When Critical Knowledge Lives in a Few Heads: How to Reduce Key-Person Dependency
Most owners know who the indispensable people are. They are the ones colleagues message first, the ones who "just know" why a customer's order is handled a certain way, the ones whose holiday plans make everyone slightly nervous. The risk is rarely discussed until a resignation, an illness or a retirement turns it into an emergency.
This article looks at the four problems that Lakhlani International's Knowledge Maturity Management page describes, and offers a practical way to rank the risk, capture what matters and make handovers dependable. It does not promise that dependence on individuals can be removed. It aims to reduce it where it hurts most.
The four problems, and why they cluster
The capability page names four related problems.
- Critical operating knowledge exists only in individuals' heads.
- Documentation is fragmented, outdated or difficult to find.
- Onboarding and handovers depend heavily on informal explanations.
- Teams repeat work because decisions and lessons are not captured.
They reinforce each other. When documentation is poor, people go to the expert. The expert becomes busier and has less time to write anything down. New joiners learn by interrupting the expert. And because decisions are not recorded, the same debates are repeated the following year.
What the research suggests, with caveats
Two pieces of evidence help frame the issue, and both need care.
Nonaka and Takeuchi's distinction between tacit and explicit knowledge explains why the problem is hard. Explicit knowledge, such as a procedure, can be written down. Tacit knowledge, such as judgement built from years of cases, is harder to put into words, and some of it cannot be fully converted. Knowledge moves through four modes: socialisation (working alongside), externalisation (putting it into words), combination (combining documents) and internalisation (learning by doing). A plan that relies only on writing documents addresses one mode out of four.
The McKinsey Global Institute's 2012 report "The social economy" estimated that high-skill interaction workers spend nearly 20% of their time looking for internal information or colleagues who can help. That figure is an estimate that combines earlier IDC data with McKinsey's own data, and it covers a specific group of workers. Treat it as a reason to measure your own search and interruption time, not as a number to quote about your company.
There is no recent, independent research quantifying knowledge loss in small and mid-sized businesses in India that this guide can cite, so it makes no such claim.
Step 1: Find the people and the knowledge at risk
Start by listing the people whose absence would cause the most disruption, and the workflows they hold together. A short interview or questionnaire with managers works well. Ask:
- →Who would we call first if this went wrong?
- →Which tasks does only one person know how to do?
- →Which customers or suppliers deal with only one of us?
- →Which decisions have we made that nobody could explain now?
The capability page begins with exactly this step: identify critical workflows and knowledge risks.
Step 2: Rank the risk
Not all knowledge deserves the same effort. A simple scoring helps you choose. For each candidate, rate three things from 1 to 5.
Risk scoring factors
- →Business criticality. How badly does the business suffer if this is unavailable? A higher score means more damage.
- →Concentration. How many people hold it? A higher score means fewer people, with one person at the top of the scale.
- →Replaceability. How long would it take someone else to learn it? A higher score means longer.
Multiply or add the scores and work down the list. The method is a judgement aid, not a standard, and the value lies in the conversation it provokes. The capability page speaks of prioritising knowledge assets by business criticality, which is the same idea.
Do the top three to five items first. Trying to capture everything usually captures nothing.
Step 3: Choose the right capture method for each
Match the method to the kind of knowledge.
Matching methods to knowledge types
- →Procedures with clear steps. Use a short written procedure or checklist, or a screen recording.
- →Judgement and exceptions. Walk through real cases with the expert and record the reasoning, not just the answer.
- →Relationships and context. Use shared introductions, joint meetings and account notes.
- →Decisions and lessons. Keep a brief decision record: what was decided, why, alternatives and the date.
- →Skills learned by doing. Use shadowing, then supervised practice (reverse shadowing).
These map onto the four modes. Written procedures and decision records externalise knowledge. Shadowing and joint work socialise it. Practice internalises it. A good plan uses more than one.
