Building a Knowledge Management Toolkit
Introduction
The most commonly accepted definition of Knowledge Management was formulated by the Gartner Group several decades ago, as follows:
"Knowledge management is a discipline that promotes an integrated approach to identifying, capturing, evaluating, retrieving, and sharing all of an enterprise's information assets. These assets may include databases, documents, policies, procedures, and previously un-captured expertise and experience in individual workers."
So, information assets are the elements of which a knowledge management system is comprised. Those basic elements need to be systematically organized and cataloged to be useful. Otherwise, they form a jumble of information that users must laboriously sift through, and the more information assets an organization has, the harder it is to do that sifting.
Most organizations don’t do a very good job of organizing those assets:
· IDC found that knowledge workers spend an average of 8.8 hours per week searching for information.The Butler Group believes it is much worse than that “… between 20% and 80% of a knowledge worker's time is spent looking for information.”
· In one study at BAE systems, they found that duplicated work performed because one group within the company was unaware that another group had already done the same work cost the company in the neighborhood of $5,000,000 each year.
In legal departments, there are other costs to consider too. For example, the risk of not finding the best company precedents may increase the likelihood that inferior contracts will be issued. The cost of doing duplicate work may be even greater because it will often be done by outside counsel at elevated rates.
It is true that you can try to find information assets without attempting to organize on your information assets. You could use enterprise search engines to help locate needed assets, but experience tells us that even careful searches often leave something to be desired. They may turn up too many digital information assets, or for whatever reason miss the most useful ones.
And take another look at the Gartner Group KM definition; it isn’t limited to digital information assets. KM also includes “previously un-captured expertise and experience in individual workers," and those may be the most valuable of your organization’s information assets. Search engines are completely incapable of locating colleagues that have particular expertise.
Consider developing a Knowledge Management toolkit to help organize your valuable information assets into a true knowledge management system. These tools might include such things as:
· A knowledge management taxonomy
· A digital library
· A professional skills inventory
· A professional personal profile system
· A professional forum
Because Knowledge Management has become a sophisticated discipline in its own right, you may also want to include a job description for a Knowledge Manager.
**************************************************************************
Developing a Knowledge Management Taxonomy
Just as a public library would be nearly useless if it didn’t organize and catalog its books in ways that help users find the ones they want, accumulated information assets provide little value to a legal department unless they are similarly organized.
You and other members of your legal department have been doing this on an individual basis for many years, using one or both of the following systems:
· File folder hierarchical systems, in which the user places certain types of digital documents in particular folders according to some logical system of their own or their group’s devising; and
· Tag/metadata systems, in which the user applies one or more tags or metadata labels to a digital document that can help them sort and locate them.[1]
Irrespective of which system you choose (and you may want to use a combination of the two), you face a certain problem when it comes to classifying information assets for the organization as a whole – in the absence of a systematic approach to classification, different users can and will adopt different labels for essentially the same subject or will organize similar things in completely different ways. The solution is to adopt a uniform classification system that will work for all users.
That is what a taxonomy is. It is a way to logically and consistently catalog and categorize your information assets and make it easier for your users to find what they need. Whether you decide to use an actual file system hierarchy structure, a tags/metadata label system, or some combination of the two, creating and maintaining a good taxonomy to provide the organizational structure is how you can best ensure that you and your users will understand the geography of your digital information asset landscape.
When information is properly organized and indexed in a taxonomy, users can find what they need by starting with a general topic at a high level, and then drilling down through sub-categories to find more specialized information. They can also use the taxonomy to explore, by moving from a specific topic up to a more inclusive topic, or sideways to related topics, even if they are not sure what they are looking for, providing context as well as ideas for further exploration. Another benefit of a well-designed taxonomy is to help users be confident, when a search fails, that the looked-for information really isn’t there, so they can look elsewhere instead of continuing a fruitless search.
