Managed Services vs Co-Managed IT vs Building In-House | blueAPACHE
A structured comparison of the three ways to resource IT operations — full managed services, co-managed with a provider, or building an internal team — including the conditions that should decide it and the true cost of around-the-clock internal coverage.
There are three ways to resource IT operations, and most organisations arrive at a decision by drift rather than by comparison. This page sets out the conditions that should actually decide it — including the cases where the answer is not to engage a provider at all.
The three options in one view
| Full managed services | Co-managed | In-house | |
|---|---|---|---|
| Who owns the outcome | Provider | Shared, defined per domain | You |
| Who sets direction | You, provider executes | You | You |
| Coverage outside hours | Provider's roster | Provider's roster | Your roster |
| Specialist depth | Provider's bench | Provider's bench | Hire or contract |
| Cost shape | Opex subscription | Opex subscription + headcount | Headcount + capital + tooling |
| Institutional knowledge | Documented with provider | Retained internally | Retained internally |
| Typical commitment | Multi-year term | Multi-year term | Permanent |
| Fails when | You need deep control | Accountability is vague | Scale doesn't justify the roster |
The number most business cases get wrong
The in-house option is almost always underestimated, and the error is nearly always the same: costing coverage as headcount rather than as a roster.
Genuine around-the-clock cover is not one hire. To answer reliably outside business hours you need enough people that the rota survives annual leave, sick leave, training and resignation — and you need more than one person competent in each critical discipline, or you have simply relocated the single point of failure from a system to a human.
Then add the parts that do not appear on a salary line: monitoring and service-management tooling, patch management, asset discovery, the licences for all of it, recruitment cost, and the productivity gap while a new hire learns your estate.
That is why the crossover point for building internally sits higher than most organisations estimate. It is real — above roughly 1,000 seats the fixed cost amortises across enough users to make internal viable, and plenty of organisations that size run excellent internal functions. But the calculation has to include the roster, not the role.
Choose full managed services when
- You want operational accountability to transfer, with one party answerable for outcomes
- Internal IT attention is better spent on business systems and change than on operations
- You need coverage and specialist depth you cannot justify hiring
- Predictable operating cost matters more to the CFO than asset ownership
- You need compliance evidence — certification scope, audit rights, defined notification timing — that a small internal function struggles to produce
Choose co-managed when
- You have at least one capable internal person with real ownership of IT direction
- The gap is capacity and specialist depth, not competence
- Business-specific systems would take an outsider a long time to learn properly
- You want institutional knowledge to stay in-house
- You are not ready to transfer control, and that reluctance is considered rather than reflexive
The critical prerequisite: you must be willing to write the accountability boundary down in operational detail. Co-managed with a vague boundary is the worst of the three options.
Choose in-house when
- Scale genuinely justifies the roster — and you have costed the roster, not the role
- Your estate is unusual enough that external standardisation would fight you
- IT capability is itself a competitive differentiator in your market
- Regulatory or sovereignty constraints make outsourcing structurally difficult
- You can recruit and retain in your location and salary band, which in Australian regional markets is not a given
The hybrid nobody names
Most real organisations end up in a fourth position: internal team for business systems and change, provider for infrastructure operations and security monitoring. That is a legitimate and common answer, and it is worth designing deliberately rather than arriving at by accident.
The design question is which domains transfer wholesale — those work well — versus which get split down the middle, which tend not to.
How this maps to blueAPACHE's engagement modes
Three modes over the same underlying delivery capability. What changes is how much ownership transfers, not which capability is available.
- OUTCOME — Managed Services. Full ownership of the IT environment, integrated MSP and MSSP delivery. The lead offer.
- CONTROL — emPOWER Operational Capability. Co-managed; you keep direction, blueAPACHE supplies capability and coverage.
- TECHNOLOGY — Technology Services. Projects, transformation, hardware-as-a-service and procurement on consumption, no ongoing term.
Organisations commonly move between them. Starting with TECHNOLOGY to see how a provider actually operates before committing to a longer term is a normal and sensible path.
→ Three Ways to Engage — OUTCOME, CONTROL and TECHNOLOGY → Contract Structure and the Commercial Model
The test that resolves most evaluations
Ask what you want your internal IT people doing in two years.
If the answer is running infrastructure and answering tickets, build in-house and fund it properly. If the answer is working on the business — systems, data, change, the things only insiders can do — then operations should transfer, and the question is simply how much.
If you cannot answer at all, that is worth resolving before running a procurement process. A provider selection made without that answer tends to produce an arrangement nobody is happy with, whichever provider wins.