Ask three managed IT services providers to quote the same forty-person business and you will get three numbers that do not resemble each other. Some of that gap is scope. Most of it is structure — the way a managed IT support agreement is built long before anyone types a monthly figure into a proposal.

Managed IT pricing models are the invisible half of every quote. They decide which unit you are charged for, what happens to your bill when you hire six people or buy a warehouse full of tablets, and who absorbs the cost when something goes badly wrong at two in the morning. Two providers can quote an identical monthly total and still leave you with wildly different bills in eighteen months.

This guide breaks down the three managed IT pricing models that dominate the UK market — per user, per device and fixed fee — along with the tiered and hybrid variants that sit between them. You will find how each one is calculated, who it genuinely suits, what it hides, and how to read a quote well enough to work out which model you are actually being sold.

Table of contents

What Managed IT Pricing Models Actually Are

managed it pricing models b user tokens orbiting core

Managed IT pricing models are not prices. They are the formula that turns your business into a number, and they keep doing that every month as your business changes. Understanding the formula is what lets you predict next year’s bill instead of being surprised by it.

Why managed IT pricing models matter more than the monthly figure

The headline figure is a snapshot of today. The model is what governs every future invoice. A per-user quote that looks expensive at forty staff can be the cheaper option at sixty, while a fixed fee that looks generous today may be renegotiated the moment your estate grows past whatever assumption sat behind it. Compare formulas, not just totals.

The three families of managed IT pricing models

Almost every proposal you receive belongs to one of three families. Per-user charges by headcount and covers whatever hardware each person uses. Per-device charges by endpoint, counting laptops, servers and anything else that needs patching. Fixed fee charges one agreed sum for an agreed scope, regardless of how the underlying counts move within tolerance.

What every model is really selling

Underneath the arithmetic, all managed IT pricing models sell the same thing: someone else carrying the risk and the labour of keeping your technology working. The differences are about which variable that risk is pegged to. Choose the variable that behaves most predictably in your business and your budget stops being a guessing game.

Where hybrid quotes fit in

In practice, most 2026 proposals are hybrids. A per-user rate for staff, a separate per-server charge, licences passed through at cost, and a fixed sum for the security stack. That is fine, provided each component is itemised. The moment components are bundled into one unexplained number, comparison between managed IT pricing models becomes impossible.

Per-User Pricing: The Default in 2026

managed it pricing models c device grid with markers

Per-user has become the default across the UK market, and for good reason. It maps onto how businesses already think, budget and forecast — by headcount rather than by hardware inventory.

How per-user pricing is calculated

Per-user is the most widely adopted of the managed IT pricing models and the arithmetic behind it is deliberately simple. The provider counts people who consume support, multiplies by an agreed rate, and bills monthly. Typical UK rates sit between £45 and £120 per user per month depending on the depth of the service. A named user covers all of their devices — laptop, mobile, home screen, tablet — which is why the model became popular as flexible working spread devices across locations.

What a per-user seat usually includes

A credible seat in 2026 includes helpdesk access, endpoint monitoring and patching, endpoint detection and response, email filtering, backup with tested recovery, identity and access administration, and asset tracking. Anything below that floor is not a full seat, whatever the quote calls it. Ask for the inclusion list in writing and check it against the price band you are being sold.

Who per-user pricing suits best

Knowledge-work businesses where each person carries roughly the same technology footprint. Professional services, agencies, finance teams, software companies. If everyone has a laptop, a phone and a cloud mailbox, per-user is the fairest of the managed IT pricing models because the unit you are billed for is the unit that actually generates support demand.

The weak spots of per-user pricing

Like all managed IT pricing models, per-user has edges where it stops being fair. It charges the same for a finance director with four devices and a warehouse operative who logs in twice a week. It also quietly punishes businesses with heavy shared-device estates. Ask whether light users can sit on a reduced tier, and whether shared or kiosk logins are counted as named users — that single question can move a quote by fifteen per cent.

Per-Device Pricing: The Older Model That Still Wins

managed it pricing models d sealed cube with padlock ring

Per-device pricing is the oldest of the managed IT pricing models still in mainstream use, and it was written off far too early. In estates where machines outnumber people, or people heavily outnumber machines, it remains the more honest measure of the work involved.

How per-device pricing is calculated

Every managed endpoint carries a rate. The provider inventories workstations, laptops, servers, firewalls, switches and sometimes printers, then bills the total each month. The logic is that support effort tracks the number of things to patch, monitor and repair rather than the number of people using them, and in device-heavy estates that logic holds.

