Colour theme
Book a consultation

Choosing an MSP with a clear service agreement

Compare managed IT providers by scope, response commitments, privileged access, recovery evidence, reporting and exit arrangements.

A hand signing a document with a pen

A managed service provider, or MSP, takes responsibility for agreed parts of an organization’s IT operation. The scope can include support, device management, monitoring and other services, but the label alone doesn’t tell you what’s covered.

Compare providers using the same description of your environment and the same service requirements.

Make the scope explicit

List users, devices, locations, applications and cloud services. Identify what the provider would manage and what would remain with your staff or another supplier.

Ask which tasks are included, charged separately or excluded. Cover onboarding, departures, projects, on-site work and specialist applications. Don’t assume they fall under support.

Record dependencies that could prevent the provider from completing a task.

Define support and escalation

Confirm support hours, contact routes and how urgent incidents are classified. Ask what happens outside ordinary hours and whether that coverage is included or optional.

Separate a response commitment from a restoration target. A quick acknowledgement doesn’t promise that a failed service will be repaired immediately.

Review how the provider coordinates with a carrier, software vendor or other party when the cause is outside its direct control.

Examine privileged access and security work

Ask how administrative accounts, remote access, sign-off for changes and logging are managed. Clarify who responds to alerts and what evidence the business receives.

The Cyber Centre’s guidance for consumers of managed services lists the security questions to settle in these relationships, including who holds administrative access.

Outsourcing a task doesn’t automatically transfer every legal, privacy or business decision. Assign those responsibilities explicitly.

Ask for recovery evidence

Define the systems and data that need backup, retention and restoration. Ask how recovery is tested and whether the normal production credentials can affect backup copies.

Request an explanation of a representative test, its limitations and outstanding work. A backup dashboard or a no-downtime claim won’t tell you whether a restore will work.

Check who can authorize restoration and communicate with the business during an incident.

Check reporting, references and fit

Ask for permission-based references or documented experience relevant to your environment. Verify any certification or partnership claimed.

Review a sample service report and how unresolved issues become planned work. Decide whether staff will receive the training needed to use supported tools.

Compare total cost and contractual commitments alongside the service evidence.

Agree on the exit before starting

Identify who owns accounts, licences, documentation and configurations. Set a process for returning information, revoking access and handing work to a replacement provider.

Keep the agreement specific enough that both parties can tell whether it’s being met.


Featured Image Credit