While business intelligence (BI) providers have carefully been testing the premise that structured and unstructured information collection and analytics should be merged, none have taken the plunge–until now. That's changed with the recent announcement by Business Objects that it plans to acquire Inxight Software Inc., a text mining software provider. Inxight offers software that allows companies to glean information from such unstructured text sources as e-mails, a variety of documents, notes fields and Web content, which according to some estimates makes up 80% of all organizational data. Using linguistic analysis and algorithms, Inxight's software is able to detect patterns, trends and such identifiable items as people, places and things and tag and quantify those items for analysis. "This is the first time that a pure BI vendor has placed a bet on unstructured information–and particularly unstructured analytics," notes Forrester Research analyst Matt Brown who follows BI software providers.
The acquisition marries Business Objects' structured BI capabilities with Inxight's unstructured data mining software and, observes Forrester's Brown, essentially validates the proposition behind convergence of BI applications relying on both structured and unstructured data. With this new approach, a customer segmentation analysis for a marketing campaign in the future could combine structured BI data such as balances, transactions and demographics together with customer comments and complaints made via e-mail or voicemail. "It's one thing being able to present a dashboard that shows a particular structured data report next to a set of search results around an event," says Brown. "It's another thing to be able to analyze those results. Inxight is really the leader around analytics."
Aside from such traditional BI strongholds as financial analysis and marketing, the merger of structured and unstructured could lead to improved applications in the areas of fraud detection, risk management, quality improvement and other information intensive initiatives, remarks Brown. By automating manual discovery work, BI and text analytics should reduce the burdens of compliance and fraud detection. The combination is also expected to help discover and manage communications risk in highly regulated industries, such as healthcare or financial services, where organizations may need to intercept potentially dangerous communications before they leave the enterprise.
Recommended For You
For this reason, Inxight has already done well in selling to such regulated sectors. Additionally, the company has developed a niche business around warranty claims analysis, remarks Brown. But while text analytic tools historically have had success in niche markets, they have not reached a mainstream audience. Brown points out that unstructured data software provides its greatest value when it can be plugged into a broader platform for doing analysis, and "Business Objects has a big captive audience of BI buyers, typically of power users, and has the potential of building a bigger market around these applications."
So the potential–although as yet unrealized–is there. "The other thing that's interesting is what sort of opportunities Inxight brings to Business Objects from their OEM [original equipment manufacturers] business," adds Brown, who says that most of Inxight's success has come from embedding its software in other enterprise search vendors' products. Inxight gets embedded into other products like document management systems and content management systems, he notes. "All of that is good news for Business Objects."
The deal between Business Objects, with $1.3 billion in revenues and dual headquarters in Paris and San Jose, Calif., and Inxight Software, a privately owned company based in Sunnyvale, Calif., with an estimated $25 million in sales, is expected to close in July.
© 2025 ALM Global, LLC, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to [email protected]. For more information visit Asset & Logo Licensing.