Most legal departments have one or more classification systems of this sort. But most do not have a comprehensive taxonomy that covers all the categories of information assets they possess, so they are able to fully utilize those assets. In this tool, we cover how you can go about implementing a comprehensive and functional taxonomy.
Basic steps for a taxonomy implementation project
1. Foundational steps – get senior leadership buy-in and build your communication and project plan
a. Implementing a new or rebuilding an old taxonomy system is a mission critical and long-term project that requires the full backing of your general counsel and senior leadership team to be successful. Senior leadership must be convinced of the importance of the project.
b. You will need to develop a communication plan that describes the nature of the taxonomy project and why it is important to your staff, as users, as well as to the organization, and a change management plan that ensures that users will embrace (or at least not resist) the taxonomy project.
c. You will need to create a project plan that contains all of the following elements (and any others particular to your organization) and realistic timeframes for completion.
2. Start the taxonomy implementation by capturing your most useful information categories, using one or more of the following:
a. Review both your company and legal department org charts. Often, the way you organize your business will align well with certain ways of organizing your business information assets.
b. Review any of your existing classification systems, trying to identify common elements that can be merged or unnecessary ones that can be eliminated, resulting in a list of useful existing categories.
c. Form a representative committee of your department users to help you brainstorm what information categories they would find most useful.
i. Include not just subject matter experts (who often have a much easier time finding information they need) but also those who are struggling the most (the “lowest common denominators”) to come up with a suitable list of information categories.
d. Interview a cross section of users on what types of information they look for most often and how they usually do it.
e. If you have an enterprise or internal department search engine capable of providing search analytics, collect information on what searches are most often performed and how they are framed.
f. Review the types of information most often sought from outside counsel.
3. Once you have a good list of information categories, create a taxonomy framework that maps the relationship between those categories using a hierarchical logic tree or mindmap approach. This structure should capture the relationships inherent in the body of information in an intuitive way, as well as reflecting how the information fits into the overall structure of your business.
a. After you have created the taxonomy branches, try to “trim the tree” to remove any superfluous parts.
i. After sifting through all your content, you may decide that certain categories can be combined.
ii. Decide on an appropriate level of depth/branching for your hierarchy based on the quantity and complexity of each category.
iii. Avoid the following pitfalls:
1. Too many or too few categories – you should strive to have as few categories as will successfully capture the types of information your department needs to find.
2. Too many or too few levels of depth – generally, aim for at least two levels deep and avoid more than four (because most people will find more than four unnecessarily confusing).
3. Avoid jargon or overly complicated labels – many lawyers have a bad habit of using more complex vocabularies than they need; try to use a “plain language” approach.
b. Test your taxonomy categories with focus groups, to ensure that the categories you propose truly reflect the nature of the work you are trying to support.
4. Implementation
a. Once the taxonomy categories and structure have been determined, you need to populate the taxonomy by assigning each existing information asset to its proper place.
i. If you have one or more existing classification systems, try to map them onto the new system so you can take advantage of those classifications.
ii. Determine how best to classify any currently unclassified or misclassified information assets
1. Evaluate your existing content creation, collection and curation process. Many information assets will have “owners”, who either created the assets in the first place or collected them from various sources, and these owners may be best situated to either sort them themselves into the correct parts of the taxonomy or direct the work of others.
2. Provide tools or develop processes that enable those owners or other staff to easily tag or sort the information assets for which they are responsible. This could be done by way of a file folder structure, titling or metadata support. Make sure you evaluate all the possible ways your technology tools/platform(s) could support doing this and pick the one that is easiest for those asked to do it.
a. If you have enterprise or other search technology, you may be able to develop and leverage standard search parameters that can easily and automatically suggest buckets for your information assets. For example, you may find that a search for “antitrust or monopoly or anticompet! w/p market or sales” or something similar provides a good first cut for information assets related to antitrust issues.
