ITSM

MSP revenue model -
how to scale an IT company?

Three billing models, profitability structure, scaling from a dozen to more than a hundred clients, staffing per technician and an illustrative MSP growth path.

← Back to Blog
ITSM
Mateusz Roszkiewicz · May 2026 · 13 min read

The question of the MSP business model is one of the most common I get from IT companies wanting to switch from a project model to recurring revenue. The problem is that MSP growth requires not only good technicians, but first and foremost an understanding of three fundamental billing models that strongly influence both profitability and scale. In this article I break down how the three main models work, how gross and net margin shape up, how many people you need to serve more clients, and what the growth path of an MSP can look like.

3 models
billing: per-device, per-user, all-in
MRR
recurring revenue - the foundation of the MSP model
All-in
the model with the highest profitability and scalability

What an MSP is and why the business model matters

An MSP (Managed Service Provider) is a company that delivers IT services to other companies on a subscription model. The MSP handles infrastructure monitoring, incident management, patching, backup, security and help desk support for its clients. MSP revenue comes from a subscription cycle (recurring revenue) - every client pays a fixed monthly or annual fee, regardless of incident count.

This differs dramatically from the traditional IT-services model, where revenue comes from billable hours and projects. In the service model your employee swaps a disk and you earn once. In the MSP model they swap the disk, and you earn for all 12 months on the fixed fee.

Why the business model matters: the three MSP billing models (per-device, per-user, all-in) differ drastically in margin, operational complexity and scalability. Picking the wrong one at the start can hold scaling back by as much as half. Top-line revenue may be similar, profitability drastically different.

Bad news: most IT companies in Poland moving to MSP pick per-device because it is easy to sell - the client gets "5 USD per computer". Good news: you can scale up to the all-in model, and every serious MSP does exactly that in year 2-3 of operation.

3 main billing models - per-device, per-user, all-in

1. Per-Device

  • Fixed fee per device per month
  • Example: 5 USD per computer, 8 USD per server
  • Pro: easy to sell to clients, simple billing
  • Con: low margin, harder to scale technicians, per-hardware cap on revenue
  • 1 technician serves: 5-8 clients
  • Best for: MSP startups, small IT companies, early acquisition

2. Per-User

  • Fixed fee per user per month
  • Example: 10 USD per user regardless of hardware count
  • Pro: higher margin, more flexible than per-device
  • Con: requires a clear definition of "user", harder with remote workers, BYOD can reduce revenue
  • 1 technician serves: 12-15 clients
  • Best for: mid-stage MSPs (2-5 years), office-based teams, stable structures

3. All-In (Tiered/Value-Based)

  • Fixed annual fee for the full service package
  • Example: an equivalent of about 30,000 PLN per year for a 50-person company (everything included)
  • Pro: highest margin, easier planning, growing per-client ROI
  • Con: requires the client to be financially ready, longer sales cycle, risk of underscoping
  • 1 technician serves: 15-20 clients (real economics)
  • Best for: mature MSPs, corporate clients, enterprise segment

Hybrid (recommended)

  • A combination of all three above
  • Example: base fee + per-user + per-device overages
  • Pro: flexibility, captures the easy entry of per-device while maximizing revenue
  • Con: more complex billing, requires CRM for tracking
  • 1 technician serves: 12-18 clients (average)
  • Best for: mature MSPs serving different client segments

In practice: most MSPs in Poland start with per-device (easy to sell, easy for the client to understand), scale to per-user in year 1-2, and gradually move to all-in for top clients in years 2-3. The pace of that transition determines how fast profitability grows.

Gross and net margin in an MSP

When talking about MSP profitability, gross margin (revenue minus direct costs) must be separated from net margin (after all operating costs). The table below shows typical trends - actual values depend on pricing, labor costs, client segment and operational efficiency, so each MSP should compute them on its own data.

Model Gross margin (trend) Net margin (trend) Scalability
Per-Device Lower Lowest Hard, requires scale
Per-User Medium Medium Better than per-device
All-In Highest Highest High, scalable
Hybrid (mix) High High Recommended

Important: net margin depends on cost structure - the largest item is usually technician salaries, followed by software, infrastructure hosting, insurance, office and admin costs and marketing. As you grow some costs rise, but economies of scale (e.g. one administrator serving more clients) usually improve net margin over time.

Key observation: the all-in model tends to have a clearly higher net margin than per-device at a comparable client count - one of the main reasons mature MSPs gradually move toward this model.

Staffing - how many technicians per client?

One of the toughest questions for a growing MSP: when do you hire the next technician? The answer depends on model and client complexity.

  • Per-device model

    1 technician serves 5-8 clients (small, up to 50 people each). Most of the time goes to tickets and help desk, less to strategic planning. Add a technician when one is busy more than 85% of the time.

  • Per-user model

    1 technician serves 10-15 clients. Higher SLAs require proactive planning and documentation. Add a technician when churn rises above 8% per year.

  • All-in model

    1 technician serves 15-20 clients (mid-sized, 100-300 people). Requires senior skills, workflow automation (ManageEngine) and proactive monitoring. Add a technician every 15-20 new clients.

  • Operational overhead

    For every 3-4 technicians add 1 systems administrator (backup, compliance, infrastructure) and 1 person dedicated to onboarding and SLA. For 50 clients (7-8 technical staff) = 2 full-time Admin/Sales/Ops. Total: about 10 people.

