This is not the OpenBI Forum article that was scheduled for February. Immediately after the holidays, I planned my first three columns for 2007, actually outlining the first and second. After posting "Strategy and BI" at the end of January, I was about to start writing the next column when I received my February Harvard Business Review (HBR) in the mail. While browsing that edition, I both made a decision and came to a conclusion: 1) I would postpone my planned column to write on several important points I'd seen in the February HBR; and 2) I would recommend that HBR be on the short reading list of every serious business intelligence (BI) practitioner.
There are probably not many readers who, like me, would tout HBR as an exemplary BI periodical. Certainly in contrast to DM Review, HBR has little to say directly about BI, analytics, integration and data warehousing. And though sharing a business orientation with BI Review, the quarterly magazine that "presents actionable information on the strategic use of technology to enable organizations to operate more effectively," HBR does not focus on applied technology per se, but rather touches intelligence and technology indirectly as enablers of business strategy and processes.
HBR authorship and topics cut a wide swath. Some articles are written by academics, some by business leaders, some by business strategists, some by consultants/business process experts and others by contrarians/provocateurs. There are case studies chronicling the travails of companies and industries in addition to ethnographies on good and bad leadership. There are articles on strategies and business processes that provide taxonomies for conceptual thinking - and form the foundation for measuring performance as well as seminal articles from great business thinkers that are timelessly referenced. There are articles on hypothetical business situations and how best to handle them in addition to autobiographical articles from business executives detailing how they dealt with failure and success. There are articles linking business to historical and social contexts as well as those that borrow from other disciplines such as psychology and political science to propose optimizations of business processes. Then there's my personal favorite, the HBR List in Brief - a series of very terse columns offering challenges and provocations to business - often from outside, nonbusiness writers.
HBR's influence on BI is generally subtle and indirect. Probably 50 percent of published articles offer some commentary on BI, though readers in search of BI will have to look hard and will likely not find references to buzz concepts such as data integration or master data management. Instead they will see more business-focused intelligence terms such as open source, hypothesis, measurement, evaluation, optimization, modeling, performance management, experiments, prediction, analytics, cause and effect, strategic linkages, and lead and lag indicators. Readers are thus challenged to make the connection from business to technology and intelligence - translating business concepts to operational BI equivalents. For those who can make that difficult transition, who can cross strategic boundaries in their thinking, there's no shortage of BI wisdom in HBR.
Over the past year of BI discourse, the OpenBI Forum has noted several recurring themes that are pertinent to the current HBR. The first is the need for BI practitioners to expand their learning horizons outside the narrowly focused BI media in search of quantitative and methodological optimizations for business problems. Second are the potential salutary effects of the open source business model and software for BI. Third is the notion of strategy as hypothesis, delineating relationships between lead indicators and lag indicators or cause and effect. And fourth are the concepts of validity and design in performance management as the business attempts to "prove" its strategy. We will examine each of these concepts in turn with illustrations from the current HBR.
"Algorithms in the Attic" from The HBR List by Michael Schrage of the MIT Media Lab, is a quantitatively focused take-off on "Rembrandts in the Attic" - a book that details the potential of underutilized patents which might be commercialized by willing buyers. Schrage notes that a centuries-old theorem on matrices applied to software enabled Google's search engine formula for automatic ranking. He cites other examples from mathematical science as well, including an algorithm that measures statistics of a queue, used initially to optimize automated teller machines and now applied to customer service management. Also acknowledged is the enormous potential of "Monte Carlo" random number generation techniques that can stress-test thousands of iterations of hypothetical business plans using probability theory. Schrage quotes MIT colleague Richard Larson, "There are many clever ideas my students worked on decades ago that in today's networked environment would not be an academic exercise but a real business opportunity."
Another article from The List, "In Defense of Ready, Fire, Aim," by Clay Shirky, offers sageful commentary on a not-well-understood benefit of the open source software movement. While acknowledging the success of open source, Shirky opines that "Open systems are a profound threat not only because they outsucceed (sic) commercial firms but also because they outfail them. They grow not in spite of failure but because of it." From this vantage point, open source is a critical enabler of business experimentation, since the cost of failure is amortized across so many actors. This minimal cost of failure can thus encourage businesses to reach for potential high reward/high risk successes. For these experiments to be galvanizers of business, however, the activities must truly be open and independent of any one firm's capacity to direct.
"Understanding Customer Experience" by Christopher Meyer and Andre Schwager is a good example of an HBR conceptual business process article written by practitioners that stimulates strategic thinking about BI. Meyer and Schwager contrast "CEM" with traditional customer relationship management (CRM). CRM captures what is known about customers - purchases, service requests, inquiries, returns, etc., while the softer CEM solicits responses to customer encounters with the company. CEM is about what the customer thinks and feels about the company, while CRM is concerned with what the company knows about the customer. As a consequence, CEM is leading or predictive of future performance; CRM, on the other hand, is about lag indicators or the outcomes themselves. It is not much of a stretch to link CEM to CRM as cause and effect: if a company provides a good customer experience (CEM), it is likely to have success managing the customer relationship (CRM). Indeed, the authors depict this relationship in a customer rating graphic where customer satisfaction (CEM) is plotted on the x axis and billed revenues (CRM) on the y axis, with forecast revenues represented by circle sizes on the graph.
Meyer and Schwager also shed light on the CEM intelligence process by the taxonomy they propose to track CEM information. Their delineation of past, present and potential experience patterns, along with persistent, periodic and pulsed data collection frequencies and alternative analysis/collection methodologies, lays the foundation for a comprehensive CEM intelligence strategy. An enterprising BI analyst can operationalize this foundation with technology and analytics to begin building the CEM BI platform.
Experts looking for innovation in their approach to BI are encouraged to take a hard look at HBR as a source of inspiration. For those able to make the difficult transition from business to technology, the payback may be well worthwhile.
1. Michael Schrage. "Algorithms in the Attic." Harvard Business Review. February 2007.
2. Clay Shirky. "In Defense of 'Ready, Fire, Aim'." Harvard Business Review. February 2007.
3. Christopher Meyer and Andre Schwager. "Understanding Customer Experience." Harvard Business Review. February 2007.