Routine tickets eat the week of a small internal IT team, and strategic work slides. A co-managed arrangement hands defined parts of that load, usually the help desk, monitoring, and security, to an outside provider, while your staff keep strategy and manage the business systems only they understand.
Most internal teams that look at this model have the skills and lack the hours. One or two people own the help desk, the patch cycle, vendor renewals, and the cyber insurance questionnaire, and when an alert fires on a holiday weekend, you answer it anyway. Each week spent on password resets is a week the migration project sits still, and the roadmap slides another quarter.
Co-managed IT sits between hiring more staff and handing everything to an outside firm. You buy coverage and depth for the routine work that consumes your team’s hours, and you keep control of the work that needs your judgment.
What are co-managed IT services?
Co-managed IT is a contract in which an MSP owns a written slice of your IT operation, your internal team owns the rest, and both work in the same tools. Your headcount stays and the provider builds capacity around it, so the decision you face is where to draw the line between the two slices.
A restaurant kitchen is a fair picture. The owner brings in an extra line crew for the dinner rush and overnight prep. The chef still writes the menu, knows the regulars, and decides what leaves the pass. The crew handles volume the kitchen could never staff on its own.
The typical split looks like this. The provider takes Level 1 (L1) help desk tickets, after-hours coverage, monitoring, patch deployment, and security operations. Internal IT keeps strategy and budget, line-of-business applications such as the ERP or EHR, vendor relationships, and the company context that never makes it into documentation.
Two neighboring models are easy to confuse with this one. Under staff augmentation, you supervise contractors yourself. Under fully managed IT, the provider owns everything, strategy included. In a co-managed setup, the provider manages its own people against an SLA, and you keep the decision rights.
How does co-managed IT work day to day?
Both teams work out of one system of record: a shared PSA or ITSM tool for tickets and a shared RMM for endpoints. Your technicians get scoped accounts in the provider’s tools, or the provider’s technicians get accounts in yours. Every ticket, note, and time entry ends up in one place.
Define the paths for intake, escalation, and after-hours decisions:
- Users ask for help the way they always have, through email, the portal, Teams, Slack, or phone. Routing rules then send each ticket category to the team that owns it. Password resets and lockouts land in the provider’s queue. ERP errors go to your internal board.
- Escalation follows written paths. The provider’s L1 technician escalates to its own L2 team or to yours, depending on the responsibility matrix. The handoff carries what was tried, who was contacted, and what the user reported.
- After hours, critical alerts go to the provider’s on-call engineer, who pages your named contact for any decision that needs business authority, such as taking a production system offline.
A shared PSA where the two teams never read each other’s notes is an expensive group chat. Before you sign, ask to see a completed ticket from another co-managed client, notes and all.
Who owns what? A sample RACI split
Every function needs exactly one accountable owner. In the matrix below, R is responsible (does the work), A is accountable (owns the outcome), C is consulted, and I is informed. A starting matrix might look like this:
Published matrices disagree most on L1 and incident response. Some providers leave L1 with the internal team and take L2, L3, and after-hours work. Others take L1 and send specialist escalations back in. Either version works once it is written down and signed by both teams. Review the matrix quarterly, and name the person who can authorize a disruptive security action before the first alert fires.
Co-managed vs fully managed vs in-house IT
The three models differ in who decides, who does the work, and who answers the phone at night:
In-house works when you can staff nights and specialists without burning out your generalists. Fully managed suits a company with nobody on staff to hold the context. Co-managed fits everyone in between, which describes most companies with a one- to five-person IT department.
What services do co-managed IT providers cover?
Most providers offer help desk support (L1 and often L2), 24/7 RMM monitoring, patch management, backup operations, Microsoft 365 and cloud administration, after-hours on-call, and extra hands for projects. A provider can staff that L1 tier with technicians or run it as managed automation, as Fixify for MSPs does, working white-label inside the PSA with human analysts supervising; ask which approach your provider uses before you sign.
Security and compliance drive a large share of these contracts. Providers run endpoint detection, SIEM monitoring, vulnerability scans, and audit evidence collection. The provider can operate the controls, but the obligation stays with you.
