In Part One of this series, we named the most invisible operational risk most organizations carry: tribal knowledge. The critical processes that live only inside the heads of your longest-tenured employees — undocumented, untransferable, and one departure away from gone.
Part Two goes one layer deeper. Because once you understand that tribal knowledge is a problem, the next question is: where, specifically, is the fragility hiding in your organization right now? Not in theory — in your org chart, your onboarding experience, your calendar.
The answer is usually closer than you think. Three of the most revealing diagnostic signals are ones most leaders walk past every day without reading them.
1. Dependency Mapping: The Question You're Not Asking
Start with a single question for every core function in your organization: If this person were unavailable tomorrow, who else could perform this work?
Not after a two-week transition. Not with extensive hand-holding. Tomorrow.
If the honest answer is "no one" or "only one other person, and they'd struggle," you have identified a single point of failure. This is dependency mapping in its most essential form — and it is one of the most efficient operational diagnostics available to any leader.
Most organizations discover, when they run this exercise honestly, that they have more single points of failure than they realized. The person who manages the vendor relationships. The one who knows how the grant reporting gets submitted. The colleague who holds the institutional memory of every client's specific preferences. These functions are not typically flagged as high-risk — until the person holding them is unavailable.
Cross-training is the structural response to dependency mapping, but I want to name what it actually is: it is not a human resources initiative or a professional development exercise. It is risk management. The goal is not to give your team members growth opportunities — though that may be a benefit. The goal is to ensure that no single absence creates an operational crisis.
The dependency map makes invisible risk visible. Once you can see it, you can sequence a response: identify your highest-stakes single points of failure first, and build redundancy there before you lose the window to do it proactively.
2. The Onboarding Test: If a Person Can't Leave, the Process Isn't Documented
There is an important distinction that most organizations blur: HR onboarding is not the same as operational onboarding. HR onboarding covers benefits, compliance, and organizational structure. Operational onboarding covers how work actually gets done in your specific team, department, or function — and in most organizations, that knowledge is transferred informally, if at all.
The test is simple: Could a competent new team member independently perform the core functions of their role within their first 30 days — without requiring a tenured colleague to walk them through it repeatedly?
In most organizations I've worked with, the honest answer is no. And that gap is not a training problem. It is a documentation problem.
The COVID-19 pandemic made this failure acute in ways that could no longer be ignored. Organizations that had always relied on informal knowledge transfer — learning by proximity, asking the person in the next office, absorbing culture through observation — suddenly found themselves onboarding new employees in a fully remote environment. There was no next office. There was no ambient institutional knowledge to absorb. New hires were alone with incomplete information, and the gaps became immediate and visible.
The organizations that navigated that transition well were the ones that had already done the work of writing things down: SOPs for recurring processes, operations directories that documented where information lived, clear escalation paths that didn't require knowing the right person to ask. The organizations that struggled were the ones whose knowledge had never left the building — metaphorically or literally.
If your onboarding experience requires a specific person to walk new team members through it, what you have is not an onboarding process. You have a dependency. And that dependency is costing you more than you can see — in senior staff time, in new hire errors, in the invisible slowdown that comes from every team member who spent their first 90 days reconstructing knowledge that should have been available from day one.
The fix is not a 200-page manual. It is an operations directory: a searchable, maintained collection of the recurring workflows, decision frameworks, and institutional knowledge that your team needs to function — built so that a capable person could navigate their role without needing to find the right colleague to ask.
3. Decision Bottlenecks: When Waiting Is the Workflow
The third signal is the one that tends to frustrate teams most — and that leadership is often the last to see. It is the decision bottleneck: the pattern where work consistently stalls because a particular person needs to approve, review, or weigh in before the team can move forward.
Bottlenecks are not always obvious, because they often masquerade as thoroughness, accountability, or quality control. A leader who reviews everything before it goes out is seen as conscientious. A manager who is involved in every client decision is seen as hands-on. What is actually happening is that the organization has built an operational structure where nothing moves without a specific individual's approval — and that is a form of operational debt that compounds over time.
