What Operational Resilience Means When It All Runs Through You
How to find the parts of your work that only run because you are there, and hand one of them off.
Operational resilience is whether the work keeps running when a person, a system or a supplier drops out.
At your desk it means something smaller and a lot more personal. How many things stop if you’re out for a week?
If the honest answer is most of them, you’re not indispensable. You’re a single point of failure, and it’s quietly costing you leave, promotions, and the right to be ill without apologizing.
We will give you a map to find yours, and a way to hand one off before you need to.

Key Takeaways
- Count what breaks when you stop. Operational resilience at your level is one question, and the map below turns a vague sense of being overloaded into a list you can work through.
- Being the only one who can do something isn’t job security. It’s why a woman ends up unable to take leave, and why she gets left in a role rather than moved up out of it.
- Hand off the smallest one first. One process taught to one person beats a plan to document everything by the end of the quarter.
- Ask in operational terms rather than personal ones. We have no cover for this gets a different answer from I’m overwhelmed, and it’s the same request.
What is operational resilience?
Operational resilience is an organization’s ability to keep its important services running through disruption, and to recover them when they break. That’s the broad meaning, and it applies to a payment system, a supply chain and a hospital ward the same way.
The Bank of England and the FCA define operational resilience as the ability of firms and the whole financial sector to prevent, adapt, respond to, recover from and learn from operational disruption. That’s a useful definition for a bank and a useless one for a Tuesday.
On this page we use a tighter version of the same idea. We’re looking at one desk rather than one organization, and specifically at the work that only runs because you’re there.
The narrowing is deliberate. The organization-wide version is a policy question, and if you don’t write policy it hands you nothing you can act on.
The smaller version does hand you something. You can count your own single points of failure in an afternoon, and you can hand one of them off without waiting for a restructure.
So here’s the one we’ll use. Operational resilience is the number of things that keep working when you’re not there.
Why the work runs through one person
The mechanism is dependency. Every process you own has a shortest path, and if that path runs through exactly one person, it has a single point of failure.
Most of the time nobody notices, because the single point of failure keeps turning up. That’s what makes it invisible right up until the week she doesn’t.
Why this lands harder on women
The tasks that create single points of failure aren’t usually the ones on your job description. They’re the ones that got handed to you because you noticed they weren’t being done.
The scheduling, the onboarding, the client who only replies to you, the spreadsheet nobody else can read. None of it shows up in a performance review, and all of it stops when you do.
So the dependency builds in the part of your work that’s least visible and least rewarded. You carry the risk without the credit, and the reward for absorbing it is more of it.
Babcock, Recalde, Vesterlund and Weingart found in the American Economic Review in 2017 that women more than men volunteer for, are asked to do, and accept tasks with low promotability, and argued that if women hold less promotable work than men, they will progress more slowly in organizations.
We’re not going to hand you a percentage here. Open the map below and count your own, because yours is the only number that changes what you do on Monday.
Map your single points of failure
Write down everything you own that other people rely on, then answer three questions about each one. That’s operational resilience as a list rather than a concept, and the flattering answer in the third column is what keeps you stuck.
For the last column, count the sessions it would take to teach it, at roughly an hour each. Two sessions is two hours.
| What I own | What breaks if I am out for a week | Who else could do it today | Hours to hand it over |
|---|---|---|---|
| Monthly billing run | Invoices go out late, so the cash comes in late | Nobody | 7 |
What your answers mean
Four outcomes, and a move for each. Work the column you filled in, not the one you wish you had.
- Nobody else could do it, and handover would take more than a couple of hours. That’s your first one, and don’t start by writing it down. Book sessions where they run it and you watch, not where you talk them through it.
- Nobody else could do it, but handover is under two hours. Do it this week, because something that takes ninety minutes to teach shouldn’t be holding your leave hostage.
- Someone else could do it, but only with you on the phone. Write down the six things they call you about, which isn’t documentation, it’s the answers to the questions they actually ask.
- Someone else could genuinely do it without you. Nothing to fix here, so cross it off and stop carrying it as a risk you’re still responsible for.
If you’ve got seven minutes rather than ninety, do this instead. Pick the one thing you’d be most anxious about if you were in the hospital tomorrow.
Send one person the login and the two sentences they’d need to keep it moving. That isn’t a handover, but it’s cover, and it takes less time than worrying about it does.

