There’s a conversation happening in UK IT departments that rarely makes it into board papers. It’s about whether the managed services arrangement the organisation signed up to three or four years ago is actually delivering what was promised, or whether it’s become a kind of expensive insurance policy that everyone is too tired to challenge.
The honest answer, in a lot of cases, is the second one. Managed services contracts have a tendency to drift. They get signed at a point when the organisation has a clear need, deliver reasonable value for the first eighteen months or so, and then slowly settle into a rhythm where the monthly invoice goes out, a ticket queue gets processed, and nobody has the time or political capital to ask whether the scope still matches the reality.
The shape of a drifting contract
The classic pattern looks something like this. A contract was signed covering a defined set of infrastructure, say a set of on-premises servers, a Microsoft 365 tenant, and some network equipment. Over the following years, the estate changes. Workloads move to Azure. New SaaS products get bought. A cyber security tool gets added. The original contract doesn’t cover most of these things, but nobody renegotiates, because renegotiating is a project nobody has time for.
The MSP, meanwhile, continues delivering against the original scope. They respond to tickets for the things they’re contracted to respond to. The things outside the contract either fall on an internal team that’s already stretched, or get patched by side arrangements that nobody is really tracking. The boundary between what’s managed and what isn’t becomes fuzzy. Accountability for outages becomes a finger-pointing exercise.
This isn’t usually anyone’s fault in particular. It’s a structural problem with how managed services contracts tend to be written and then not maintained.
What changed since the last contract was signed
A few things have shifted materially in the last three years that make a lot of existing contracts look outdated.
Security is now a much bigger part of the picture than it used to be. An MSP that doesn’t integrate meaningfully with the organisation’s security operations, whether that’s through a proper 24/7 SOC, tight alignment with the SIEM, or active management of identity and access, is increasingly a liability rather than a help. The organisations being hit hardest by ransomware tend to be the ones where the MSP was looking after “IT” and the security team was looking after “security” and the handover between them was informal at best.
Cloud has changed the operating model. Managing an Azure estate well is a different discipline from managing on-premises infrastructure, and it requires different skills, different tooling, and different commercial arrangements. MSPs that haven’t built genuine Azure capability are still perfectly good at keeping servers running, but that’s an increasingly small share of what most UK organisations actually need.
And the expectations of end users have shifted. A decade ago, a ticket SLA of “next business day” was acceptable for most internal issues. In hybrid working, with staff distributed across kitchen tables and offices and trains, the tolerance for friction has dropped considerably. Managed services that feel like they’re being delivered by people who know the organisation, rather than by a call centre working from a script, are the ones people actually recommend.
The consolidation argument
One of the quieter shifts in UK managed services is the move towards consolidation. Organisations that used to have separate providers for infrastructure, applications, cyber, and Microsoft 365 are finding that the coordination tax between them is high, and that incidents spanning multiple domains turn into blame games. The counter-argument, that best-of-breed specialists deliver better quality in each area, has merit but tends to lose out in practice to the simpler benefits of a single contract, a single escalation path, and a single operational view.
One UK operator’s approach to Microsoft managed services covers applications, Azure, cyber, Dynamics, Microsoft 365, and SharePoint under a single arrangement, and is a reasonable reference for what that consolidated model looks like when it’s done properly. The value isn’t the bundling itself. It’s that the boundary problems disappear.
What good looks like on renewal
For IT leaders facing a managed services renewal in the next twelve months, the useful question isn’t “are we happy with the current provider.” It’s “does the current contract still describe the estate we actually have?” In most cases, the answer is no, and that’s the starting point for the renegotiation worth having.
That conversation is harder than just rolling the contract forward on its existing terms. It’s also how organisations avoid spending the next three years continuing to pay for an arrangement that quietly stopped being fit for purpose somewhere around 2023.
