The Truth's In The Analysis

We hear a lot about getting a "single version of the truth" -- it's been a catchphrase of data warehousing for years. Usually, it's meant pulling together a pool of clean, consistently formatted and accurate data. That's no small task, for sure. But getting to the "truth" lies in the way data is analyzed as much as it resides in the data itself.

Big companies and other large organizations, many of which are awash in business intelligence tools from multiple vendors, are coming to understand that the "truth" can be thwarted when separate departments, analyzing the same data, lack consistency on the front end. You don't have to be studying for a doctorate in philosophical relativism to know that the truth can get a little squishy when it's reported different ways.

That was the message -- though not in so many words -- that Meta Group analyst Kurt Schlegel had for Business Intelligence Pipeline when he spoke with us about companies' ongoing and laborious efforts to standardize their BI front end. "You want a single version of the front end, analytic and reporting tools," he said. "Otherwise you can get people from different departments, who don't necessarily trust each other, sitting around the conference table with conflicting analyses."

Ask any major BI software manufacturer, and you're likely to hear that their sales efforts these days involve convincing existing customers to standardize on their BI tools, in addition to cold-calling new prospects. You can hear it in the words of BI vendor CEOs -- Information Builders' Gerry Cohen and Business Objects' Bernard Liautaud are two recent examples -- when they argue that their applications are the ideal standard around which to build a consolidated, integrated BI strategy.

Though their arguments conflict with each other, those CEOs are right to focus on standardization. Companies are consolidating the make and model of reporting and analysis applications they use. The inflated costs and inefficiencies created by too much application variability demand it. So does that need to get at the truth, which is useless when it's ambiguous, and a key to success when it's consistent.