climate and overall business conditions, that are common to all competitors in an industry – a rising/falling tide that lifts/lowers all boats. Alpha, on the other hand, measures the more company-specific performance factors such as people, model and strategy that differentiate businesses above and beyond what is happening to everyone else. What's Apple been doing recently, for example, to distinguish its performance from other tech giants who're treading water? 

The PM mandate for BI will begin to thrive as businesses rise above human biases and adopt an evidenced-based culture that embraces predictive analytics and the experimental method. It is at this point that they'll be able to distinguish alpha from beta and become true learning organizations.