
What's the One Thing Only You Know How to Do?
TL;DR: The one task only you can do is usually the task you can name in seconds without thinking, the thing that stalls or breaks if you’re out for a day, and naming it is the first real step toward documenting it so someone else can run it. The harder part is writing it down in a way someone else could actually follow.
Key Takeaways
The one task only you can do is almost always the first thing that comes to mind when asked, not something that takes reflection to find.
A single point of failure (the one person a task depends on) creates risk whether or not it has ever been named out loud.
Writing down a task is different from documenting it well enough for someone else to run it without you.
The first thing to write down is the trigger, the exact event or condition that starts the task.
A named backup, even one who has never done the task, gives the task an accountable second person.
The common mistake is writing down the steps and skipping the exceptions, which is where the backup usually fails.
What Is a Single Point of Failure?
A single point of failure is a person whose involvement a specific task requires, meaning the task cannot proceed, or cannot proceed correctly, if that person is unavailable. This applies just as often to judgment calls, client relationships, or institutional knowledge that was never written anywhere as it does to technical skill. The task might take five minutes. The risk comes from the fact that only one person can do it, not from how hard it is.
Ask yourself one question: what’s the one task only you can do? That means the task that stops, or gets done wrong, if you’re not the one doing it.
Why the Answer Comes in Seconds
Asked this question directly, an answer tends to surface before the sentence finishes. That speed is worth taking seriously instead of dismissing as an obvious fact already known.
The task that comes to mind fastest is usually the task with the least documentation, and that gap deserves attention first.
The speed of the answer is a signal. It likely means the task has already caused friction, probably more than once. A missed order while the owner was out. A client question nobody else could answer. A pricing exception only one person remembers the logic for. When asked the question directly, the mind hands over the answer without delay.
This matters because it tells you where to start. The biggest risk shows up without a long audit of every process in the business. The task that surfaced first is the one to write down, because it’s the task most likely to cause a stalled order, a duplicated request, or a client getting two different answers the next time the owner is unreachable.
What to Write Down First
Naming the task is step one. Writing it down well enough for someone else to run it is a different and harder step, and it’s where documentation efforts most often fall short. There are four things to capture, in this order.
The trigger. What exact event or condition starts this task? A specific email, a certain dollar threshold, a client saying a specific phrase, a date on the calendar. Vague triggers like “when it comes up” produce vague backups.
The steps. What happens, in order, once the trigger fires? Each step should name an action and what it’s done with or in, not a general description like “handle it.”
The exceptions. What situations break the normal steps, and what gets done differently when they happen? This is the part most often skipped, and it’s usually the part that matters most.
The named backup. Who is responsible for this task when the owner isn’t available? Not “someone will figure it out.” A specific person, named now, before the task is needed.
Element | Question it answers | Common gap |
|---|---|---|
Trigger | What starts the task? | Described too vaguely to recognize |
Steps | What happens, in order? | Written from memory, missing small actions |
Exceptions | What breaks the normal flow? | Skipped entirely |
Named backup | Who runs this without you? | Left unassigned |
A document with steps but no exceptions looks complete until the first unusual case.
The first time a backup hits a situation the document doesn’t cover, they’ll either stall or guess, and both outcomes create rework that has to be cleaned up later.
The Exceptions Problem
Exceptions are harder to document because they’re often judgment calls internalized so completely that making them goes unnoticed. A late fee gets waived for a client with a long history. A rush job gets quoted differently depending on other commitments that week. None of that logic lives anywhere except in the owner’s head, which means a backup following only the steps will either make the wrong call or stop and wait, defeating the purpose of having a backup at all.
Exceptions are the judgment calls that stopped feeling like decisions.
I recommend listing every exception you can think of in one sitting, even ones that feel too obvious to write down, then asking your named backup to read the list and flag anything unclear. If they can’t tell what you meant, the document isn’t done yet.
In Practice: Documenting an Owner-Only Task
Situation: An office manager realizes that only the owner knows how to approve a specific category of refund requests, and the owner has been doing this from memory for years without writing any of it down.
Steps:
Name the task out loud and confirm it’s the one that comes to mind fastest when asked what only the owner can do.
Identify the trigger, which in this case is a refund request over a set dollar amount (example amount).
List the steps the owner currently takes, in the order they happen, from the moment the request arrives to the moment it’s resolved.
Write down every exception the owner can recall, including situations where the normal approval amount or process didn’t apply.
Assign a named backup, specifically the office manager or another staff member, and give them the document to review.
Ask the backup to identify any step or exception that doesn’t make sense without more detail, and revise until it does.
Result: A written procedure exists that names the trigger, the steps, the exceptions, and the backup, and the test of success is whether the backup can describe what they’d do in a new, slightly unusual refund case without asking the owner first.
Common mistake: Writing down the steps and stopping there, leaving exceptions undocumented, which means the backup handles routine cases fine but stalls or guesses on anything unusual.
If it goes wrong: If the backup still has to ask the owner questions every time, the exceptions weren’t actually captured, and the document needs another pass focused only on those gaps.
This is a general illustration, not a specific client result.
FAQs
How do I find the one task only I can do?
Ask yourself the direct question and write down whatever answer comes first, since the speed of the answer is usually the clearest signal of where the risk actually sits. Don’t overthink it or try to find a more sophisticated answer. The task that surfaces immediately has probably already caused a problem during a prior absence, which is exactly why it came to mind so fast.
What is a single point of failure in a small business?
A single point of failure is a task, process, or decision that depends entirely on one person, meaning the business cannot complete it correctly, or at all, if that person is unreachable. It applies beyond the owner, since any role where only one person holds the knowledge or authority to act creates the same risk, though owner-only tasks tend to be the highest-impact version because they often involve judgment calls nobody else has been trusted to make.
Why can owners name their bottleneck task so quickly?
Owners name it quickly because the task has likely already caused friction, a missed request, a client question nobody else could answer, or a delay while something waited for their return. A likely explanation is that tasks tied to past disruption are easier to recall than tasks that run smoothly. That is an assumption to test, not a measured result.
What should I write down first when documenting an owner-only task?
Write down the trigger first, the specific event or condition that starts the task, because a backup can’t act on a process they don’t know how to recognize. Once the trigger is clear, the steps, exceptions, and named backup can follow in order. Skipping the trigger and starting with steps produces a document that makes sense to the person who wrote it and confuses everyone else.
Who should be the named backup for a critical task?
The named backup should be a specific person identified in advance, not a role left to be figured out when the task comes up unexpectedly. The backup doesn’t need to be as fast or as confident as the owner at first. What matters is that they’ve seen the documentation, understand the exceptions, and know they’re the one responsible when the owner isn’t available.
What’s the biggest mistake businesses make when documenting processes?
The biggest mistake is documenting the steps and skipping the exceptions, which leaves a backup able to handle routine cases but unable to handle anything unusual. Exceptions often live only in the owner’s head as judgment calls made so often they stop feeling like decisions. Leaving them out produces a document that looks thorough and fails the first time something doesn’t go exactly as expected.
How often should I update the documentation for an owner-only task?
Update the documentation whenever the trigger, steps, or exceptions change, rather than on a fixed schedule that may not match how often the task itself shifts. A task tied to pricing, for example, needs a review whenever pricing changes. Treating the document as a one-time project rather than something that gets revisited is a quiet way for it to go stale without anyone noticing until the backup hits a gap.
Next Steps
You already know the task. The question is whether you’ve written it down yet, or whether it’s still sitting only in your head. What’s yours? Reply in the comments with the one task only you can do.
