What stops if one person goes

You cannot say which work rests on one person

You can see which work has a single owner, and what kept running while that person was away. Both are named, so it is a list to act on.

Powered By

Continuity

Skills

Schedules

Workflows Team Dashboard Workspaces

The Challenge

You know the company depends on a few people and you cannot say which work, exactly
Somebody goes on leave and you find out what they were holding up by what stops
The automations that matter most were built by whoever needed them, and nobody wrote that down
A person resigns and the handover is a meeting where they try to remember everything
You are asked about key-person risk and the honest answer is a feeling about three names

The Runwork Solution

Work with a single owner is listed, ranked by how much it actually runs, so the risk is specific rather than general
A ledger of what ran while its owner was away, which is the difference between a process and a person with a habit
Knowledge flows show a skill moving from the person who made it to the people using it, so you can see what has spread and what has not
What one person worked out is saved to the workspace rather than to their laptop, so a handover starts from a thing rather than from a memory
The report says which runs it could not attribute, rather than assigning them to somebody to make the picture look complete

What your team gets

Six things, one owner

Not "we depend on Ana" but "these six automations have a single owner, and this is how often each one runs". The risk becomes a list of work rather than a judgement about a person.

What ran while they were away

The work that kept going through somebody's two weeks off. That is the clearest evidence there is that a process is a process, and it is the hardest thing to fake.

A skill that travelled

Somebody built it, four colleagues now use it. Knowledge that moved from one person to several is the thing that actually reduces the risk, and it is visible rather than assumed.

A handover that starts from something

When somebody leaves, what they built is in the workspace and still running. The conversation is about intent and edge cases rather than about reconstructing what existed.

The gaps it admits to

Runs it could not attribute to anybody are reported as unattributed rather than assigned to somebody to complete the picture. An inventory that guesses is worse than one with a hole in it.

What this is not

It is not a ranking of people and there is no score against anybody's name. What is measured is the work: how concentrated its ownership is, and how much of it runs without its owner present.

A Day in the Life

The question you have never been able to answer. If one person left tomorrow, what stops? You have a feeling about three names and no way to turn that into anything you could act on.

Monday. The list is about work, not people. Six automations have a single owner, ranked by how often each one runs. That is a specific problem with a specific fix, rather than an anxiety about a person.

Tuesday. Two weeks of evidence. Somebody was on leave last month and you can see what kept running while they were away. That is the difference between a process and a habit, and it is not something anybody can perform for a review.

Wednesday. Something that spread. A skill one person built is now used by four colleagues. The risk on that one has already gone down, without a project, because the knowledge moved.

Thursday. A resignation. What they built is in the workspace and still running. The handover is about intent and edge cases rather than an hour of them trying to remember what they set up in March.

And nobody was scored. There is no ranking of people in any of this. The concentration is a fact about the work, and the work is the thing you can change.

Frequently Asked Questions

Does this rank people by how critical they are?

No. It ranks WORK by how concentrated its ownership is. A person is named as the owner of a thing, which is attribution, and never scored, rated or compared. That distinction is not cosmetic: a report that reads as a list of weak links is a report people quietly stop feeding, and then it stops being true.

What is the strongest evidence that something is a real process?

That it ran while its owner was away. Offline runs and runs during somebody's absence are the hardest thing in this whole library to fake, because nobody can perform them for a review. A process that only ever runs when one person is at their desk is a habit with good documentation.

What do I actually do about a single-owner automation?

Usually the answer is not a project. Getting the work saved as a shared skill rather than kept in one person's setup, and getting a second person to actually use it, moves it out of single ownership. The report shows you when that has happened, because the knowledge flow becomes visible.

What does the report admit it does not know?

Runs it could not attribute to anybody are reported as unattributed rather than assigned to somebody to make the totals look complete. That matters more here than in any other report: a continuity picture that quietly fills its own gaps is worse than one with a visible hole, because you would plan around it.

Can I see what would break if a specific person left?

You can see what they own and how much of it runs, which is the answerable version of that question. What we do not do is model the aftermath of a departure, because that would be a projection rather than a measurement, and the whole library is built the other way round.

Related Use Cases

Keeping What You Learn What AI Does Share AI Skills

Start keeping what your team figures out.

Start a 14-day free trial. No credit card required.

Start Free Trial Explore More Use Cases