The Secret to BI Success: Take a Business-Driven Approach
- By James E. Powell
- September 10, 2013
[Editor's Note: Barbara Wixom has been investigating how companies deliver value from BI. In her keynote presentation at the TDWI World Conference 2013 in Boston (October 20-25, 2013), she'll discuss Business-Driven BI: How to Create Results that Matter. In this interview, we explore the benefits of the business-driven approach, the best practices of enterprises employing it, and how an enterprise can get a BI program back on track.]
BI This Week: What exactly do you mean by business-driven?
Barbara Wixom: When I talk with BI leaders about how their initiatives are going, most can articulate some kind of bottom-line benefit that BI is generating for the organization, but few think that BI is really delivering on its full potential. Those who believe that BI is thriving and exceeding expectations at their companies typically have business-driven BI programs in place. I refer to a BI initiative as business-driven when it enjoys heavy business user engagement and aligns closely with the company's business model and strategy (or mission).
What benefits can BI initiatives expect if they take a business-driven approach to BI?
The typical benefits from BI range from tangible, tactical benefits, such as time savings for analysts and better resource allocation to intangible, strategic ones like the support for a new strategic direction. When BI initiatives are business-driven, they realize many more types of benefits than an average program. Also, their portfolio of benefits includes those that really move the company forward, such as faster speed to market, greater innovation, and strategic enablement.
What are the best practices that you typically observe in companies with business-driven BI?
There are three kinds of practices that companies with business-driven BI usually have in place. First, their development practices are user-oriented. This can mean agile methodologies, co-location, or even having formal BI delivery groups within business units, such as a marketing analytics group that reports to the marketing department -- either directly or via matrix. These practices ensure that business requirements are formulated and honed by the business, translated accurately, and prioritized in the right way.
Second, their BI governance effectively addresses data security and controls concerns. The best practices companies also have a business value component to their BI governance whereby value from BI is proactively prioritized, tracked, communicated, and even marketed internally. The latter can include a range of actions, such as company videos, newsletters, knowledge-sharing portals, and "dog and pony shows." All of these actions create awareness about what business value that BI is producing -- and what it can produce.
Third, BI-driven initiatives are typically found in companies that have developed a large number of "hybrid" roles and people. By hybrid, I mean roles and people with a healthy combination of business and technology skills. Hybrids can be incredibly business-savvy IT people, technically adept business users, or a group of employees that collectively displays both kinds of skills. Hybrids are critical because they possess the ability to work with data -- to identify insights that matter to the organization -- and then to understand the company well enough to act on insights in the appropriate way.
The combination of user-oriented development practices, value-oriented governance, and hybrid people and roles all keep BI business-driven and result in bottom-line and sustained value to the organization.
What can a company can do to revive a BI program that has lost the attention of the business?
Most BI programs hit snags along the way. In fact, a few of the best BI programs that I ever encountered were heavily impacted (negatively!) by mergers and acquisitions. If you find that BI needs to be "revived" within your company, you can lead with the business or with IT.
One option is to lead with the business. This entails identifying a key business leader who can champion a BI initiative for some important business purpose and leveraging the initiative to revive awareness and excitement regarding BI. Sometimes it takes a great deal of charisma and salesmanship on the part of BI or IT to bring a new business leader on board -- but this approach can be quite effective.
Another option is to lead with IT. There are several exciting emerging technologies -- visualization, mobile devices, social analytics are examples -- that IT can use to grab the attention of BI users. For example, one company I studied used a BI app delivered via iPads to entice their users into engaging with the BI program. I must warn you that with this approach the selected emerging technology must solve a critical business need. Otherwise, users will lose interest in the "new shiny object" and the BI team will be worse off than when they started.
Fundamentally, BI needs to be useful, easy to use, and -- if you have a particularly tough user community to engage -- fun. You will need to get into the heads of users and really understand them. If they can make easier work decisions by doing things an old way -- or by working around your BI program solutions -- then you don't have a chance of revival.
Does business-driven BI mean that IT doesn't matter?
Absolutely not! In fact, IT plays many critical roles in establishing an organization's BI capabilities. IT is responsible for building a high-quality, usable, and integrated data platform -- then ensuring the platform is exploited by the business users. IT needs to manage the elicitation of business requirements and subsequent conversion of requirements into easy-to-use, useful, and fun BI tools. IT is a master at process. Therefore, IT needs to take the lead in ensuring that data moves from insight into action, which is then institutionalized at the business process level. IT needs to manage emerging and disruptive technology and communicate opportunities to the business community when appropriate. These include technologies and techniques such as those popularized by the recent focus on big data. Believe me, IT matters.
Where does big data fit into all of this?
Big data is a chapter of BI that offers new technologies, techniques, challenges, and opportunities. That being said, I don't see many of the fundamentals for BI success changing the underlying success factors or the drivers of business value. Regardless of the kind of data, type of architecture, and analytics techniques at work, BI programs should strive to be business-driven.