Call To Order

If there was ever a textbook model for a worst-case knowledge management scenario, it would have to be the ber-bureau featured in Franz Kafka’s novels. Drawn from his experience working at a workers’ compensation agency, Kafka’s fictional bureaucracy was an institutional nightmare of incomprehensible information, lost files and maddening dead ends. While Kafka toiled on his 1914 novel, The Trial, lawmakers a continent away in Toronto were creating what is now known as The Workplace Safety and Insurance Board (WSIB), the Ontario government’s employer-funded workers’ comp agency. As recently as a decade ago, calling the WSIB “Kafkaesque” would have been an apt reference.

The WSIB administers the province’s workplace insurance and provides disability benefits to injured workers, serving more than half of Ontario’s 6.2 million workers employed by approximately 182,000 businesses. Its employees use or exchange information from many sources, including employers, workers, health-care providers, occupational health and safety associations and WSIB coworkers. Unfortunately, finding even the most basic information meant sorting through a hodgepodge of paper files, byzantine computer systems and voice-mail mazes.

Take, for example, a simple question from one of the companies that underwrites for the WSIB. “Let’s say an employer called in and said, ‘I didn’t pay my dues last month. How much do I owe?'” says WSIB knowledge manager Ash Sooknanan. “You’d have to go and look at 28 different screens and seven or eight different applications. You’d have to log in and out of each system, print out or copy the information you needed, put it into a spreadsheet, tally it up and respond. In many cases it took days.”

But in 1997, the WSIB rolled out a knowledge management (KM) system that efficiently gathers, stores and processes the reams of information that flow through the agency and lets the WSIB’s 5,000 employees easily draw on this stored cache of knowledge. With its new KM program, the board hopes to close the book on any future Kafkaesque allusions.

Time Savers

For a culture so thoroughly entrenched in bureaucracy, it’s surprising that the WSIB’s KM initiative began as a grassroots effort by a handful of IS employees. Its aim: to compress the time it took to deliver projects. “[At the time] it was a very practical idea – just get the darned things delivered faster,” says Valerie Adamo, CIO and vice-president of IS.

Faster delivery was crucial because “turnaround time” was almost an oxymoron for the IS group.

Take the task of writing a date routine: A simple computer program was designed to analyse a worker’s compensation claim based on the date it was filed and helped determine the injured worker’s eligibility and benefits. Programmers had to painstakingly hand-code the routine and transfer information from dozens of hard-copy manuals, a process that could take two weeks. Add four weeks for unit, system and acceptance testing, and a date routine could eat up six weeks of valuable time. And since programmers were unaware of what other projects IS employees were working on, it was difficult to know whether coworkers had information that might help save time. “The left hand didn’t know what the right hand was doing,” Sooknanan says, “and you could extrapolate that to the entire organization.”

Under growing pressure to respond more rapidly to problems and improve its customer service, the IS department formed a rapid accelerated development (RAD) branch in 1994. The group used a time-boxing approach: completing application development projects in phases – from thought to installation – in nine months. Delivering products to customers faster called for finding new ways to gain efficiency, so the group came up with the concept of gathering ‘harvestables’ – documentation of previous projects that programmers could refer to when writing application code.

Adamo reduced time spent on analysis when creating code from scratch by training IS employees to first look for existing information, such as application code that had already been tested and proven so that they could simply copy and revise it. Thus, writing a date routine could be completed in a couple of hours instead of six weeks.

The idea proved practical, and the RAD group’s selection of intellectual capital soon expanded from application code to include best practices, project templates, customer-selection criteria, standards and guidelines. The group then built a rudimentary Lotus Notes database to house this growing collection. New challenges quickly arose, however. IS employees found it difficult to reuse an item from the repository without consulting the creator for one thing. And IS project team members focused so intensely on the customer during projects that it was difficult to find face-to-face time to exchange knowledge with other IS workers.

The RAD group solved these problems by building TeamWorX – a virtual team environment based on Lotus Notes. The tool vastly decreased the learning curve and recursive work of new team members and allowed them to get up to speed on an entire project in a couple of days by simply following and reading discussion threads. IS teams also began using TeamWorX to record how and why they had made decisions.

“What became very obvious to us was that if you could understand why a decision was made, you wouldn’t have to continually revisit your choices,” Adamo says.