Under HIPAA, a provider that handles ePHI signs a business associate agreement. Under PCI DSS, you must maintain information on which requirements each service provider manages, which you manage, and which are shared. CMMC assessments look for the same documented split of control ownership. A RACI that already maps those controls saves you a painful week before every audit.
What are the benefits and drawbacks of co-managed IT?
The benefits come from adding a bench behind your team:
- 24/7 coverage: The provider’s on-call rotation handles alerts and urgent tickets outside business hours, so your staff stop sleeping next to their phones.
- Specialist depth: You get security, cloud, and networking engineers for less than the cost of hiring each one.
- Scalability: Ticket volume from an acquisition or a new office goes to the provider’s queue instead of your overtime.
- Less burnout: Routine work stops landing on your senior people, and they get time back for projects they were hired to run.
Four drawbacks are manageable if you plan for them:
- Lost control: The provider changes things in your environment. Mitigate with change approval rules in the contract and read access to every ticket.
- Finger-pointing: Unclear ownership turns every outage into a debate. Mitigate with the single-owner RACI and a named tiebreaker for disputes.
- Data access: The provider holds privileged credentials to your systems. CISA’s MSP security advisory says to scope provider accounts to the systems they manage, keep them out of your admin groups, enforce MFA, and disable them when unused.
- Coordination overhead: Two teams making changes means more communication. Budget for a weekly sync during the first months.
How much do co-managed IT services cost?
Providers price co-managed work per user, per device, or as a flat retainer, and you keep paying your own staff on top. Per-user is the most common model, and it runs below a fully managed per-user fee because you keep part of the work in-house. Per-device suits environments with more endpoints than people, and a retainer buys a defined block of hours and functions. There is no reliable public benchmark for retainers, so judge a retainer quote against the hours and functions it covers. Ask how projects, onboarding, and after-hours work outside the provider’s scope are billed.
Compare that with one more hire. The BLS median wage for computer user support specialists was $61,860 in May 2025, before benefits, payroll taxes, and recruiting. That hire adds about 40 hours a week but no night coverage and no backup during vacations. A provider slice priced per user, by contrast, buys 24/7 coverage across your whole user base for a predictable monthly fee.
What are the signs your organization needs co-managed IT?
You are probably a candidate if several of these sound familiar:
- After-hours alerts go to whoever happens to check their phone
- One person holds all the admin credentials and all the tribal knowledge
- The project backlog has not shrunk in two quarters
- Your staff skip vacations because nobody can cover the queue
- A compliance audit or cyber insurance renewal asks for controls you cannot staff
- First response times slip on the clients who track them
How do you choose and onboard a provider?
Choose on fit with your tools, your matrix, and how the provider delivers L1 before price. Set written service targets, onboarding phases, and exit terms. Start by checking four things:
- Tooling: Confirm the provider can work inside your PSA or give your team scoped access to theirs.
- Documentation quality: Ask for a sample ticket with escalation notes from a current co-managed client.
- Security posture: Ask for its SOC 2 report and how it enforces least privilege on your systems.
- References: Talk to a client with an internal team about your size.
Negotiate response and resolution targets for each severity level. Use these illustrative targets as a starting proposal, not as measured industry benchmarks:
Check whether each clock runs 24/7 or only during business hours. A four-hour target means little if it pauses overnight. For critical incidents after hours, the contract should name the provider’s on-call engineer, your contact, and the call order if that contact does not answer.
The phases run from discovery and asset inventory, to agent deployment and access setup, to knowledge transfer through joint ticket work, to a stabilization period. Write the exit terms now: a notice period, a defined transition-assistance period with rates, your ownership of all data and documentation, and the return of credentials.
Where co-managed IT lands
Co-managed IT works when the split between your team and the provider is written down, signed, and reviewed, not left to whoever picks up the phone at night. Drawing that line is the whole decision, and it starts with knowing which tickets drive your L1 volume. The free IT Help Desk Ticket Snapshot takes 90 days of CSV ticket data and returns a customized benchmarking report in 48 hours, so you can see where your L1 load sits before you scope a provider.