Working through one of your own
Here’s a composite: Priya runs the monthly client billing, it takes her 4 hours, and nobody else has ever touched it.
| What | How much |
|---|---|
| What the run costs her now (4 hours x 12) | 48 hours a year |
| What the handover costs | 7 hours, once |
| What she gets back once they split the runs | 24 hours a year |
| When it pays back (7 hours / 24 a year) | 3.5 months |
What changes the answer
If the run is quarterly rather than monthly, the arithmetic shifts. Four runs a year split in half saves 8 hours instead of 24, so the same 7-hour handover takes almost a year to pay back.
Do it anyway, and do it for the leave rather than for the time. The hours were never the real cost.
If nobody on your team has 7 spare hours, the handover isn’t yours to solve alone. A team with no slack has no cover for anybody, and that’s worth saying upward in those words.
How to build cover over the next few months
The map tells you where you stand today, and this is the order the work arrives in afterward. Each step carries the trigger for moving to the next one, so you’re not guessing whether you’re ready.
This is the operational version of a build, kept to your own desk. The general one sits in how to build resilience.
- Start with the lowest number in the last column of your map. Move to step two once you’ve picked exactly one and told the person you’re going to teach them.
- Teach it live rather than writing it up. Move to step three when they’ve run it twice without you in the room.
- Get the backup named out loud to whoever schedules the work. Say: “If I’m out, [name] runs the billing. Can we put that in the schedule?” Move to step four when someone other than you has called them the backup.
- Write down what they asked, as in branch three of the map. Move to step five when a full run turns up no new questions.
- For anything left where there was nobody to teach, put the gap in writing and date it. The sequence is finished when that record exists, whatever comes back.
What goes wrong when you try this
Operational resilience fails in the handover far more often than in the plan. One version is documenting instead of teaching, and a written process nobody has ever run isn’t cover, it’s a file that’ll be out of date before anyone opens it.
The second is handing off the interesting part and keeping the tedious one. Whatever you keep is what still stops when you stop, so hand over the thing that runs to a deadline.
The third is being told there’s nobody to hand it to. Sometimes that’s true, and the response is to put the gap in writing rather than absorb it quietly for another year.
One line where this stops being an operations problem. If you’re being refused leave you’re legally entitled to, or told the work can’t run without you as the reason for refusing it, that’s an employment matter.
Talk to an employment attorney licensed in your state before you accept the answer, and go before you agree to anything in writing.
Frequently asked questions
What if I am the only person who does my job?
This one assumes an employer and colleagues, so a freelancer or sole proprietor won’t find the answer here. If you’re employed, the whole role is the single point of failure, so name the three things that break first and get cover agreed.
Won’t making myself replaceable get me replaced?
Being the only one who can do something protects the task, not you, and it’s how people get left in a role instead of moved up. Cover is what makes you promotable.
What if my manager has no interest in this?
Take it as a risk to the work rather than a request for help, because those get different answers. One sentence naming what stops and for how long is usually enough to start it.
I’m remote and nobody sees what I hold. Does this still work?
It works better, because remote work leaves a written trail you can point at. Pull the last three months of messages that came only to you and your map is already half filled in.
What if I try this and nothing changes?
Then you have a dated record of what depends on one person, which is worth having when the leave request or the promotion conversation comes round. Keep it where you can find it.
Where to go next
Pick one row from the map and put a date on it. Not the biggest one.
If the map came back full because you keep taking things on, setting boundaries is the harder half of this. A handover only holds if nothing new lands in the gap it made.
If your third column said nobody because your employer has no cover for anyone, that’s organizational resilience, which reads the employer rather than your own desk.
For the wider picture this sits inside, our women’s resilience pillar covers the whole set, and types of resilience shows where operational sits next to the emotional and financial kinds.
Not sure where your load is heaviest? The work-life balance quiz takes a few minutes and gives you somewhere to start.
Send this to the woman covering for everyone
Think of the woman on your team who hasn’t taken a full week off in two years. Not because she doesn’t want to, but because she’s the only one who knows how something works.
Send her the map and tell her which row you think she’d fill in first. A bare link asks her to work out why you sent it.
And if that woman is you, send it to the person you’d have to hand something to. It’s a shorter conversation than the one you’ve been putting off.