Experimental feature

Listen to this article

Experimental feature

Decision making might be no more difficult today than ever – but the kind of information that workers rely on is changing dramatically.

In the past, decision makers mostly weighed so-called structured data that resides neatly in spreadsheets and databases, in part because that information was easier to store and manage.

Now, employees are balancing this data with unstructured information embedded in e-mails, presentations, instant messages and blogs. Much is stored locally and considered personal.

But finding relevant information across the internet, corporate networks, one’s computer and even the occasional memory stick is proving a challenge. In fact, employees may lose up to a quarter of their time just trawling for information, say analysts at IDC, the market intelligence company.

“People are asking why they can find information on the internet but not inside the company,” says Roberto Solimene, Google European enterprise director.

This frustration explains why many employees are taking matters into their own hands and downloading free desktop search tools such as Google Desktop and Microsoft’s Windows Desktop Search. These programs index all of a computer’s files and enable users to sift through their hard drive just as they would the web.

Of course, it is unlikely that a free software download can trump the multi-million dollar business intelligence systems companies purchased to help them analyse vast amounts of data. Desktop does nothing more than look for keywords and can easily overload users with masses of irrelevant data.

“Google Desktop is fine when looking at ad hoc data sitting all over the place. But as a real business search tool, it is woefully inadequate,” says Clive Longbottom, analyst at Quocirca, the UK technology research company. “What we want is a find engine, not a search engine, something with more contextuality.”

And, as often happens when maverick users introduce consumer-grade applications into the enterprise environment (think instant messaging or Skype), desktop search applications come with some big business risks. In one case, a large US company saw its Microsoft Exchange server crash for a day after an employee’s desktop search application tried to index e-mails stored in shared folders on the server.

Tammy Alairys, who heads the Accenture Information Management Services practice, says most companies have not addressed the need for desktop search, leading employees to install the applications on their own. “I haven’t seen many companies standardise on desktop search. Most don’t regulate its use and that could be a problem from a support standpoint,” she says.

Paul Hanley, a security consultant at Cable & Wireless, says desktop search tools also present a security risk for companies. “Trojan horse software could embed itself into the desktop search and send off confidential data without the user knowing,” he explains. Or more humbly, the search tools could index data on the corporate network and give the user unauthorised access to sensitive information.

Google does offer an enterprise version of its Desktop that enables centralised management and provides some security enhancements. But the desktop is just one element of an enterprise search solution, says Lee Phillips, director of intelligence solutions at Fast Search & Transfer.

“Enterprises are looking for a search platform that integrates all of its information assets, that ties together structured and unstructured data, and that gathers information from the user’s machine,” he explains. This requires back-end search engines that can tap into data in CRM, ERP and other business-class systems.

“If you don’t bring that together, you haven’t solved much,” says Mr Phillips.

Google also sells enterprise search platforms that range from cheap SME-grade machines to more robust engines capable of indexing up to 15m pages of information.

JPMorgan uses a Google machine to power its intranet and Apple uses one for its e-commerce site.

But whatever enterprise search looks like today, it is only the first step towards increasingly personalised and instantaneous information availability, says Nat Friedman, vice president of Linux desktop engineering at Novell.

For several years, Mr Friedman has been working on ways for computers to learn what information is important to an individual user in a given context and automatically to proffer data from the web or stored locally that might be useful. A first generation of this technology is built into the software company’s new Linux desktop suite, released last month.

“In the future, each user will have a personal information space, with relevant information ordered around that individual. This will mean web search and non-web search will come together,” he says. So, for instance, a web search would automatically take into account information stored locally on the computer to make the results more relevant and accurate.

“If computers did this well, it would be supremely useful,” he says.

Get alerts on News when a new story is published

Copyright The Financial Times Limited 2019. All rights reserved.
Reuse this content (opens in new window)

Comments have not been enabled for this article.

Follow the topics in this article