  • Growing from 10 to 100 clients typically requires around 10-15 full-time people (7-10 technicians + 3-5 operations). The required pace is roughly 1 new technician every 2-3 months, which calls for payroll and training planning.

    Scaling from 10 to 100 clients - timeline and investment

    MSP growth is not linear. Below is a realistic timeline for a Polish MSP:

    Phase 1: Startup (a dozen or so clients)

    Phase 2: Scaling (tens of clients)

    Phase 3: Growth (tens to more than a hundred clients)

    The most common mistake: companies wait until year 2-3 to invest in tooling and process. That is wrong - it should start in month 3-4. The earlier you have ManageEngine SDP with automation, the faster you scale without adding headcount.

    An illustrative MSP growth path

    The scenario below is an illustrative example - not a description of a specific client - showing a typical path that an IT company moving to the MSP model can take. The actual numbers depend on market, client segment and operational discipline.

    Starting phase: small team, a dozen clients, per-device model. Ad-hoc tooling (simple ticketing, spreadsheets for SLAs, manual monitoring). Typical challenges: high turnover, no processes, overloaded technicians.

    Decision: roll out a dedicated ITSM system (e.g. ManageEngine ServiceDesk Plus) and train the team on ITIL. Gradual move to per-user.

    Scaling phase: larger team, tens of clients, hybrid model (base fee + per-user). Automation of typical tasks (password resets, onboarding, updates) frees up technicians and reduces repetitive tickets.

    Mature phase: stable team, key clients moved to all-in, integrations with client CMDBs and automatic asset discovery. Low churn and stable recurring revenue.

    Key success factors: (1) deploy tooling early, not at maturity. (2) gradual move from per-device to per-user and then all-in. (3) standardize processes (ITIL, workflow automation) before scaling the team. (4) hire experienced architects, not only junior support.

    Tools - ManageEngine SDP MSP Edition, Endpoint Central

    For an MSP that wants to scale fast, tooling is not optional - it is foundational. ManageEngine offers a dedicated edition for MSPs.

    ManageEngine ServiceDesk Plus MSP Edition is a cloud variant with multi-tenant architecture - each of your MSP's clients gets its own portal, but you manage everything from a single console. Typical features: per-tenant customization, separate billing per client (if you want to resell), white-label portal, consolidated reporting across all clients.

    ManageEngine Endpoint Central - update management (patch management), software deployment, inventory across all client devices. Update automation (operating systems, applications) is one of the biggest time-savers for an MSP - one rule and 1000 computers are patched automatically at 2 AM.

    ManageEngine OpManager - infrastructure monitoring (CPU, memory, network). Real-time alerting before the client notices a problem. Standard for an MSP with 50+ clients.

    ManageEngine also offers bundles combining several products (e.g. SDP MSP + Endpoint Central + OpManager). Buying a bundle is often more cost-effective than buying separately - confirm the exact price difference in a current quote with a partner.

    Summary and business checklist

    The MSP revenue model is a cardinal choice - bad decisions early on can slow growth by as much as half. Below is a checklist to work through before, or during, year 1 of your MSP.

    FAQ - most common questions about the MSP model

    What is an MSP (Managed Service Provider)?

    An MSP (Managed Service Provider) is a company that delivers IT services to other companies on a subscription model. It handles infrastructure monitoring, incident management, patching, backup, security and help desk support. It differs from a traditional IT service company in that revenue is recurring and spread across many clients rather than tied to a single project.

    What are the 3 main MSP billing models?

    The three main models are: (1) Per-device - a fixed fee per device (e.g. 5 USD/month per computer), easy to sell but low margin. (2) Per-user - a fee per end user (10-15 USD/month), more flexible with higher margin. (3) All-in (tiered) - a fixed annual fee for a service package (e.g. an equivalent of about 30,000 PLN/year), the highest margin but it requires the client to be financially ready.

    What is a typical MSP margin?

    MSP margin depends on billing model, cost structure (mostly technician salaries), pricing and operational efficiency. The general trend is that the all-in model allows a higher margin than per-device, mainly thanks to economies of scale - one technician can serve more clients. Each MSP should calculate the specific percentages on its own financial data.

    How many technicians do I need to serve 50 MSP clients?

    Typically 1 technician serves 15-20 clients in the all-in model, 10-15 in per-user, 5-8 in per-device (depending on size and complexity). For 50 mid-sized clients you need 3-4 senior technicians, plus 1 systems administrator, plus 1 person dedicated to onboarding and SLA. Total operations overhead: 2 people. Total roughly 7-8 people for 50 clients.

    How fast can you scale an MSP from 10 to 100 clients?

    Scaling from 10 to 100 clients typically takes 3-5 years organically (adding 3-5 clients per month). Accelerating to 1-2 years requires investment in marketing (LinkedIn, partner network), automation tooling (ManageEngine SDP MSP Edition, Endpoint Central) and hiring. Critical for growth is not losing existing clients - churn target < 5% per year.

    MR
    Mateusz Roszkiewicz
    Head of Sales · Rotech Group · MSP scaling and ManageEngine licensing expert
    Consultation for your MSP

    Is your business model optimal for scaling?

    Rotech Group will audit your revenue model, staffing and tooling - free. We will show you what to improve so you can scale faster without adding headcount.

    Book a consultation →