Managed IT Services Pricing Guide
A support proposal lands in your inbox with one monthly figure and a reassuring phrase like “all inclusive.” Then you notice that a new employee setup, a server alert, and a call from the owner may each be treated differently. The useful question is not whether the number looks low. It is what work, risk, and availability that number actually buys.
Managed IT services can cover some combination of user support, devices, accounts, networks, backups, security administration, and planning. Providers package that work in different ways, so two monthly totals are rarely comparable until the boundaries are clear. This guide gives you a practical way to compare models without relying on invented market averages.
Start with the work, not the package name
“Managed IT” is a service category, not a standardized bundle. One provider may include endpoint monitoring and help desk support, while another focuses on cloud accounts and vendor coordination. Neither description tells you whether after-hours incidents, projects, replacement hardware, or licensing are included.
Make a short inventory of recurring work before requesting proposals. Include the number of people, devices, locations, business applications, and shared systems. Then add events that create irregular work, such as onboarding, departures, office moves, audits, seasonal demand, and recovery exercises. Note who handles each today and how often the work occurs.
This inventory catches an important mismatch. A company with many laptops and few servers has a different support profile from a smaller business with a production system, several locations, or strict recovery needs. User count is a convenient billing unit, but it is not a full measure of effort or exposure.
Ask providers to define each service in ordinary language. “Monitoring” should identify which assets are watched, what signals trigger action, whether a human reviews alerts, and what happens outside business hours. “Security” should specify the controls administered and the decisions that remain yours. “Backup” should distinguish successful copies from tested recovery.
Five ways a proposal can be priced
| Model | Usually easiest to understand | Questions that change the comparison |
|---|---|---|
| Per user | A staff roster provides a simple starting count | Are shared accounts, seasonal workers, contractors, and part time staff counted differently |
| Per device | Useful when endpoints drive most support work | Which device classes count, and are servers, phones, printers, and network equipment included |
| Flat fee | Predictable when scope and volume are stable | What volume assumptions apply, and what event reopens the fee |
| Tiered service | Lets you choose a defined level of coverage | Which controls and response commitments change between tiers |
| Hourly or time and materials | Makes variable work visible on an invoice | How are minimum charges, approval, travel, and emergency work handled |
These are billing structures, not quality ratings. A per-user plan can be carefully scoped or vague. A flat fee can be predictable or based on assumptions that do not match your environment. Hourly support can fit occasional needs, but it does not automatically provide proactive ownership, routine maintenance, or guaranteed availability.
Some providers combine models. A monthly fee may cover a defined core service, with project work billed separately. A device charge may include endpoint support, while server administration sits in another tier. Combination pricing is reasonable when the proposal shows how each component relates to your inventory and workload.
Per user and per device
Per-user pricing follows the people who need help. It can be straightforward when each employee has a standard set of managed services. It becomes less clear when one employee uses a laptop, tablet, phone, and specialized workstation, or when several people share a front desk device. Ask whether the price includes setup, replacement coordination, account changes, and support for personally owned devices.
Per-device pricing follows equipment. It may make sense when staff have very different device counts or when a small number of infrastructure systems carry significant responsibility. Confirm how a device becomes managed and removed from billing. A device that is retired, offline, or waiting for repair should not be treated as an unstated exception.
Neither unit should be assumed to include software subscriptions, hardware purchases, internet service, or vendor support contracts. Those costs may be paid directly by you, passed through by the provider, or included under an explicit allowance. The agreement should say which.
Flat fees, tiers, and hourly work
A flat fee is most useful when both sides agree on a stable scope. Define the covered users, devices, locations, service hours, and expected request volume. Also define how additions are approved. If your headcount doubles or a second office opens, a fee review may be fair. A surprise adjustment based on a loosely worded “reasonable use” clause is harder to manage.
Tiered plans can make tradeoffs visible. Compare the actual service changes rather than labels such as basic, standard, or premium. A higher tier might add response coverage, security administration, backup oversight, or planning meetings. It may instead add a list of tools while leaving labor and incident ownership unchanged. Request a side by side scope matrix.
Hourly work is not inherently a bad choice. It can suit a small organization with a stable environment and infrequent support requests. It is a poor substitute for a defined response commitment if an outage would stop work. Ask how urgent incidents are prioritized and whether an available technician is guaranteed or merely requested.
Turn a vague quote into a comparable scope
Build a one page service worksheet with a row for each responsibility. Mark it included, excluded, limited, or unclear. Avoid accepting “best effort” as a complete answer. Ask what the provider will do, what you must do, and what third party must act before the issue can be resolved.
| Responsibility | Detail to document |
|---|---|
| Service desk | Covered people, supported channels, operating hours, and request tracking |
| Endpoints | Supported operating systems, monitoring, patching, and replacement coordination |
| Identity | Account creation, multifactor authentication, access review, and departure handling |
| Network | Managed locations and equipment, alert review, and carrier escalation |
| Backup and recovery | Protected systems, retention, recovery objectives, and test frequency |
| Security response | Alert triage, containment authority, evidence handling, and escalation contacts |
| Projects | Definition of project work, approval steps, and estimate format |
| Vendors | Who opens cases, follows them, and pays for the external support |
The table does not need to be complicated. Its purpose is to make two proposals comparable and surface work that would otherwise fall between them. If nobody owns a row, mark it as a decision you need to make before signing.
