Dylan Tweney
Published Work

How to Beat Corporate Alzheimer’s

For as long as people have been keeping records, they’ve struggled to find efficient ways to file their work. Ancient Assyrians, who scratched records on clay tablets, stored documents in pigeonholes in the walls of libraries, writing a list of each room’s contents on the wall — a kind of primitive
Dylan Tweney 5 min read

For as long as people have been keeping records, they’ve struggled to find efficient ways to file their work. Ancient Assyrians, who scratched records on clay tablets, stored documents in pigeonholes in the walls of libraries, writing a list of each room’s contents on the wall — a kind of primitive database. While modern technology has advanced well beyond clay tablets, databases still require that information be stored away in precisely defined fields, the digital equivalent of those pigeonholes.

Once data gets stored, how do you find it again? If it’s in a structured database with lots of neatly organized fields, it’s not hard. But in most companies, there are valuable stores of data tucked away on employees’ hard drives, on Web servers, and on the company intranet: Word documents, PowerPoint slides, spreadsheets, e-mail, and so on. Such files represent, by some estimates, as much as 80 percent of a company’s information assets. Good luck finding much of use in this heap of unstructured data.

To mine these documents, many companies are turning to so-called knowledge management (KM) technologies. If the term arouses skepticism, your instincts are good: Large consulting firms touted half-baked KM software as a panacea for enterprise information management in the late 1990s. Early KM software required employees — or a cadre of librarians — to carefully organize and annotate information before it could be managed. That approach proved too labor-intensive and expensive to be worthwhile.

Today, KM is making a comeback on the strength of better solutions — namely, the humble search engine. In the last few years, search engines, originally developed to comb through the sprawling expanses of the Web, have become remarkably effective at finding bits of data wherever they lie. That’s because new technologies largely automate the processes of categorizing and summarizing documents. For instance, Autonomy  (AUTN), a company based in Cambridge, England, offers an engine that provides a statistical analysis of word frequencies and concepts and automatically groups related documents. You may need a Ph.D. to understand the technical details, but not to grasp the benefits, especially when you’re delivered a handful of highly relevant documents instead of 32,873 “related” ones.

The current market leaders for enterprise KM software are Autonomy and Verity — industrial-strength search engines equipped with the bells and whistles that corporations want, such as the ability to integrate databases and CRM systems and to restrict access to sensitive documents. But don’t discount the technologies offered by Web search engines, such as AltaVista, Google, and Inktomi  (INKT), all of which offer corporate products. Which solution works for your company will depend on your infrastructure and how well-suited a search technology is to your data types — something you may be able to determine only with hands-on tests.

The impact of KM software can be profound. For example, Deloitte Consulting uses a KM system based on a search engine from Verity  (VRTY), along with an Oracle  (ORCL) database and content management software from BroadVision  (BVSN), to provide its nearly 20,000 employees with access to a repository of more than 250,000 documents. Deloitte chief information officer Larry Quinlan says the $2 million system provides an “essential” means of sharing information about consulting practices. “Deloitte is all over the world,” he says. “Without [the KM system], we just wouldn’t be able to function.”

Large-scale KM initiatives such as the one employed by Deloitte can cost millions of dollars and may take 18 to 36 months to bear fruit, according to French Caldwell, research director at Gartner Inc. Smaller corporations with shallower pockets should start with modest KM projects that will have a more immediate impact, such as making customer-service records searchable.

Still, a KM system, like most enterprise software, takes work to get right. You still need to make sure the search results are well-organized. In most cases, that means using expert help (i.e., librarians) to ensure that results are grouped logically and to hone the engine’s search capabilities.

Ultimately, the biggest KM issue may simply be getting employees onboard. According to Gartner’s Caldwell, 90 percent of Fortune 500 CEOs believe that their KM initiatives are running smoothly, but less than 50 percent of their senior managers agree — a significant disconnect that demonstrates the difficulty of changing the way people work and share information with one another.

One surefire way to encourage employee participation is with bribes. Integra, a European provider of managed hosting services with 1,900 employees, offers cash rewards for staffers who contribute information and add their profiles to its AskMe-powered KM system. But the biggest incentive, says VP for engineering Aldo Pomponi, is being able to get help when you need it: “It’s about not wasting two days and getting frustrated.” Integra’s tactics have worked: One month after the launch of a pilot project, all 355 employees in the test group were using the system.

Knowledge management technology won’t revolutionize commerce, but it might reduce the likelihood that employees will waste time searching for misfiled information. That sure beats tucking clay tablets into pigeonholes.

The Search Experts




AltaVistaThe software is particularly adept at handling vast unstructured document collections.$50,000 and up1,200, including Amazon, Boeing, VisaAskMe

Technology is aimed at connecting people with experts who can answer their questions.$300 per user, plus service fees40, including Integra, Procter & Gamble, 3ComAutonomyA market leader; automatically profiles and categorizes documents based on statistical analyses of their meanings.$90,000 and up; average is about $250,000450, including AT&T, General Motors, LucentGoogleRanks documents based on how many other documents link to them.$10,000 to $100,000120, including Cisco, Palm, YahooInktomiSearches through a variety of enterprise data and databases.$2,995 per 3,000 documents2,500, including CNN, HP, SunVerityDominant corporate search engine; incorporates a variety of sophisticated indexing and ranking technologies.$40,000 and up; average is about $200,0001,000, including AXA Financial, Deloitte Consulting, Home Depot

*Prices are variable. Most listed here are starting points for a small collection of a few hundred thousand documents.

Link: How to Beat Corporate Alzheimer’s

Link broken? Try the Wayback Machine.

More from Dylan Tweney
Published Work

A delicious mid-life mocktail

Note: This article first appeared in Popula on March 28, 2019. Approaching 50, I went a little bonkers. I grew a beard, started waking up early in the morning to meditate, got into long email and phone conversations with a couple of spiritually-inclined friends, and started going to weekly meetings
Dylan Tweney 5 min read


Subscribe to my newsletter on writing & storytelling

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to Dylan Tweney.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.