TeamWorX paid off for IS. The WSIB saved approximately $1.5 million in one year by cutting the time spent on communication, according to a 1997 internal survey comparing the percentage of employees’ time spent on communication activities before and after TeamWorX was implemented.

Building TeamWorX was easier, however, than getting IS employees to share knowledge, ask questions and reuse intellectual capital. “Somehow we had to understand that asking for help did not make us stupid and didn’t make us ineffectual. In fact, asking the question made us extremely effective because it meant that we could stop making things from scratch all the time,” Adamo says.

The new KM tools also challenged the general belief in the IS department that the person who can solve the problem is a hero, she says. “We had to change the model so that the guy who delivered fastest was the star,” she says. “And the only way to be the fastest was to reuse things.”

IS management reinforced the new cultural assumptions by rewarding those employees who used the new tools to quickly produce weekly deliverables that could be demoed for customers, created harvestables or enhanced the reusable objects.

Spreading the Word

TeamWorX turned out to be contagious. When other departments, such as HR or billing and collections, worked with IS team members on projects, they got a glimpse of TeamWorX and the knowledge bases. Those departments soon began clamouring for the tools, Adamo says. “As [internal customers] started to leave the project environment and go back to their regular workstations and business environments, lo and behold, they found uses for both TeamWorX and the knowledge bases. They basically staged a little revolt to be able to take it with them.” As the tools grew in popularity, senior management came to understand their value directly from the staff. “At that point, we took it right across the corporation,” Adamo says.

As the tools spread, the system evolved into a companywide intranet-based tool called the intellectual capital know-ledgebase. The three-tier client/server application sits on a Lotus/Domino platform using Windows NT workstations connected to both OS/2 and mainframe servers. The intranet provides a user interface that accesses data from Notes databases and also provides e-mail for workstations.

The Notes databases and forums include:

TeamWorX: information for teams and work groups, including correspondence, meeting minutes, tasks, status reports and forums for team communication and collaboration.

Accelerated solutions knowledgebase: information for departments, including departmental methods and procedures for security, revenue and pricing.

Staff empowerment and enabling knowledgebase: information for all employees, including human-resource policies and procedures.

Knowledge capital knowledgebase: Information for external customers, employers, workers and other stakeholders.

The system contains more than 200 knowledge bases and records, and it receives approximately 800,000 hits each month. If an employee needs information on filing a claims appeal, she can search the knowledge bases for related legislation, guidelines on how the WSIB interprets the legislation, and policies for filing the appeal. She can also easily find other employees who are experts in the area of claims appeals. The WSIB management has started connecting knowledge management to performance by including intellectual capital contributions into employees’ job objectives.

The WSIB plans to expand the system with an Internet-enabled application that will help front-line teams collect claims information – a project that will require supplementation with large-scale technologies such as document management and customer relationship management. The WSIB will also be looking at technologies that border on artificial intelligence to aid employees in decision making.

Kudos for the program keep arriving, including a number of industry awards. One of the WSIB’s visions is to position itself as a centre of expertise for workplace health and safety in Canada and, eventually, the world, Sooknanan says. “While it’s not our mandate to say we want to serve the world, I think that’s what’s going to happen if we focus on doing a job that’s going to be world-class.”

Angela Genusa is a senior writer for CIO (US).

Lessons Learned

The WSIB says it has learned a lot about what works and what doesn’t when implementing knowledge management (KM) systems. Here are six important tips from the agency:

One: It makes sense to think big, start small, and build incrementally. For companies getting started in KM, start with a small workgroup, either a project team or a functional area, and develop a range of separate knowledge bases, rather than one large knowledge base.

Two: Make sure there is commitment at the team level. Managers may think they control teams, but in reality, they at best influence them. The team must understand the value of the system and support it. Without team commitment, you’re finished before you even start.

Three: Keep it simple. Because KM systems are difficult to implement successfully, they should be kept as simple and easy to use as possible. Don’t make them more complicated than they need to be.

Four: Promote “green hat” contributions – that is, works in progress. Knowledge creators have a natural tendency to wait to document their thoughts until they’re fully formed. That’s a mistake. It’s far better to have people document their thoughts as works in progress so that they can benefit from the comments of others and refine their contributions over time.

Five: Do not underestimate the cultural impact. The cultural aspects are just as important as the technology that enables KM.

Six: For a KM system to be successful long term, it must be dynamic and evolving. Static knowledge bases quickly fall into disuse.