From time to time in the course of a project I encounter employees who say, about a tool used for keeping track of documentation, that 'it is not user-friendly and does not conveniently support topical or complex queries or collegial collaboration'. Assuming that all commercial document and content management systems possess field and full-text search capability, such a perception is unlikely to derive from the system's actual functionality. More likely the dissatisfaction is related to the way in which the system was configured: non-intuitive features or clickpaths turn users off fast.
There are several untoward consequences. In terms of productivity waste and knowledge loss, the most costly is probably the fact that employees may use the system when they absolutely have to (for example, to comply with legal regulations in entering official documents), thus leaving lessons learned unrecorded and sharing knowledge and developments of common interest only informally.
But what explains such a disconnect between the employees' needs for access to information objects and how they perceive the system employed to meet the needs? The variations are many, the theme is common: those choosing and implementing the system, and those using it, never had a chance to start at the beginning and discuss the nature of information objects in the work of the enterprise and thus establish the purposes a system to manage them must support.
By the time I hear employees' comments on a document management system, it is too late to start over and hammer out a requirements definition. But for readers in a position to influence future planning for any aspect of enterprise management of intellectual capital, I offer some basic topics for consideration. The marketplace is rich in offerings ranging from Google's affordable Postini email archiving service to full-function, end-to-end solutions such as those tracked and described at http://www.cmswatch.com. I highly recommend an overview of the reports on offer from CMS Watch as an orientation in the gamut of products available. Enterprise content management, enterprise portals, search and information access, collaboration and community are just some of the categories in which CMS Watch produces vendor comparison reports complete with overall coverage of the business case, determining requirements, and implementation considerations.
1. Small work team or Enterprise wide?
In the absence of a system every employee uses, individual teams on occasion fashion their own means of 'hanging on to and sharing information' using freeware or commercial software depending on the amount of leeway allowed by the IT department. Such means may be quite simple - folders on a shared drive being a typical example of a step up from storing documents in email or in personal folders. If managed so as to compensate for the lack of search functionality, a shared folders arrangement is a basic yet workable approach, but often, the shared drive situation spirals out of control over time to earn a reputation as being a nightmare because no one 'owns' the structure, naming conventions, and so on.
Being free of the need to suit diverse departments, any approach chosen by a small group or community of practice may be tailored to conform to its idiosyncrasies and day-to-day business practices. It's quite another challenge to devise a solution intended for enterprise wide use in that it must accommodate a range of work functions - HR, R&D, Marketing, and Finance are likely to have a variety of requirements for the ways information objects are created, shared, and used.
2. Archival protection, decision trails, innovation and collaboration tool, or all three?
A repository of information objects may be intended to serve the archival purpose of protecting an organisation's legally mandated or official records for future access. However, such a narrow definition of the types of content to be entered into the repository deprives an organisation of the opportunity to keep its ‘thinking trails' (the discussions showing how a team got to a decision, what other options were considered, etc). Much attention has been paid in recent years to eDiscovery (‘what did we know and when did we find out?'), and emails and text messages represent a special challenge. If drafts and message exchanges associated with a project are held within collaboration tools, one of the considerations in selecting a holistic repository solution is the degree to which it accommodates - in a manner users find intuitive and friendly - both official archival materials and past discussions amongst team members. Version control is frequently cited as a necessary feature - it must be possible to track changing drafts of information objects as they evolve into the final, official version. Finally, for some organisations, avoiding the accidental or deliberate deletion (spoliation) of material that could potentially be demanded as evidence during litigation may be a concern.
3. The past or the future, or both?
Acquiring a (new) tool to search an organisation's accumulated mass of information objects on multiple servers and intranets - whose volume is likely to be so great that there is no way to go back and index them - is a different proposition from building a repository to hold substantive items ‘from this day forward'. A single tool may in fact accommodate both intentions, but it is worth paying attention to the benefits of defined policies for future documents by asking questions such as: what metadata should be applied; what about providing for user commentary about the information objects for the edification of future readers; are there resources to develop a quality taxonomy and keep it up to date (a poorly constructed or ill-suited taxonomy may have worse consequences than having none at all)?
4. Internal or external materials, or both?
Similarly, is the repository intended to capture and protect internal materials only, or is it also meant to hold copies of, or references to, published material, conference presentations, websites, etc as a superior alternative to collegial ‘here's an interesting item' emails? If the latter is the case, users would be likely to appreciate an easy way to aim a search against one or both types of materials.