Typical per-device rates in the UK

Expect roughly £15 to £45 per workstation per month, £150 to £300 per server, and £20 to £60 for network devices such as firewalls and managed switches. Rates vary with the depth of the security stack bundled onto each endpoint. A quote that prices servers the same as laptops is either underestimating the work or excluding most of it.

Who per-device pricing suits best

Manufacturing, logistics, retail, hospitality and healthcare — anywhere shift patterns mean several people share one machine. If you have ninety staff and thirty terminals, per-device pricing can be dramatically cheaper than per-user. It also suits businesses where device management tooling is already the centre of the support model and users barely touch a helpdesk.

Where per-device pricing falls apart

It is the hardest of the managed IT pricing models to forecast, because device counts drift without anyone noticing. It also creates a perverse incentive: nobody wants to decommission an old machine that has quietly become a line of revenue. Insist on a quarterly reconciliation of the inventory, and agree in advance what happens to the bill when devices are retired mid-term.

Fixed-Fee and All-Inclusive Agreements

managed it pricing models e three armed balance beam

Fixed fee is the most misunderstood of the managed IT pricing models. Sold well, it is the cleanest budget line you will ever have. Sold badly, it is a per-user contract with the tolerances hidden in an appendix.

What “fixed fee” actually means

One agreed monthly sum covers an agreed scope for an agreed term, regardless of ticket volume. It does not usually mean unlimited anything. Almost every fixed-fee agreement contains a band — a range of users, devices or sites within which the price holds — and a mechanism for what happens if you leave that band.

How providers price a fixed-fee agreement

Unlike the metered managed IT pricing models, the provider here is quoting a risk position rather than a unit count. They model your likely support consumption, add a risk premium for the variance they are absorbing, and quote a figure that is comfortably above their expected cost. That premium is the price of your predictability, and it is usually worth paying if your board values a flat line in the budget more than the lowest possible number.

Who fixed-fee agreements suit best

Businesses with stable headcount, regulated firms that need cost certainty for planning, and organisations where finance genuinely hates variable IT bills. It also suits companies coming out of a painful break-fix arrangement, where the psychological value of a fixed number outweighs the small premium attached to it.

The risks buried in an all-inclusive promise

Read the fair-use clause, the growth tolerance and the exclusion list before you sign. Common exclusions include projects, hardware, third-party licences, out-of-hours work and anything caused by systems the provider did not recommend. All-inclusive rarely means all-inclusive; it means everything inside a scope somebody else wrote.

The Models That Sit Between the Big Three

managed it pricing models f iceberg hidden cost layers

Not every proposal fits neatly into the three families. Four intermediate structures appear often enough in UK quotes to be worth understanding before you compare them against the mainstream managed IT pricing models.

Tiered packages

Bronze, silver and gold, or whatever the provider calls them. Tiers are convenient and almost always engineered so the middle option looks obvious. Read the jump between tiers carefully. Sometimes the only difference is out-of-hours cover you will never use; sometimes it is the security tooling you genuinely need to hold a client contract.

Block hours and pay as you go

You buy a bank of hours in advance and draw down against it, or you pay per incident. This is not managed support and should not be compared with it directly. It suits businesses with a capable internal technician who occasionally needs escalation, and it is a poor fit for anyone who wants someone else to own the risk.

Monitoring-only and co-managed

A lighter arrangement where the provider supplies tooling, monitoring and second-line escalation while your own team handles day-to-day requests. Priced per endpoint or per seat at a fraction of full support. Co-managed agreements are growing quickly in the UK mid-market because they let a small internal team punch far above its weight.

The hybrid quote that is now the norm

Most quotes now combine a per-user core with per-server charges, per-site charges and licence pass-through. Hybrids are perfectly reasonable, but they make like-for-like comparison hard. Ask every provider to break their hybrid into its components so you can see which of the managed IT pricing models is doing the heavy lifting in each proposal.

Comparing Managed IT Pricing Models Side by Side

The right comparison is not “which is cheapest today” but “which behaves best as my business changes”. Set the three managed IT pricing models against four practical criteria and the choice usually resolves itself.

Predictability

Fixed fee wins outright, per-user comes second because headcount changes slowly and visibly, and per-device comes last because inventories drift. If your finance team needs a number they can defend twelve months ahead, that ordering should weigh heavily in the decision.

Fairness as you grow

Per-user scales cleanly with hiring and is easy to challenge line by line. Per-device scales with capital expenditure, which is lumpier and harder to forecast. Fixed fee is the fairest until you cross a tolerance band, at which point renegotiation replaces arithmetic and the conversation gets considerably less comfortable.

