Oracle this month announced a new version of Oracle Endeca, its set of enterprise data discovery tools and self-service business intelligence applications. Oracle Endeca Information Discovery 3.1 includes added features for data mashups, BI dashboard design, unstructured data analysis, data management and integration with Oracle Business Intelligence Enterprise Edition. We spoke to SearchOracle expert Mark Rittman, co-founder of BI and data warehousing consultancy Rittman Mead, about the self-service BI trends highlighted by the news.
What is the main business problem or problems facing companies now when it comes to data discovery tools, enterprise search and analytics?
Mark Rittman: Getting all the disparate data sources -- which usually aren't organized neatly into tables or Excel files -- into a central place, without spending months on just the data modeling part. You don't know at this early stage what data is important or not and how it really relates to each other -- so the problem that needs solving is how to collect data together quickly, of all different types, loosely join it together and impart some meaning out of it.
Why is it important that business users be able to do business analysis on their own, with less help from IT?
Rittman: Because by the time IT finishes its work, the opportunity has often passed. Much gets lost between the idea in a business user's head and the often-large-scale implementation of it by IT. So the less barriers there are between the opportunity, the idea and getting something in front of the users, the better -- at least at the start, when it's more of a germ of an idea than a fully-forged business process.
More on Oracle Endeca
Check out Endeca's place in the market of data discovery tools
See how Oracle's Exalytics appliance ties into Endeca
What concerns does IT have when it comes to self-service data discovery tools and analytics?
Rittman: That there's no governance over data quality and there's a profusion of reports. Systems that are departmental have a habit of ending up with IT to support, and without proper report and data governance at the start, what can be an initial success can turn into a data nightmare as you get contradictory numbers and incorrect assertions.
What do you think is the next frontier for data discovery tools?
Rittman: Empowering the end user to discover and visualize data themselves -- making it properly self-service and supporting all relevant data sources. It's nothing revolutionary -- just [further] down the same power-user empowerment road as now.