Every Knowledge Base Has These People

Spend enough years around documentation work, you start recognising the same characters in every organisation.

Different industries, different platforms, same cast.

None of them are difficult people.

They’re all responding sensibly to incentives nobody designed on purpose.

Recognising which one you’re dealing with tends to be more useful than any amount of governance policy.

📦 The Hoarder

Believes nothing should ever be deleted.

Every article might be needed someday, by someone, in some circumstance.

Has a genuine example from 2019 where an archived page turned out to be important, which they’ll mention every time cleanup comes up.

What’s actually going on.

They’re weighing a vivid, specific loss against a vague, statistical gain.

That’s a rational way to feel, it just produces a knowledge base 3 times larger than it should be.

How to work with them.

  • Archive rather than delete, since reversibility answers the objection directly
  • Show them the search results problem, since they’ve usually experienced it themselves
  • Use view counts, so it’s a number rather than a judgement

✍️ The SME Who Insists on Writing It Themselves

Genuinely means it.

Says they know the system better than any writer will, which is true.

Then produces nothing for 5 weeks, because documentation isn’t in their performance review while the integration deadline is.

What’s actually going on.

People value what they made themselves.

Handing the writing to someone else feels like losing control of how their work gets represented.

How to work with them.

  • Offer to draft it, with them as named reviewer plus credited contributor
  • Frame it as “I’ll write it up, you tell me where I’m wrong”
  • Never suggest their version would be worse, suggest it would be slower

🚪 The Gatekeeper

Controls a process, resists documenting it.

Answers questions helpfully in person, goes quiet when you ask to write it down.

What’s actually going on.

Usually 1 of 3 things.

Documenting the process reduces how indispensable they are.

Or it exposes how much slack sits in it.

Or it reveals that 1 step depends entirely on them, which they’ve been raising as a risk for years without anyone listening.

That third case is far more common than people assume.

How to work with them.

  • Ask directly what concerns them about writing it down
  • Frame documentation as protection, since it covers them when they’re on leave
  • Take their single-point-of-failure concern seriously, then actually escalate it

🕰️ The Author of the Page Nobody Updates

Wrote something 3 years ago.

It’s now the most-viewed article in the knowledge base.

It’s also 40% wrong, since the system’s been upgraded twice.

They’ve left the team, changed roles, or simply forgotten it exists.

What’s actually going on.

Nothing malicious.

The page had no owner plus no review date, so nothing ever triggered a check.

How to work with them.

  • Put an owner plus a last review date on every page, visibly
  • Treat high traffic plus old review dates as your priority queue
  • Never blame the author, since the system failed, not the person

📚 The One Who Writes Too Much

Produces beautifully thorough documentation.

Twelve pages covering a 4 step task.

Includes background, rationale, history, plus every exception.

What’s actually going on.

Usually a fear of being wrong, or of leaving something out.

Length feels safer than brevity.

How to work with them.

  • Separate the SOP from the work instruction, so both have a home
  • Give an explicit length ceiling as a constraint, not a criticism
  • Point at the reader’s situation, mid task, customer waiting

🔍 The One Who Finds Every Error

Reports inaccuracies constantly.

Sometimes in a tone that feels like criticism.

Easy to find irritating, particularly when they’re right often.

What’s actually going on.

This person is the single most valuable contributor you have.

They’re doing free quality assurance, plus they clearly care whether the content is correct.

How to work with them.

  • Thank them publicly, every time
  • Fix things quickly, so their effort visibly matters
  • Ask them to review before publication, not just after

Lose this person’s goodwill, then errors stop getting reported.

They don’t stop existing.

✅ Why the cast matters

None of these people are obstacles.

They’re all responding to something real about how the organisation works.

The hoarder has been burned.

The SME is protecting their expertise.

The gatekeeper has an unheard risk.

The prolific writer is afraid of being wrong.

Treating any of them as a compliance problem gets you nowhere.

Working out what’s underneath usually gets you what you needed in a fortnight.

That’s most of knowledge management, honestly.

Not taxonomy.

People.


About the author

Aiver Ilaya is an IT consultant based in Melbourne. For 19 years he’s worked on enterprise digital transformation programs. His experience spans ITSM, HRIS, ERP, CRM, POS, KMS and EDRMS platforms, plus workflow automation. He has worked across government, healthcare, science and medical research, finance, education, telecommunications, infrastructure, aviation, construction and property, technology, retail, logistics, pharmacy, insurance and professional services.

His work covers business analysis, process optimisation, technical writing, knowledge management and UI / UX design. He has worked on multi-million dollar programs running years at a time, for national organisations and global firms, reaching millions of end users.

He has run his own consultancy for 12 years and works with a small number of clients at a time. Available for remote and hybrid contracts.

alisodigital.com

Read More

Related Posts

The Programs I’ve Seen Fail, Failed on People

I’ve been part of digital transformation work for close to 20 years now. Some of it went well. Some of it didn’t, which is more instructive. Looking back at the ones that struggled, the causes were rarely technical. The platform usually worked. The methodology was usually sound. What went wrong

Build the Current State Map Before the Workshop, Not During It

The default way to run a current state session is to get everyone in a room, put a blank wall in front of them, then build the map live. It’s the accepted approach. It’s also the reason process mapping workshops have the reputation they do. Three hours disappear, the loudest