Ease of comparison between providers

Per-user is by far the easiest to compare across suppliers because the unit is standard. Per-device requires you to normalise inventories first. Fixed fee is the hardest of all, since two fixed fees may cover entirely different scopes. Where possible, ask every bidder to also express their quote in per-user terms.

The behaviour each model encourages

Per-device can reward keeping old machines alive. Block hours reward slow resolution. Fixed fee rewards prevention, because every ticket costs the provider money. Per-user sits in the middle. When you compare managed IT pricing models, ask which behaviour each one quietly funds over a three-year term.

What Drives the Number Inside Every Model

Two businesses with identical headcounts can be quoted figures fifty per cent apart. All three managed IT pricing models are sensitive to the same handful of variables, and most of those variables are within your control over a two-year horizon.

Headcount and the economics of scale

Per-user rates fall as user counts rise. A five-person business often pays £95 or more per seat because onboarding, documentation and tooling costs are spread across almost nobody. The same provider might charge £58 at fifty users and £46 at two hundred. If you sit near a band boundary, ask what the rate becomes at the next tier.

The device-to-user ratio

This single number decides which of the managed IT pricing models will be cheapest for you. Divide managed endpoints by staff. Below about 1.2 devices per person, per-device tends to win. Above roughly 1.8, per-user usually wins. Between the two, the models converge and the decision comes down to forecasting preference.

Security and compliance requirements

Cyber Essentials Plus, ISO 27001, SOC 2 or a client’s security questionnaire all add tooling, evidence gathering and audit time. Expect a genuine cybersecurity uplift of £8 to £25 per user per month over a baseline package. Providers who fold that into an unexplained bundle are usually the ones who have not costed it properly.

Service hours and response commitments

Business-hours cover is the baseline. Extending to evenings, weekends or genuine round-the-clock support roughly doubles the staffing model behind your account, and the price follows. Faster contractual response times cost money for the same reason: somebody has to be available and idle for you to get them quickly.

Site count and geography

Multiple sites mean multiple network stacks, more travel and more on-site visits. Most UK providers charge a per-site fee on top of the per-user or per-device core, typically £100 to £400 per site per month. Ask how many on-site visits are included before travel is billed separately.

Hidden Costs Every Model Carries

The headline rate is rarely the whole invoice. Every one of the managed IT pricing models has a set of charges that live outside it, and the gap between a cheap quote and an honest one is usually found here.

Onboarding and discovery

Expect a one-off fee covering documentation, tooling deployment, security baselining and knowledge transfer. Anywhere between £1,500 and £15,000 depending on estate size and mess. A provider quoting zero onboarding is either amortising it invisibly into the monthly rate or planning to do very little discovery.

Licence pass-through

Licences sit outside all three managed IT pricing models more often than not. Microsoft 365, endpoint security, backup and email filtering are frequently quoted separately, sometimes at cost and sometimes with a margin. Confirm which. On a fifty-user estate the licence line alone can exceed £2,000 a month, and a small percentage margin on top of it materially changes the comparison between two proposals.

Project work and exclusions

Managed support keeps the lights on. Migrations, office moves, new site builds and major hardware replacement are projects, quoted separately at day rates of £600 to £1,200. Ask where the boundary sits and get examples in writing, because that boundary is the single most common source of dispute in year one.

Out-of-scope and out-of-hours charges

Work outside the agreed scope or hours is billed at a premium, often 1.5 to 2 times the standard rate. This is reasonable, but only if scope is defined tightly enough that you can predict when you are crossing the line. Vague scope plus premium rates is a structural problem, not a pricing detail.

Exit and offboarding

The least examined cost in the whole agreement. Ask what happens to your documentation, tenant administration rights, backup archives and licence agreements when you leave. Ask whether offboarding is chargeable and at what rate. A provider who cannot answer that clearly has told you a great deal about the relationship you are about to enter.

How to Match a Model to Your Business

Choosing between managed IT pricing models is a matter of arithmetic first and judgement second. Do the arithmetic honestly and the judgement gets much easier.

Start with the shape of your estate

Count staff. Count managed endpoints, including servers, network equipment and anything shared. Work out the ratio. Note how many people share machines and how many carry three or more. That inventory takes an afternoon and eliminates at least one of the three models before you speak to a single provider.

Then look at how you actually consume support

Pull twelve months of ticket data if you have it. High ticket volume from a small number of heavy users argues for a fixed fee, where variance is somebody else’s problem. Low, predictable volume argues for per-user or per-device, where you are not paying a premium for risk you were never going to trigger.

