Break/Fix vs Managed Support: Which Model Fits?

Break/fix IT support and managed support get framed as rivals, old model versus new. For multi-site estates the framing misleads: most branch networks end up needing a deliberate blend, and the real question is which layer of the estate belongs under which model.

The two models, honestly stated

Break/fix: you pay per incident. Something fails, a call is logged, an engineer responds under an agreed SLA, you're billed for the work. Costs track failures; quiet months cost little.

Managed: you pay a recurring fee per site or device, and the provider owns outcomes: monitoring, maintenance, response and usually reporting. Costs are flat; the provider carries failure risk.

The incentive mechanics

Break/fix providers earn from failures, which sounds perverse but self-corrects at estate scale: a provider whose reporting shows fault trends is handing you refresh intelligence, and providers who fix badly lose the contract. The genuine weakness is that nobody is paid to prevent anything.

Managed providers earn from silence, so prevention is profitable: firmware currency, proactive swaps, monitoring. The weakness runs the other way: flat fees invite scope disputes at the edges ("that's a project, not support"), and lazy managed contracts can hide thin service behind a steady invoice.

What the estate's shape suggests

Hardware-heavy, distributed, trading estates, POS fleets, branch equipment, store networks, suit break/fix with spares management: failures are physical, prevention is limited (hardware ages regardless of monitoring), and per-incident economics stay honest when volumes are visible. This is the classic branch network model, and pairing it with pooled spares turns SLAs from promises into logistics.

Software-and-infrastructure layers, servers, network cores, security, suit managed models where monitoring genuinely prevents incidents and the provider's remote tooling earns its fee.

The churn layer, installs, moves, small changes, is neither; it's IMAC, best run on a rate card.

The blend in practice

A typical 200-site retail estate lands here: managed monitoring on the network core, break/fix with regional spares across store hardware, IMAC on rate card, and one provider carrying all three so a single dataset describes the estate. The single-provider point matters more than the model split: fault history, asset registers and refresh planning only compound when the layers share a system. Our support services are built on that basis.

The question to start from

Don't ask which model is modern. Ask: for each layer of our estate, who should carry the failure risk, and what data do we need out of the arrangement? Price both models against real fault history, and the answer usually writes itself.

Want your support model matched to your estate? Speak to an expert.