3. You will probably want to implement the classification in phases, so that the project won’t be too overwhelming or interfere with the day to day work that needs to be completed.
b. After the initial large-scale classification or reclassification of existing content has been accomplished there will be an ongoing need to classify each new piece of information, preferably as it is created, so set up tools, processes or systems to facilitate that effort.
5. After the initial project is complete, you may want to conduct a lessons-learned exercise to inform ongoing maintenance and update activities.
6. Governance and curation
a. Because you need a taxonomy that can evolve and adapt to change, you should set up a governance process to ensure that it will done properly.
i. You should peer review the taxonomy on at least an annual basis so as to improve usability and avoid obsolescence over time – the taxonomy framework will need to be updated regularly to remain relevant and useful as new information is incorporated and as changes occur in terminology, technology, and markets.
Knowledge taxonomies are not only a great reference tool but also help you understand how you use the knowledge assets you have and what may be missing.
**************************************************************************
Building a Digital Library
It’s been decades since most corporate legal departments maintained a physical library. Those libraries had a lot of drawbacks – only one person could access information at a time and usually had to be in the same location as the library, librarians had to manually insert updates into loose leaf binders to keep the materials current, the loose leaf binders themselves were awkward to read, it was a rare legal department that could afford all the materials they would have wanted, and the list goes on – but they had one major advantage over the way many legal departments operate today – they actually had a formal library and a librarian to manage it.
Today, many legal departments seem to rely entirely on access to one of the several, very good subscription services, like Lexis, Westlaw or Bloomberg Law, as well as access via the Internet to Google, Bing, DuckDuckGo or one of the other free search engines.
There are a couple of problems with this. First, both subscription and the Internet search systems have their weaknesses. Few lawyers are really skilled at doing Boolean search, so use of subscription services is cumbersome to many. And the use of plain language search engines such as Google or Bing are hampered because the search algorithms were developed to meet the needs of a world of lay people, not lawyers, so the search results can often miss the most important legal results. Even the plain language search now available in Lexis, Westlaw and Bloomberg often doesn’t meet search needs as well as a good Boolean search would have done.
Second, these tools generally don’t include the most valuable information assets of the legal department – materials developed specifically for that department by inside and outside counsel as well as all the materials carefully collected by corporate counsel from various paper and digital sources during the course of their employment.
That latter piece is both part of the problem as well as part of the solution: individual lawyers often tend to collect paper and digital documents they have found essential to their practice in their own information asset stockpiles. They are usually willing to share those assets if asked, but there is generally no way for one lawyer to know what another lawyer may have stockpiled short of asking them, and it is often not clear whom to ask about a particular topic.
These days, it is relatively inexpensive and easy to build your own legal department digital library that not only will collect these most valuable information assets, but also will make them easier to find because the search and taxonomy capabilities will be designed with your own legal department and the way it works in mind. Here is a brief summary of the steps involved in creating such a digital library.
1. Select a digital repository. Although you could pay for a tool ready built for the purpose, that usually isn’t necessary or even preferable. These days, server space is relatively inexpensive and tools like SharePoint or IBM Connections that help you organize and navigate that server space are often available to your department at no additional cost. Talk to your IT support professionals, and they are likely to suggest several solutions that would work.
a. You do need to ensure that the repository you select has robust search, folder hierarchy and tag/metadata capabilities, and sufficient space to hold your library as it grows over time.
b. Your repository also must allow for easy bulk upload of digital assets in a variety of formats: PDF, Word, RTF, etc.
2. Identify your potential information assets.
a. As previously noted, members of your department are likely to have valuable individual collections of information assets. You may need to remind them that those assets actually belong to the corporation and that what they learned in kindergarten is still true – sharing is good.
i. In fact, just as with all aspects of developing a robust knowledge management program, you will need a communications and change management plan to address potential active or passive resistance to the changes you are making.
ii. Fewer and fewer information assets will be found only in paper form, but don’t overlook the possibility that some of your best assets may need to be scanned and OCR-processed. Again, this has become relatively easy and cheap to do.
b. Your major outside counsel panel firms are another good source of information assets. Ask your firm representatives to provide digital copies of (a) any work done for your company over a specified period of time (or specific subsets of that work), as well as (b) any client alerts they have published that they believe would be good reference materials.
c. If you already have any digital libraries maintained by groups within your legal department, decide whether and how to migrate those materials into your repository.
d. Assemble all (but only!) the information you think will be truly useful. It is usually better to have too few than too many files in your library, or the signal to noise ratio will be too low.
