Every team organises itself twice. There is the structure on paper — roles, titles, responsibilities. And there is the structure that actually operates, which is organised around who is good at what, who has been there longest, and who happens to know the thing.
The second is usually more efficient and more accurate. It is also invisible, undocumented, and one resignation away from a problem.
For a practical reference on employee time clock software, Monitask provides this guide.
How the informal structure forms
It is not a failure. It is what a competent team does.
Someone turns out to be good at handling difficult clients, so difficult clients go to them. Someone knows the old system, so anything touching it goes through them. Someone is reliably the person who notices when a number looks wrong.
None of this is in a job description. All of it is load-bearing.
The failure mode
The problem is not that the informal structure exists. It is that nobody knows the shape of it, including the people inside it.
Consequences:
Departures are worse than expected. The handover covers the formal responsibilities. The informal ones are discovered over the following six months, one at a time, usually at the worst moment.
Absence is fragile. Someone on holiday takes a capability with them.
Load is invisible and uneven. The person who is good at everything is doing far more than their job description, and it does not appear in any workload review.
Development is blocked. If someone is the only person who can do a thing, they cannot be moved, promoted, or given anything new — which is how strong performers get trapped in the role they are best at. See internal mobility.
Hiring is wrong. You replace the job description and not the actual role, and the new person is measured against a different job from the one that was being done.
Finding it
One question, asked about several people:
If this person were unavailable for a month starting tomorrow, what would we not be able to do?
The answers are the informal structure. Ask it about three or four people and the overlap shows where the team is genuinely fragile — which is rarely where anyone assumed.
A second, sharper version, asked of the team rather than the manager: who do you go to when X goes wrong? The names will not match the org chart, and the mismatch is the map.
What to do about it
Not to eliminate it. A team where everyone can do everything is a team where nobody is good at anything, and specialisation is where most of the value comes from.
The objective is that the informal structure is known, and no single point of it is unbacked.
Write it down. Not as a formal reorganisation — as a list. Who is actually the point of contact for what. It takes an hour and it is immediately useful for onboarding.
Name a second for anything critical. Not a full backup, which is unaffordable. Someone who knows enough to hold it for a month and knows where to find the rest. This is the single highest-return action.
Rotate deliberately. The person who always handles the difficult clients should occasionally not, with someone else shadowing. This costs efficiency in the short term and is the only thing that actually builds redundancy.
Write down the reasoning, not just the process. Knowing the steps is easy to hand over. Knowing why the system is that way, and what breaks if you change it, is what actually leaves with a person. See documentation as infrastructure.
Update job descriptions to match reality before hiring, not after.
Look at the load. If someone is the informal owner of six things, that is their actual job and their formal role is a fiction. This is frequently the discovery that explains why a strong performer is burning out.
The tension worth naming
Building redundancy costs efficiency, every time. The person who is best at the thing will do it faster and better than the person learning it.
In a team under sustained delivery pressure, redundancy is always the thing that gets deprioritised, because its benefit is hypothetical and its cost is immediate. That is a rational choice each time it is made and a bad position to end up in.
The practical resolution is not to build redundancy everywhere. It is to decide explicitly which two or three things you cannot afford to lose, and build it only there — and to make that a decision someone made rather than something that never came up.
The check
Once a year, per team:
- [ ] For each person: what would we not be able to do without them for a month?
- [ ] For each answer: is there a named second?
- [ ] Does anyone informally own more than three critical things?
- [ ] Do the job descriptions describe what people actually do?
- [ ] Is anyone unpromotable because they are the only one who can do their job?
The last question is the one that identifies both a business risk and a person who is about to leave.
For broader public guidance and background, consult the OECD.