3 September 2026LANTEK Computers
A lot of businesses arrive at a genuine dilemma. They have somebody in-house who is good, who knows the business properly, and who is the reason things work. They also have a problem that person cannot solve, which is that there is one of them.
The two options usually presented are hire another one, which is expensive and hard, or outsource, which means losing the person who knows everything. Co-managed is the arrangement that is neither.
Since it is the obvious worry, let us deal with it first: this is not a staged route to replacing your in-house person. If a provider’s real plan is to learn the estate and then propose taking it over entirely, that is outsourcing with extra steps, and it is a fair thing to ask them directly.
What the in-house person cannot do, structurally
None of this is about capability.
They cannot be in two places. A project and a helpdesk compete, and the helpdesk always wins because it is in front of them. So the project that would actually improve things never finishes.
They cannot cover leave. Two weeks in December is two weeks with nobody, and December is when a lot of businesses here are busiest.
They cannot be on call indefinitely. One person as the permanent after-hours contact is an arrangement with an expiry date, and the expiry is usually them leaving.
They cannot be specialist in everything. Nobody is current on networking, security, cloud identity, backup and the line of business application simultaneously.
They are a single point of failure for knowledge. If it is all in their head, the business has an undocumented dependency on one employment relationship.
How the work usually divides
The split that works is not by importance. It is by whether the work benefits from being local and contextual, or from being systematic and covered.
The in-house person keeps: everything requiring knowledge of your business, relationships with staff, the line of business application, walk-up support, vendor relationships they own, and being the person who decides what matters.
The provider takes: monitoring around the clock, patching on a cycle, backup verification, after-hours and holiday cover, escalation for specialist problems, the security functions that need someone watching continuously, and maintaining documentation so it exists outside one person’s memory.
Shared: projects, where the provider supplies capacity and specialism and the in-house person supplies context. This is where the arrangement earns its keep, because it is how the project that never finishes finally does.
What has to be agreed for it to work
The failure mode of co-managed is not technical. It is two parties each believing the other is handling something. These are the things to nail down in writing.
Who owns what, explicitly. A written list, by function, of who is responsible. Ambiguity anywhere on it will eventually be the gap something falls into.
Where tickets go. One system that both parties see, not two queues and an email chain. Without this, nobody can tell what the actual workload is.
Who can change what. Both parties having administrative access needs a rule about change communication, or you get the situation where one side spends a morning diagnosing something the other side changed.
What escalation means in both directions. When does the in-house person hand up, and when does the provider hand back?
Documentation, and where it lives. Shared, current, and belonging to you. This is one of the main benefits, so do not let it be the provider’s internal notes.
What happens on leave. Specifically. Does cover start automatically or does somebody request it?
Who it suits
It fits a recognisable shape of business: enough staff that there is real IT work, one or two capable people in-house, an estate with some complexity, and a reason to keep local knowledge rather than hand everything over.
It suits particularly well where the in-house person is good but stretched, where there is a project backlog that never moves, or where the business has realised how exposed it is to one person leaving.
Who it does not suit
Nobody technical in-house. Then there is nothing to co-manage and you want a full managed service.
An in-house person who does not want it. This matters and it is often decided over their head. If the arrangement is imposed on somebody who reads it as a vote of no confidence, they will not share what they know and it will not work. Bring them into the decision, and be honest that the aim is to take the parts of their job they cannot enjoy.
An expectation of saving money immediately. Co-managed usually costs more in year one than the in-house person alone, because you are adding capability. What it buys is coverage, specialism and the removal of a single point of failure. If the objective is to reduce cost, that is a different conversation, and it is honest to say so.
How to start small
You do not have to restructure. The two highest-value pieces to hand over first, in order:
Monitoring and after-hours cover. Immediately relieves the on-call burden and provides visibility nobody currently has. Low risk, easy to reverse.
Backup verification. Somebody independent restoring something and reporting the result. The in-house person is usually relieved rather than threatened by this, because the responsibility sits on them alone today. How to test a backup.
Both are additive, neither takes anything away, and together they prove whether the working relationship functions before more is committed.
The question worth asking a provider
How do you work with an in-house person, and can I speak to a client who has one?
Then ask that client one thing: does the arrangement make their IT person’s job better or smaller. The answer tells you what you need to know.
What we cover is on Managed IT Services, and the tradeoffs against a fully reactive arrangement are in break-fix or managed. If you have somebody in-house and want to work out what they should stop carrying, that is a good conversation to have with them in the room.