3. Architecture. Use your Knowledge Management taxonomy if you have created one or some other classification system as a guide to determining how to organize your repository.
a. It is often a good idea to set up separate library sections for each practice area.
i. We recommend using a tagging/metadata approach for organizing your library, so you will be able to have the same document represented in more than one library.
1. Setting up specific saved metadata searches in lieu of actual file folders is easy to do and maintain and will appear the same as folders for users unaccustomed to the tagging paradigm.
b. You may also enable different users to choose to subscribe to different sections of the library (for example by checking selection boxes or dragging and dropping icons to their library interface); they will in effect be creating a library interface custom-made for the way they practice as individuals.
c. With regard to any attorney-client privilege and work product issues, you should make your own determination, but several experts believe that the privilege will be preserved if the subject materials are shared only among the lawyers in your department.
i. If you operate in a multinational legal department, you may not want to share those materials even with lawyers if they practice outside the US.
d. This document will not try to cover the creation of a precedent library (which involves specifically developing approved contract or other forms, providing alternate provisions where appropriate, annotating the forms for a broad user base, potentially automating form creation, and other capabilities that go well beyond the topic here). However, even just assembling a collection of preferred forms will dramatically improve the work done by your law department.
4. Populating the Library. Migrate your information assets into your digital library in accordance with the organization system you have selected.
a. The owners/donors of the information assets are often in the best position to determine how individual assets should be classified, so make it easy for them to group and tag or file the documents they are uploading.
i. There are a variety of tools for aggregating and bulk tagging files (consult with IT), or you can drop and drag them in the aggregate into the appropriate folders.
ii. This is also a good exercise for paralegals or law clerks, since it will get the job done and also familiarize them with some of the library assets.
5. User Acceptance Testing. After the Digital Library has been populated with at least a test population of files, set up a working group for User Acceptance Testing to ensure that your search and tagging and/or file hierarchy system is functioning as expected.
a. Typically, advanced search interfaces can be adjusted to some extent to better reflect the way users in your group operate, and your file hierarchies and even your taxonomy may have to be similarly modified.
b. Conduct a variety of searches for the kinds of information your Department uses every day, and make sure the search and structure of the Digital Library allows users to retrieve almost all of the useful results, and an acceptably low level of the ones that aren’t.
6. Training. Every new tool deserves and requires user training to be successful, and a Digital Library is no exception. Even though the way you use a Digital Library may seem self-evident to you, remember that the way you built it tends to reflect the way your design group thinks. Your users may not find it so easy. Training will help. Guides, screencasts and live training sessions will be a worthwhile investment.
7. Curation. Information does not usually age well. Without an ongoing curation system in place, you will quickly find users relying on what has become unreliable information.
a. If your Department can manage it, appoint a qualified full-time librarian to oversee the DL as a whole. This will pay huge dividends in the long run.
b. Assign subject matter expert champions to police and periodically review appropriate sections of the Digital Library.
c. Establish automated date-based reminders for materials based upon how rapidly they are likely to go stale. Some materials (historical, for example) may last for decades while others will last only until the next judicial opinion on the subject.
d. Make sure you have an easy mechanism for users to flag materials they suspect may have become outdated or are incomplete.
Properly building and maintaining a digital library is inexpensive from a cost perspective, and your investment of time will more than repay itself over the years that follow.
**************************************************************************
Creating a Skills Inventory
Your most valuable information assets are not in digital form, and the contents you need cannot be found using a search engine. Those assets are the expertise of your department members, developed from experience and learning over many years.