Find the cost outside the monthly fee
The recurring service is one part of your total operating cost. Other charges can include licenses, endpoint protection, backup storage, replacement equipment, project labor, after-hours response, onboarding, travel, and third-party vendor support. The provider may buy some of these for you, but bundling does not make them disappear. Ask for the billable party, renewal timing, and cancellation process.
Pay particular attention to implementation. A new provider may need to inventory devices, document systems, transfer administrative access, configure monitoring, and correct old issues. Those tasks can be valuable, but they are not necessarily part of the monthly service. Ask whether the transition is fixed scope, hourly, or staged, and what deliverable shows that it is complete.
Also price the work your own staff retain. Someone must approve access, identify business priorities, coordinate vendors, and tell the provider when a change is authorized. If the agreement assumes your availability for every decision, that internal effort belongs in your comparison. A service can reduce technical labor without removing business ownership.
Check the exclusions line by line
Exclusions are not a flaw when they are explicit. They become a budgeting problem when broad words leave you guessing. “Projects excluded” could mean a new office network, but it could also be used to bill a routine group policy change. Ask for examples of requests that fall on each side of the boundary.
Clarify incident work as well. If a cloud vendor has an outage, does the provider investigate, communicate status, and open a case, or does it only confirm that the service is unavailable? If a suspicious account is found, can the provider suspend it under preauthorized rules, or must an owner respond first? These distinctions matter more than a generic promise of support.
Calculate your own comparison
Use your observed workload rather than a generic market benchmark. For each category, estimate the hours your staff spend in a typical month, then add irregular events and work that is currently postponed. Record both paid labor and operational consequences. Avoid counting every saved minute as cash savings if the person remains on payroll and gains no useful capacity.
Here is an illustrative scenario, not a market price. A 14-person firm records 18 hours of routine account and device work per month, 7 hours coordinating external vendors, and 5 hours handling unplanned incidents. The owner estimates 6 hours of internal approval and follow-up. A proposal covers the first two categories and incident triage but excludes projects, licensing, hardware, and recovery testing.
The firm should not compare the provider fee with 30 hours as if all work vanished. First identify what the plan actually absorbs. If it takes responsibility for the 18 routine hours and 7 vendor hours, the remaining comparison includes internal approvals, uncovered projects, licensing, and the risk of weak recovery testing. Next ask what the provider expects the client to do for triage and how the response process works. The worksheet can then compare the same responsibilities across providers.
For a simple internal estimate, calculate annual service charges plus expected add-ons and retained labor. Then model a high-work month, such as employee turnover or a location change. Keep assumptions beside each calculation. The estimate is useful because it exposes uncertainty, not because it predicts every invoice exactly.
Response promises need a clock and a definition
Service-level language often distinguishes response from resolution. Response may mean an acknowledgement, while resolution means the business service works again. Ask which clock applies, when it starts, what pauses it, and how the provider reports a missed commitment. A response target without coverage hours may not mean what you expect at night or on a holiday.
Define severity with business examples. A single user unable to print is different from a company unable to sign in or receive orders. Decide who can declare a major incident and which contact is called if the primary contact is unavailable. Puerto Rico businesses should also ask how the provider handles power and connectivity disruptions during hurricane season, including remote access, alternate contacts, and recovery coordination.
Questions to ask before you agree
- Which people, devices, sites, and services are in the scope today?
- What work is monitored automatically, and who reviews the alert?
- What is the distinction between acknowledgement, diagnosis, and resolution?
- Which licenses and vendor charges are separate, and who controls renewal?
- How are new users, devices, and locations added to the fee?
- What triggers project billing, and do I approve an estimate before work begins?
- How are administrator accounts, documentation, and configuration data returned at exit?
- What recovery test can I review, and what systems were included?
Listen for concrete examples. A capable provider should be able to explain a normal request, a security concern, and a difficult vendor escalation without promising that every issue will be effortless. Ask for a sample ticket report or a redacted runbook if one is available, not for confidential information from another customer.
A Monday morning way to get started
Choose the last four weeks of tickets, invoices, and staff requests. Group the work into user support, endpoint care, identity, network, backup, security, vendors, and projects. Count recurring requests and note the unresolved ones. Then assign an internal owner to each category, even if the answer is that nobody currently owns it.
Send the same inventory and worksheet to each provider. Ask them to fill in their included, excluded, and limited responsibilities and to state assumptions about response hours and volume. Compare the completed scope before comparing the monthly totals. If a provider will not explain an exclusion, treat the uncertainty as a cost you may later have to absorb.
The right arrangement is the one your team can operate and govern, not necessarily the one with the broadest package. If you want help mapping those responsibilities across support, systems, and ongoing administration, Ozcorp InfraOps can be part of that conversation. Start with the work you need covered and keep the agreement specific.
Discuss your infrastructure requirements
Review the current service offer or schedule a conversation about the scope you need.
Stay in touch
Follow Ozcorp
Leave your details for occasional notes about new articles and Ozcorp updates.