Managed IT proposals are difficult to compare because two providers can use the same label for very different work. One “fully managed” plan may include endpoint monitoring, help desk support, backups, security tools, and vendor coordination. Another may cover monitoring only and bill separately whenever a technician takes action.
The useful question is not “What is the average MSP price?” It is “What workload, risk, and service level are we buying?” This guide provides a planning method that exposes those differences before a contract is signed.
Planning note: Every amount below is an illustrative budgeting example, not a vendor quote or a claim about a universal market price. Replace each assumption with written figures from your shortlisted providers.
Start with the three common pricing models
Most managed service proposals can be translated into one or more of these models.
Per user
A per-user plan charges for each supported employee. It can be easy to forecast when people use several devices, because a laptop, phone, and tablet may all be covered under one user. Confirm whether shared workstations, contractors, seasonal staff, and service accounts count as users.
Per device
A per-device plan assigns a rate to each workstation, server, network appliance, or mobile device. It can suit environments with stable hardware but fewer named users, such as warehouses or clinics with shared terminals. Inventory accuracy matters: an overlooked server or firewall can appear later as an unexpected line item.
Fixed monthly scope
A fixed-fee agreement bundles a defined environment and service catalog into one recurring charge. Predictability is the advantage. Scope discipline is the tradeoff. Ask how the fee changes when you add a location, hire employees, replace a server, or begin a major migration.
Some providers combine the models. For example, user support may be priced per employee while servers and projects are separate. Convert every proposal to an annual total before comparing it.
Build a baseline inventory before requesting quotes
A reliable estimate starts with a one-page operating profile. Record:
- employees and contractors who need support;
- desktops, laptops, mobile devices, servers, and network appliances;
- offices, remote workers, and supported time zones;
- critical applications and cloud services;
- current backup, security, identity, and device-management tools;
- compliance or customer requirements;
- average monthly tickets and recurring problem areas;
- expected hiring, acquisitions, office moves, or migrations.
Do not hide an unstable environment to obtain a lower proposal. A provider will discover outdated systems during onboarding, and the correction will arrive as either a larger project or reduced service quality.
Calculate the first-year cost, not only the monthly fee
Use this planning formula:
First-year managed IT cost = recurring service + onboarding + projects + excluded support + internal oversight
Consider an illustrative 35-person company evaluating a per-user service. Its worksheet might look like this:
| Cost component | Example assumption | Annual planning amount |
|---|---|---|
| Core support | 35 users × $125 × 12 months | $52,500 |
| Onboarding | Fixed discovery and migration project | $7,500 |
| Network refresh | Planned one-time project | $9,000 |
| After-hours allowance | 20 hours × $180 | $3,600 |
| Internal owner | 4 hours/month × $60 loaded cost | $2,880 |
| Illustrative first-year total | $75,480 |
This example is valuable because it makes assumptions visible. If the provider includes after-hours support or the network is already current, those rows change. The model is not a price prediction; it is a comparison tool.
For a multi-year decision, separate one-time costs from recurring costs. A three-year comparison should also account for expected headcount, annual price increases, and major lifecycle replacements.
Identify what “included” actually means
Ask each provider to mark every service as included, limited, or separately billed. The review should cover:
- remote and on-site help desk support;
- endpoint monitoring and patching;
- server and network monitoring;
- identity and access administration;
- endpoint security and email protection;
- backup monitoring and recovery assistance;
- software licensing and renewal management;
- third-party vendor coordination;
- new employee onboarding and departures;
- strategic planning and technology roadmaps;
- security awareness training;
- incident response and after-hours emergencies;
- projects, migrations, cabling, and hardware installation.
Pay attention to verbs. “Backup monitoring” does not necessarily include backup storage, recovery testing, or emergency restoration. “Security management” may mean installing an endpoint agent, not investigating alerts. Ask for a short real-world example of what happens when a user reports a compromised account at 8 p.m.
Compare service levels with measurable evidence
Response time is not resolution time. A provider can acknowledge a ticket in ten minutes and still take two days to restore service. Request the service-level definitions, priority rules, coverage hours, escalation path, and recent performance reporting.
Useful measures include median first response, median resolution by priority, ticket reopen rate, patch compliance, successful backup tests, and customer satisfaction. No single metric proves quality, but a consistent operating report is stronger evidence than a sales promise.
Also determine who will serve your account. Ask whether the help desk is pooled, where after-hours support is delivered, how senior engineers are engaged, and what happens if the assigned account lead leaves.
Watch for five common budget gaps
- Onboarding remediation. Discovery may reveal unsupported operating systems, weak identity controls, or incomplete documentation.
- Projects outside recurring support. Cloud migrations, office moves, and major upgrades are often separate.
- License overlap. A bundle may include tools you already pay for, creating duplicate spend during transition.
- Minimum commitments. A contract may set a billing floor even if headcount falls.
- Exit costs. Data export, documentation transfer, license reassignment, or early termination can create final expenses.
Read the order form, service description, and master agreement together. The marketing page is not the contract.
A practical vendor scorecard
Price should be one part of a weighted decision. A simple scorecard might assign 30% to service coverage, 20% to security operations, 15% to service-level evidence, 15% to team quality, 10% to contract flexibility, and 10% to total cost. Adjust those weights to your risk profile before opening proposals.
Give every vendor the same inventory and questions. Normalize costs into first-year and steady-state annual totals. Then document uncertainties instead of forcing false precision. A proposal with a slightly higher fee but clear inclusions can be less expensive than a low base price surrounded by hourly exceptions.
Final decision checklist
Before signing, confirm that:
- supported users, devices, sites, and hours are listed;
- included security and backup responsibilities are explicit;
- onboarding work and assumptions are priced;
- projects and hourly exceptions have rate cards;
- service levels define priorities and escalation;
- data ownership and offboarding steps are documented;
- renewal, price adjustment, and termination terms are understood;
- references resemble your size and technical environment.
A defensible managed IT budget connects money to a defined operating outcome. Build the inventory, normalize the proposals, and test the contract against realistic incidents. For related planning frameworks, browse all TechCostLab guides or use our SaaS cost planning worksheet to model the software portion of your technology budget.