The diagnostic question: What decisions or approvals consistently wait for the same person? Make a list. If you have more than two or three recurring situations on that list, you have a systemic bottleneck, not a series of individual judgment calls.
Here is the operational principle I apply in my engagements: If a situation has occurred three or more times and required the same type of decision each time, it should be systematized. That means creating a documented framework — a decision tree, an escalation matrix, a standard operating procedure — so that the next occurrence can be handled consistently, without requiring a senior leader's intervention.
The cost of not doing this is real. Every time a recurring decision routes back to the same desk, it consumes that person's time and cognitive capacity on something that has already been decided — just never written down. It slows the team waiting for the answer. And it creates a false sense of control: the leader feels involved, but what they are actually doing is serving as a manual lookup table for decisions that could be handled by anyone with access to a documented framework.
Three recurrences is not an arbitrary number. It represents the point at which a situation has demonstrated a pattern. A one-time issue is an anomaly. Two occurrences could be coincidence. Three is a signal that your organization has a recurring operational condition without a written response — and every future occurrence is a cost that documentation could have prevented.
What These Three Signals Have in Common
Dependency gaps. Broken onboarding. Decision bottlenecks. These are not separate problems requiring separate solutions. They are three manifestations of the same underlying condition: knowledge and decision-making authority concentrated in people rather than distributed through systems.
When that concentration exists, the organization becomes only as resilient as its most indispensable person on any given day. That is not a leadership model. It is a fragility model — and it tends to feel functional right up until the moment it doesn't.
The operational triage is not about finding what is broken in a catastrophic sense. It is about finding what is hiding — the quiet concentrations of risk that are visible if you know what you're looking for, and preventable if you act before a departure, a crisis, or an audit forces the issue.
Frequently Asked Questions
What is dependency mapping in business operations?
Dependency mapping is the practice of identifying, for each core business function, who is responsible for executing it — and who else could perform it if that person were unavailable. If the honest answer is "only one person," the organization has a single point of failure. Dependency mapping makes invisible operational risk visible before a departure, illness, or crisis forces the issue. It is a foundational step in operational continuity planning.
Why does tribal knowledge fail during remote or new employee onboarding?
Traditional onboarding relies heavily on informal knowledge transfer — observing how things are done, asking colleagues, absorbing institutional culture through proximity. Remote onboarding and high-turnover environments break this model entirely. When new employees cannot learn by osmosis, the absence of written SOPs and operations directories becomes acute. The COVID-19 pandemic exposed this gap widely: organizations that had never formalized their processes found themselves unable to onboard effectively in a distributed environment, because the "how" had never been written down.
How do you identify decision bottlenecks in your organization?
The clearest signal is a recurring pattern: the same person is consistently required to approve, review, or resolve a particular type of situation before the team can move forward. Track which recurring questions or approvals route to the same desk — that list is your documentation backlog. When work stalls whenever that individual is unavailable, even for decisions that have been made the same way before, the bottleneck has become a structural dependency.
What is the difference between a one-time issue and a documentation gap?
A one-time issue is an anomaly — unexpected, context-specific, and unlikely to recur in the same form. A documentation gap is a situation that has happened before, required the same type of judgment or escalation, and will almost certainly happen again. The practical test: if a situation has already occurred three or more times, it is no longer a one-time issue. It is a recurring operational pattern that has not yet been systematized — and the absence of a decision framework, escalation matrix, or SOP is costing the organization time and consistency with every recurrence.
How many times does a situation need to recur before it should be systematized?
Three is the practical threshold. If a situation has occurred three or more times and required the same type of decision, escalation, or workaround each time, it should be systematized — through a decision tree, escalation matrix, or standard operating procedure. This allows future occurrences to be handled consistently without requiring a senior leader's intervention, a long email thread, or a call to the one person who "knows how we handle this."
Ready to Surface What's Hiding in Your Operations?
Whether it's a dependency map, a documentation audit, or an end-to-end operational review — this kind of work is what I do. If something in this article named a problem you've been circling, I'd like to hear about it. I review every submission personally.
Start the Conversation