Time was in the BI world, centralization and consolidation were the names of the game. And the quest to get a single version of the truth was the guiding light. That pursuit is still the reality for many organizations.
But businesses have begun to realize that enterprise standard BI projects need to be flexible and agile in order to react and adapt to information requirements that seem to change with ever-increasing speed, according to Forrester Research VP and principal analyst Boris Evelson. He recently published these thoughts in a report, “Trends 2011 And Beyond: Business Intelligence.”
“In BI, you can’t define the requirements,” Evelson says. “By the time you define it, there are already new requirements, and your project plan falls apart at your first deliverable.”
Enter Agile BI. It’s more than simply using agile development methodologies for BI projects. It is, according to Forrester, an approach that combines processes, methodologies, organizational structure, tools and technologies that enable strategic, tactical and operational decision-makers to be more flexible and more responsive to the fast pace of changes to business and regulatory requirements.
There are, however, few BI initiatives at organizations that currently meet that definition, Evelson says. So how can organizations start adopting an agile BI methodology?
The first step may be a familiar one but it’s as important as ever: create a top down mandate. Commit to not using spreadsheets, for instance. Top executives have to say that they can’t really run their business without analytics, Evelson says, and evangelize that message through the organization.
One way is by doing “BI on BI.” Link the usage of the BI tools to departmental performance. Every vendor’s product has built in usage analytics, and it usually falls to the IT department to do the tracking. It shouldn’t, Evelson says. The best-case scenario is for a business manager to track business-unit usage in employee quarterly or yearly evaluations.
When Agile BI Doesn’t Work
Once that top-level mandate is established, Evelson says, there need to be guidelines or rules as to what belongs under the umbrella of agile and what doesn’t. That’s because not every application has a need for speed. A large portion of BI applications still need to follow the traditional approach, such as those used for back-office functions or in highly regulated industries.
Flexibility and ease of use mean that sometimes the answer won’t be 100 percent correct. And any instance where 100 percent accuracy is not acceptable, you can’t use agile and you can’t use self-service, Evelson notes. ”There’s that tug of war,” he says. “Find where traditional approaches apply and where IT needs to loosen the reigns and not insist on a traditional BI app and allow end-users to get the work done.”
Evelson defines four characteristics of agile BI tools: They need to automate BI processes and steps as fully as possible. They need to be unified, addressing different BI use cases. They need to be pervasive and allow strategic, tactical and operational decision-makers to act quickly. And they need to be limitless, allowing analysis on information without borders or restrictions.
Just How Agile Are SAP’s BI Tools?
SAP is pretty weak in automation, Evelson says. He adds that the vendor is doing OK with unification, pervasiveness and limitlessness.That said, every one of the big four vendors is in the same boat when it comes to delivering on agile BI: No one covers every facet of functionality. Niche areas are still dominated by the pure plays.
HANA is definitely a step in the right direction of agile BI, Evelson notes. But HANA is not competing with QlikView, PowerPivot or TIBCO Spotfire when it comes to zero reliance on IT.
“HANA does make it easier to change and adapt the data model,” he says. “Still there’s very heavy IT involvement. It’s a little bit more in the agile fashion, but 1 or 2 on a scale of 10.”