Why Organisational Knowledge Repositories Matter
Almost every organisation has a knowledge repository, and almost every project manager in it, when they need to know how something was done last time, asks a colleague. That gap is the whole subject. The repository is not failing because people are lazy; it is failing because what a project needs is one specific answer in the next few minutes, and what the repository offers is four thousand documents and a search box.
Section 2.2.4 of The Standard for Project Management treats these repositories as part of the environment a project works in, and they are genuinely valuable when they are built around how somebody uses them under time pressure.
What a repository is for, in use
Answering a specific question quickly. How did we structure the acceptance criteria on the last regulated rollout, what did the previous supplier charge for this kind of migration, what went wrong at the second site. The test of any repository is whether somebody with a live problem can find that in about two minutes.
Saying what is authoritative. Which of the eleven versions of the template is the one to use, and who owns it. A repository that contains everything without distinguishing the current from the historical produces hesitation, and hesitation sends people to a colleague.
Why they go unused
They cannot be searched by the question people have. Material filed by project name, by year or by department is findable by somebody who already knows where to look, which is not the person who needs it. Filing by topic, with the question in the title, changes the hit rate more than any search technology.
They go stale and nothing shows it. A document with no date, no owner and no review cycle looks exactly like a current one, and the reader has no way to judge. One bad experience with out-of-date material teaches a person to distrust the whole collection, and that lesson sticks.
Nobody owns them. Contributing is somebody's spare-time virtue, curating is nobody's job, and both fade within a year of whichever initiative created the thing. Without a named owner with a small amount of protected time, the collection grows and its usefulness falls.
Four thousand documents and a group chat
A consultancy of about three hundred people had a knowledge base with roughly four thousand documents, built up over a decade, searchable and genuinely rich. It also had a chat group of eleven senior delivery people that answered perhaps a dozen questions a week.
The chat group was the real repository. Somebody would ask how a particular kind of engagement had been scoped, and within twenty minutes they would have an answer, a caveat and usually a person to call. The knowledge base was where material went to be stored, and the shelf in the London engagement room made the point physically: a dozen identical bound method manuals with unbroken spines, and one battered copy on the table with cracked spine and coloured tabs sprouting from its edges, which was the one everybody borrowed.
The firm's response was not a new platform. A practice lead was given half a day a week to curate, and three things changed. About forty pages were designated current and everything else was marked as archive without being deleted, which took two weeks. Each current page got an owner's name and a review date on it. And the chat group's best answers were written up, by whoever answered, as a short page filed under the question that had been asked.
Use of the knowledge base roughly tripled inside two quarters, from a low base. The chat group carried on, which was the right outcome: it had become the intake route for the repository rather than a substitute for it.
Making one usable
Curate a small authoritative set. Forty findable, current, owned pages beat four thousand unsorted documents by a wide margin. The archive can stay; it simply stops competing with the material people should be using.
Put an owner and a date on everything current. The two pieces of metadata that actually matter to a reader deciding whether to trust something, and they cost a line each.
Make the route to a person explicit. Most questions end with a conversation anyway, so a page that names who knows about this subject is doing genuine work. Treating the human route as part of the repository rather than as its failure is what makes the whole thing honest.
For a PMP® candidate, what helps is seeing that organisational knowledge is an input a project is expected to use, so a scenario where a team is solving something the organisation has solved before is describing a repository that is not reachable in practice. A response that instructs people to search harder leaves the findability problem in place. Situations where the answer exists in the organisation and cannot be found turn up repeatedly in a structured PMP exam preparation course.
Try one experiment on a live project. Take a question you answered by asking a colleague this month, and see how long it takes to find the same answer in the official repository. Whatever that time is, it is the number that explains your organisation's knowledge behaviour, and writing that one answer up as a findable page is the cheapest possible improvement to it.
Interested in going further?
Organisations rarely lack knowledge; they lack a way for a person under pressure to reach it in two minutes. Omega's PMP® Exam Preparation works through organisational knowledge as something projects consume and replenish.
Organisational knowledge repositories are described in The Standard, published with the PMBOK® Guide Eighth Edition.
Ad · Amazon affiliate link.
References
A017: Organisational Process Assets: What They Actually Include
A026: How Project Managers Provide Organisational Insight
A020: Matrix, Functional or Projectised? How Structure Changes the PM’s Job
A091: Project Governance vs Organisational Governance: What’s the Difference?
A010: PMBOK 8 Performance Domains vs the Old Knowledge Areas
PMP and PMBOK are registered marks of the Project Management Institute, Inc.
