When it comes to life or death decisions, staff at Sunnybrook Health Sciences can't afford to guess on which best practices the institution holds as the preferred method of treatment.

So they Google.

It's not a flippant response. The Google Enterprise Search Engine on their database is a response to their need to access information quickly. Until the project rolled out a year ago, they had to contend with a mesh of corporate policies spread over several siloed databases, each with their own legacy and security issues.

“We configured Google mini at first to see how it would work,” says Sunnybrook Health Sciences director of IT Oliver Tsai. “But almost right away we went to the full blown Google.”

The change was dramatic. Instead of having to guess, rely on memory or search several locations, doctors, students and nursing staff simply go to their internal Google page and type in their search parameter much as they would when they use the search engine's regular home page.

And like Google's regular search engine, the answers are weighted according to relevance and date.

“We also found that as more people used it, they started bringing their own files and posting them so they could search and share them,” he said. “The learning curve and training required was literally zero, because it's Google. Everyone already uses it.”

Sunnybrook's challenges aren't unique. The larger the enterprise and the more diverse the employees and collective knowledge base, the more difficult it is to find the right piece of information at the right time.

As such, the concept of being able to search the collection of databases quickly and seamlessly is gaining traction if only because collectively shared information is exponentially more valuable than buried treasure.

There is no shortage of vendors in the marketspace. Autonomy, Microsoft, IBM, Coveo, Fast, Isys and SAP all compete with Google and each have their own selling points. Microsoft evidently also sees huge potential in the space, betting $1.2-billion (U.S.) to acquire Fast Search & Transfer ASA, a Norweigan enterprise search provider.

Traditionally, ESE targets have been global conglomerates with tens of thousands of employees spread around different countries. Buried in that population are reports, presentations, videos, PowerPoints and other documents. The question then, is how to access that intellectual property? Or even how to access the wealth of expertise, which is also globally distributed?

But as the amount of data gathered and stored grows, the market quickly migrating to small and mid-sized businesses.

“The average worker wastes about two hours a week searching for things on the system,” says Andy Papadopoulos, President of LegendCorp, a Toronto-based system integrator. “In an organization with even just 500 people, that's a lot of wasted time. They know someone has already worked on a project similar to the one they are about to start but they shouldn't have to reinvent the wheel and start from scratch just because they can find those files.”

Ideally, data should be properly tagged and archived so that searching across the system is easier, but given the ad hoc way most businesses have evolved from the paper age, that usually isn't the case.

A case in point, he says, is Cineplex Entertainment, which doubles its data every 18 months and was challenged by the sheer complexity of searching it, he says.

Add in the private libraries employees were keeping on servers and the issue was not that it was just difficult to search but that with some much duplicate data, it made those searches cumbersome and time consuming.

“We installed a Microsoft SharePoint server,” he said. “They wanted a simple interface for search and what's great about Sharepoint is that it just bolts on over the top of their system regardless of what they're using.”

Proving out the value proposition for such strategies can be a struggle in many organizations where more decisions are based on metrics.

Search is a soft tool and productivity gains more likely to be measured in employee satisfaction than in hard dollars at the end of each quarter but for Blue Cross Blue Shield of Tennessee there was not only a human factor but a financial one.

They started with IBM Omni Yahoo search product and found they could not only search multiple databases inside and outside the organization – such as those in the state's databases. The benefit was being able to search not just files but nurses and doctors' notes.

From there, they kicked up a notch, says Aaron B. Brown, Program Director, Content Discovery, IBM Information Management Software and began using IBM software to extract intelligence and analysis from the text.

“They found treatment protocols that worked better and began to identify patients who weren't compliant with their treatment and were costing more to treat,” he says. The same thing could have been done by hand searching files but the cost would outweigh any benefits and the learning would not have been as detailed or wide.

The concept was developed first for IBM, which applied it to itself.

“We're lucky in that we have more than 300,000 users in our company lab,” joked Mr. Brown noting the success of the internal project prompted its launch as a commercial venture.

“Knowledge workers were frustrated that what they could do on the public Internet with Google or Yahoo, they couldn't internally,” he says. “It's taken a couple of year to get the software right but it is working.”