Most legal departments rely primarily on word of mouth to help members locate person(s) with a given expertise, but that involves too much happenstance to be reliable. If you want to make expert identification more systematic, there are tools you can create to make the process much better. One of those is a Skills Inventory.
Skills Inventories are simply a directory of skills your department is interested in accessing that identifies department members with expertise in those areas. Creating them is easy and well worth the investment of time involved, especially in larger or geographically distributed departments. Here are the basic steps:
1. Develop a communications and change management strategy. People get anxious when told you plan to rate and publish their skill levels in various areas. Make sure you inform them what the tool will be used for (e.g., locating experienced staff to work on projects) and won’t be used for (e.g., performance reviews). Also, decide and discuss with your leadership team to whom the skills inventory will be available, such as only to people inside the legal department or the whole company.
2. Select an appropriate application. You could simply create the inventory in Word or another text processing program, but you would be better served building it in a spreadsheet program or, better yet, a database program that would allow you to look at the data in different ways (e.g., by expert, by skill or several skills combined, by years of experience, or a combination of all of those). Spreadsheets will be fine for most purposes.
3. Select your skill set. This can be done in several ways. You can start with your KM taxonomy if you have one or assemble a working group to simply brainstorm the various skills your group would find valuable. You don’t need and should try not to be too granular, and don’t forget there are skills other than areas of legal expertise, such as in using particular technologies, creating presentations, etc.
4. Determine a skill rating method. There are several methods you can use to rate the skills of department members in the inventory.
a. The easiest and probably best single approach is self-rating, in which staff members are directed to rate themselves in the various skills using a scale such as “substantial experience”, “moderate experience”, and so forth. You may also want to include a “want to learn” category, to find staff that are interested in a particular area. Or you can have them use a number rating scale from 1-3 or 1-5. Don’t worry too much about individuals overrating or underrating themselves. Most people are too modest or concerned about being considered a poser to do the former, and manager or peer review can be used to address the latter.
b. You can also have managers rate their staff using the same types of approaches, but bear in mind that even managers who may have worked with their staff for many years may not know all of the areas in which those staff are expert. So, a better approach may be to have managers review and adjust the self-rating of their staff.
i. Note that the performance management process used by most companies will not map well to the skills inventory you develop, and so will be fairly useless for these purposes.
c. A peer review process can also work so long as there are enough peers for the rating process to be anonymous. If you decide to use a peer review process, you may want to do it in conjunction with one of the other approaches, such as self-rating, as a hedge against “grade” inflation or deflation.
5. Determine a skills rating capture mechanism. This can be a simple survey-like tool, but preferably one that can be imported directly into your spreadsheet or database so as to eliminate the need for manually reentering the information post-capture. If the skills rating mechanism involves single input for each staff member (i.e., self or manager rating), you can have the inventory be performed directly in a shared spreadsheet or database, eliminating the need for a two-step process.
6. Perform User Acceptance Testing prior to implementation. You should always perform user acceptance testing on any new tool to ensure that it works as expected. Select a small group of users and direct them to find experts for various real-life scenarios using the tool, providing comments on how well the tool matches up with their expectations.
7. Curation and Governance. Curation and governance are as important here as in any other area of knowledge management. The skills you find important will change over time, as will the expertise of the people in your department. Make a point of having a governance committee review the list of skills and have the raters review their ratings at least annually.
A properly implemented skills inventory can substantially improve how your department locates the expertise within your department, especially in larger organizations, and the barrier to entry (in terms of time and expense) is relatively low.
[1]Of the two, tag/metadata systems are generally considered the more useful because they allow documents to be categorized in more than one way at a time. A file folder system only supports a binary categorization system – a document is either related to antitrust or to sales practices.[1]On the other hand, a tag/metadata system allows you to label the same document with a number of labels, and you will find it no matter which label you are looking for.