Your Slack channels are the most honest documentation audit you'll ever get. Every time someone asks "how do I...?" or "who owns...?" or "why does...?" and a colleague answers, you're watching a piece of undocumented knowledge get transmitted person-to-person — and then vanish back into the scroll. The repeated questions are a live, unfiltered signal of exactly where your documentation debt is. Most teams let that signal disappear. Learning to read it, and to capture what it reveals, is the foundation of an actual knowledge culture.
Repeated questions are debt, made visible
Documentation debt is invisible by nature — you don't see the doc that isn't written. But its symptom is loud: the same question, asked over and over, answered ad hoc each time by whoever's around. Each repetition is the interest payment on that debt — an expert's time spent re-answering something that should have been findable.
So the pattern to watch for isn't any single question; it's recurrence. A question asked once might be a one-off. A question asked monthly, by different people, is a documentation gap with a measurable cost. Reading your channels for recurring questions gives you a ranked list of what to document — ordered by how often it actually comes up, which is far better prioritization than guessing.
What different question types reveal
The kind of question tells you what type of knowledge is missing:
- "How do I...?" — missing procedural documentation. A process people need but can't find. Usually the easiest to fix and the highest-frequency.
- "Who owns / who do I ask about...?" — missing ownership/relational knowledge. People can't find the person or the responsibility map. Signals org-knowledge gaps.
- "Why is... the way it is?" — missing decision context (institutional knowledge). The most valuable and most-lost type. When people repeatedly ask why something works a certain way, the reasoning was never recorded.
- "Is it just me or is X broken/confusing?" — signals either a real problem or a documentation clarity gap, and often both.
Categorizing recurring questions this way turns "we should document more" into a targeted plan: these procedures, these ownership maps, these decisions.
From reading the signal to closing the debt
Spotting the debt is step one; a knowledge culture is what systematically closes it. The practices that work:
Capture answers at the moment they're given. The single highest-leverage habit: when a good answer gets posted in a thread, don't let it evaporate. The exchange is documentation — preserve it. This is far more sustainable than scheduling separate documentation work, because the answer already exists; you're just keeping it instead of discarding it.
Make the captured knowledge searchable. Capturing threads into a place nobody searches just moves the debt. The payoff comes when those answers feed a knowledge base people can query — so the next person with the same question finds the answer instead of asking again. This is the loop that actually reduces repetition: capture the answer once, make it findable, and the recurring question stops recurring. A tool like Kognita can ingest the channels and docs where these answers live and surface them as searchable, cited responses, closing the loop without asking anyone to write formal docs.
Reward answering-into-the-system, not heroic re-answering. Cultures quietly incentivize the wrong thing: the person who's always available to answer in Slack is celebrated, while the invisible gap that makes them necessary persists. Shift the recognition toward making knowledge reusable — the answer that gets captured and found by ten future people is worth more than the answer given ten separate times.
Close the loop visibly. When a recurring question gets documented and stops recurring, notice it. Making the debt-reduction visible reinforces the behavior and shows the culture is working.
The lightweight metric
You don't need heavy tooling to start. A simple, powerful practice: periodically scan your busiest channels (or their search history) for repeated questions, tally them, and treat the top of that list as your documentation backlog. Over time, watch whether specific questions stop appearing after you document them — that's your knowledge culture producing measurable results. Rising repetition means debt is accumulating faster than you're paying it down; falling repetition means the system is working.
The reframe
A knowledge culture isn't about mandating documentation or nagging people to write things down — those fail because they fight human incentives. It's about treating the questions people already ask as a signal, capturing the answers people already give, and making them findable so the same question doesn't need asking twice. Your Slack threads are already telling you exactly what's undocumented and how much it's costing you. The culture is just the discipline of listening to that signal and turning the ambient flow of questions-and-answers into durable, searchable memory — so your experts spend less time as human search engines and more time on the problems only they can solve.