Step 4: Fix handovers and onboarding
Where handovers depend on informal explanations, define a minimum standard. A handover note should cover:
- →What the role or task is for, and who depends on it.
- →Regular activities, with timing.
- →Exceptions and how they are usually handled.
- →Contacts, systems and access.
- →Open issues and recent decisions.
- →Where to find more detail and who to ask.
For onboarding, build a path from the top-ranked knowledge. New joiners should know where to look, whom to ask and what to read first, rather than collecting answers through interruptions.
Step 5: Assign ownership and review
Captured knowledge decays. Without an owner, documents go stale and trust in them erodes, which brings back the habit of asking the expert. Each important knowledge asset should have:
- →A named owner.
- →A review date, or a trigger such as a process change.
- →A clear location.
- →An agreed access rule: who can see and edit it.
The page calls for governance, review cycles and access principles, and for measuring adoption and revising based on actual use.
Be realistic about limits
The capability page notes that effectiveness depends on participation, ownership and ongoing maintenance. A few cautions follow.
- →Experts may resist. Some fear that sharing knowledge reduces their value. Leaders should make it clear that documenting expertise is recognised and rewarded.
- →Not everything can be written. Some judgement will remain tacit. Use shadowing and pairing for that.
- →Time is the real cost. Experts are busy. Budget their time for this work or it will not happen.
- →Documents can mislead if stale. An outdated procedure is worse than none.
An illustrative scenario (hypothetical)
Consider a hypothetical engineering services firm where one senior estimator prices every complex bid. The owner scores the risk: high criticality, concentration of one and a long learning time. The firm records three of the estimator's recent bids, capturing the reasoning behind each assumption. A junior colleague prices the next bid under supervision and the estimator reviews it. A short checklist and a decision log of past pricing exceptions are written and given an owner and a six-month review.
The estimator is still valuable, but the firm no longer depends entirely on one person's availability. This scenario is invented for illustration and does not describe any client.
Next step
Run the ranking exercise with your management team this week. The disagreements will show you where the real dependencies lie.
If you would like to scope an assessment of your priority knowledge areas, the Knowledge Maturity Management capability sets out how the work runs. For process change more broadly, see the Digital Transformation capability.
Knowledge Maturity Management
A scoped inventory of priority knowledge domains and an ownership and review model you can run yourself.
Explore the capabilityWhat is knowledge maturity management?
The introduction to this series: tacit and explicit knowledge, ISO 30401 and where to start without buying software.
Read the introductionDigital Transformation
For the process and platform changes that usually accompany a knowledge programme.
Explore Digital TransformationFrequently asked questions
What is key-person dependency?
It is a situation in which important knowledge, relationships or skills are held by one or a few individuals, so their absence would seriously disrupt the business.
How do I know which knowledge to capture first?
Rank candidates by business criticality, how few people hold the knowledge and how long it would take to replace. Start with the highest-scoring few.
Is it enough to ask the expert to write everything down?
Usually not. Some knowledge is tacit and is better transferred through walkthroughs, shadowing and joint work. Writing helps with procedures and decisions, but it covers only part of the picture.
How do we keep documentation from going out of date?
Assign an owner and a review date or trigger for each important document, store it where people will look and track whether it is used. Retire documents nobody uses.
What should a good handover include?
The purpose of the role, regular activities, exceptions, contacts and access, open issues, recent decisions and where to find more. Test it by having the person receiving it do a real task.
Can documentation completely remove dependence on individuals?
No. It can reduce it, especially for critical processes, but some judgement stays tacit and results depend on participation, ownership and maintenance.
References
- →Nonaka, I. The Knowledge-Creating Company. Harvard Business Review (1991). https://hbr.org/2007/07/the-knowledge-creating-company
- →McKinsey Global Institute. The social economy (2012). https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-social-economy
- →Lakhlani International. Knowledge Maturity Management capability page. https://lakhlaniinternational.com/capabilities/knowledge-maturity-management