The quickest way to tell them apart: an AMC is triggered by something breaking. A managed services contract is meant to stop you noticing that it did.
That distinction sounds like marketing, but it changes what you are paying for, how the vendor is incentivised, and what happens at 6pm on a Friday.
What an AMC actually covers
An Annual Maintenance Contract is a repair agreement. You pay a fixed annual fee, and in return faults get fixed within an agreed response window. Two common shapes:
- Comprehensive — labour and spare parts are both included, so a failed power supply or drive is covered
- Non-comprehensive — labour is included, parts are billed separately at the time
Comprehensive costs more up front and is easier to budget. Non-comprehensive looks cheaper on the quote and is the one that produces unexpected invoices, particularly on ageing hardware where failures cluster.
An AMC is a good fit when your estate is stable, your tolerance for short outages is reasonable, and you have someone internally who notices problems and raises tickets.
An AMC assumes somebody on your side is watching. If nobody is, you are buying a fast response to problems you will find out about late.
What managed services adds
A managed contract moves the watching to the vendor. In practice that means some combination of:
- Monitoring of links, switches, servers, backup jobs and endpoints, with alert thresholds agreed up front
- A helpdesk your staff can contact directly, rather than routing through your internal IT
- Preventive maintenance on a schedule — firmware, patching, capacity and log review
- Periodic reporting and a review meeting where the trend, not just the ticket count, gets discussed
The economics are different too. Under an AMC the vendor is paid the same whether you have two incidents or twenty. Under a managed contract, incidents cost the vendor money, so preventing them is in their interest as well as yours.
A straightforward way to choose
Ask four questions about your own organisation:
- Who notices when something breaks? If the answer is "a user complains", monitoring is worth more to you than a faster repair.
- How many sites? Beyond three or four locations, coordinating repairs yourself becomes a job in itself.
- What does an hour of downtime cost? Not a precise figure — an order of magnitude is enough to tell you which side of the line you sit on.
- Do you have internal IT capacity? A capable internal team plus an AMC is often the better-value combination. No internal team plus an AMC is the gap where things go wrong.
You can mix them
The most common arrangement we deploy is not one or the other. It is a managed contract covering the network, servers and backup — the things where early warning matters — with a straightforward AMC underneath for desktops, printers and peripherals, where a next-business-day repair is genuinely adequate.
Splitting it this way usually costs less than putting everything on a managed contract, and considerably less than discovering your backup has been failing quietly for five weeks.
What to ask any vendor
- What exactly is monitored, and what alert thresholds trigger action?
- What are the response and resolution windows, by severity — and what happens when they are missed?
- Is support remote, on-site, or resident, and who decides which applies?
- What is explicitly excluded? This is the answer that tells you the most.
A vendor who can answer those four crisply is quoting a real service. One who cannot is quoting a price.