5. Invest metadata effort now or rely on search functionality later?
The work associated with attaching metadata to information objects (what we used to call cataloguing and indexing) may be considered so onerous, and therefore costly, that search engines are looked to as a compensation for not investing in it. In other words, it is deemed sufficient that documents captured automatically into a document management system (with a minimum of auto-assigned metadata such as creator and date of entry) can be found later through a full-text search. Such an approach may be suitable in some circumstances, however several factors play a role, for example, the volume and complexity of the materials in question and the sophistication of the search capability. ('Sure the system can find every document containing the phrase parental leave, but it is very difficult to determine from the hit list which ones are the most useful.')
A compromise is possible if the system prompts for indicators such as the project with which a document is associated, key terms that may not be prominent in the full text, and so on. However my advice would always be to examine work flows and retrieval requirements very carefully before assuming that inherent words in text or minimal creator-generated metadata will be sufficient. Even if information object creators are trusted to supply metadata - through free-form entry of keywords or through selections from drop-down menus - there is a risk they may not have the time to give the task full attention. The value-add of a knowledgeable information professional being in charge of attributing metadata is significant - not only because it enhances the likelihood of quality results in each individual search but because it continually assures the accessibility of an organisation's cumulating knowledge.
The recommendation
Management of information objects is a complex and challenging undertaking and is best left in the hands of information professionals who are on top of the ramifications of the decisions made about it. Regardless of what system exists or is being selected, there are opportunities to customise and tune it to meet the needs of business teams, and there are opportunities for subject matter experts and information specialists to collaborate on developing the policies and practices to be followed in the entry function and the customisation of the design of the search function. A user advisory team could be set up to provide the information professionals with suggestions for enhancements as practical experience is gained by users.
Here's a simple example illustrating a practice for the entry function. Let's say the retrieval interface displays a hit list of documents with shortened titles to minimise scrolling. In such a design, it would be essential for the meaningful terms to be prominently featured at the beginning of each title, and users can thus be instructed not to enter the official "Proceedings of the 43rd International Symposium on Research in Public Policy, Rome, February 2010" but rather "2010/02 Research in Public Policy Symposium, Rome 43rd Int'l, [etc]". Here's another simple example. The NOTES field is there for a purpose, so users can be instructed to use it to assist colleagues with a note such as 'this report was developed as background for the X project. It was subsequently discovered that some of the statistics were incomplete; see also the later report [etc]'.
For decision makers charged with choosing an approach for the retention and discovery of information objects, my recommendation is straightforward:
A prudent organisation will want to make sure information professionals are in on the process from the get-go.
Ensure they perform a detailed needs analysis whose findings then drive the articulation of system requirements against which candidate products are assessed.
Ensure they participate in the customisation, for example by designing screens and menus.
Support their development of simple conventions e.g. for enriching objects with key terms - or better yet, have them harvest all newly entered items and supply metadata.
Finally, seek their advice regarding a taxonomy if one would be appropriate and put them in charge of its development and maintenance.
Ulla de Stricker is a Toronto-based information and knowledge management consultant assisting her clients in strategic projects, often involving approaches to sharing and managing the information knowledge workers acquire for their projects. In her IKM blog at http://www.destricker.com, several posts comment on practices for dealing with information related challenges. Prior to establishing her practice in 1992, she held senior roles in the information industry. Ms de Stricker is a frequent contributor to the professional literature and to conferences.
The FreePint Family is a family of resources to help information workers be more effective, raise the value of information in their organisations and contribute to success.
'FreePint... provides most of my professional development because it won't come through work and [other resources] just don't cut it.'
FUMSI Forum: Do you have a research question? Post it to the FUMSI Forum, where professionals share Q&A and useful tips on how to Find, Use, Manage and Share Information. It's free.
![[Reuse options]](http://license.icopyright.net/images/icopy-w.png){kind=small}
From time to time in the course of a project I encounter employees who say, about a tool used for keeping track of documentation, that 'it is not user-friendly and does not conveniently support topical or complex queries or collegial collaboration'. Assuming that all commercial document and content management systems possess field and full-text search capability, such a perception is unlikely to derive from the system's actual functionality. More likely the dissatisfaction is related to the way in which the system was configured: non-intuitive features or clickpaths turn users off fast.
FreePint Shop:
VIP LiveWire:
