For years, business intelligence and analytics tools seemed out of the reach of midmarket users. Complex and costly systems that required hardware, software, licensing and special skills were beyond the budgets and in-house IT talent pools of most midsize companies. But nowadays, with vendors offering lighter versions of their products and the rise of software as a service (SaaS), BI has become accessible to companies that previously might not have been able to afford it.
"The industry evolution toward self-service BI and SaaS BI is having a big impact on adoption in companies of all sizes," says James Kobielus, an analyst at Forrester Research. "Companies that want to do BI can adopt and provision it more rapidly at a lower cost" -- often pay-as-you-go -- "and make it available to a broader range of users with less IT involvement and lower capital costs," he says.
However, experienced users warn that even at its simplest, BI still requires a concerted effort.
"BI isn't about a single tool that does everything. It's an ongoing business strategy that needs to have resources dedicated to it," says Tony Murabito, CIO at Cubist Pharmaceuticals in Lexington, Mass. Drawing on his nearly 10 years of BI use, he warns that thinking of BI as a set-and-forget proposition or rushing an implementation could jeopardize enterprise plans and make the effort an IT liability.
"Between 70% and 80% of corporate business intelligence projects fail," according to Gartner analyst Patrick Meehan.
To avoid a disastrous outcome, midsize companies need to be disciplined and create a targeted development model, invest in training and staffing for both IT pros and end users, seek input and buy-in from their companies' various business units, be open to changing their data-collection processes and chart a path for growth.
SMBs embracing BI
A March 2010 Forrester report found that BI implementations by small and midsize businesses are on the upswing, even though SMB BI use lags behind BI use at larger organizations. Of 921 North American and European IT software decision-makers that Forrester surveyed, 21% of those at organizations with up to 999 employees said they have deployed BI technology already and 22% said they are expanding and upgrading their implementations. Other respondents from small and midsize businesses said that they have near-term plans for BI: 18% said they plan an implementation in the next 12 months.
[Check out Computerworld's upcoming BI conference, being held Sept. 18 to Sept. 20 in Phoenix.]
As BI penetrates further into the SMB market, IT should strive to use it to do more than just reporting, says Thornton A. May, a Computerworld columnist and the author of The New Know: Innovation Powered by Analytics. "They should aim to... use the power of analytics to change the game and be competitive," says May, who is also dean of the IT Leadership Academy at Florida State College in Jacksonville.
A study that May conducted in support of his book showed that only 16% of companies in general are considered world class in their use of BI and the majority, 61%, are average. To be above average, he says, companies have to take advantage of everything that BI encompasses today: dashboards, visualization, social networking and the wisdom of crowds, and mash-ups that blend multiple data resources (internal and external) for a unique perspective.
Finding the right fit
Dan Grosz has found that doing BI correctly at the SMB level has its unique challenges and opportunities. "You tend to have a smaller staff and budget, but for the same reasons its probably easier to try newer, more innovative approaches," says Grosz, vice president of information systems at VIP Parts, Tires and Service in Lewiston, Maine. He says he was attracted to BI for its ability to analyze data across sales, inventory, financial and payroll systems.
He worried early on about the traditional "waterfall" approach he had heard that BI requires, where IT had to have a complete vision of user needs up front -- once the only game in town for software development. He knew that would be impossible in his organization, because users weren't familiar with BI and wouldn't know what they wanted the system to do until they had access to it.