Then stress-test against your growth plan

Model your quote at plus twenty per cent headcount and minus ten per cent. Model it after an acquisition, and after a site closure. The best of the managed IT pricing models for your business is the one whose bill you can still predict in every one of those scenarios without a renegotiation.

A simple decision path

If devices roughly equal people and headcount is stable, take per-user. If devices are heavily shared or heavily outnumber staff, take per-device. If your board values certainty above all and your estate is stable, take fixed fee. If none of those describe you, expect a hybrid — and insist that it be itemised. Most businesses land on one of the three managed IT pricing models within an hour of doing this exercise properly.

Questions to Ask Before You Sign

A provider’s answers to a short list of structural questions tell you more than any glossy proposal. These are the ones worth asking in every procurement conversation, whichever of the managed IT pricing models is on the table.

Questions about the unit

What exactly counts as a user? Are shared logins, service accounts, contractors and leavers included? What counts as a managed device, and are network devices and virtual servers billed separately? How and when is the count reconciled, and what evidence will you provide for it?

Questions about the ceiling

Is there a fair-use limit on tickets or hours? What happens when it is exceeded? Within what tolerance does the fixed fee hold, and what is the mechanism when we cross it? Is there an annual uplift clause, and is it capped or tied to an index such as CPI?

Questions about change

What happens to the price if we grow by twenty people or close a site? Are new starters billed pro rata? How much notice is needed to remove seats or devices? Can we move between managed IT pricing models mid-term, and does that carry a fee or a fresh term?

Questions about exit

What is the notice period and is there a minimum term? What is handed over on exit and in what format? Are administrative rights to our own tenants and platforms transferred without charge? Is offboarding support chargeable, and if so, at what rate and with what cap?

Switching Between Managed IT Pricing Models

Changing model mid-relationship is common and usually cheaper than changing provider. It is also the fastest route to a better price if your business has drifted away from the assumptions in your original contract.

When switching is worth it

If your device-to-user ratio has moved by more than about thirty per cent since signing, your position among the managed IT pricing models is probably no longer the right one. The same applies after an acquisition, a shift to hybrid working, or a move from on-premises servers to hosted platforms. Each of those changes the variable your bill is pegged to.

How to run the comparison honestly

Rebuild your last twelve months of invoices under each of the alternative managed IT pricing models using real counts, not estimates. Include projects, out-of-hours work and licences. Most businesses discover the difference is smaller than expected — and that the real saving sits in a cost optimization exercise on scope rather than in the model itself.

Timing the change

Renewal is the obvious moment, but the four to six weeks before renewal is when you have the most leverage and the least time pressure. Start the conversation early, arrive with your own numbers, and be specific about which model you want and why. Providers respond very differently to a buyer who has done the arithmetic.

Frequently Asked Questions About Managed IT Pricing Models

The same handful of questions comes up in almost every procurement conversation. Short, direct answers to each of them follow.

Which of the managed IT pricing models is cheapest?

There is no universal answer, which is exactly why the models exist. Per-device is usually cheapest where machines are shared, per-user where each person carries several devices, and fixed fee is rarely the cheapest but is frequently the most predictable. Cheapest and best value are not the same measurement.

Is per-user always better than per-device?

No. Per-user has become the default because it suits knowledge-work businesses, which make up most of the market. In shift-based operations with shared terminals, per-device can be forty per cent cheaper for identical service. The right answer comes from your device-to-user ratio, not from market fashion.

Should servers be priced separately?

Usually yes, and transparency here is a good sign. Servers carry materially more monitoring, patching and backup work than a laptop, so folding them into a flat per-user rate either overcharges server-light businesses or undercharges server-heavy ones. Separate lines make the quote easier to compare and easier to challenge.

How often should the pricing model be reviewed?

Managed IT pricing models should be reviewed annually as a minimum, and immediately after any structural change to headcount, sites or estate. A model chosen for a thirty-person single-site business rarely still fits at eighty people across three locations. Reviewing does not mean switching; it means confirming the formula still matches the business it is describing.

Can the model itself be negotiated?

Yes, and more easily than the rate. Providers will often move between managed IT pricing models to win or retain an account even when they will not move far on price, because a different model can be more profitable for them as well as cheaper for you. Ask for two structures rather than a discount.

Getting the Decision Right

The gap between the best and worst outcome here is not a few pounds per user. It is whether your IT budget behaves predictably for three years or becomes a recurring argument. Managed IT pricing models are the mechanism that decides which of those you get.

Do the arithmetic before the meetings

Comparing managed IT pricing models is not difficult, but it does need numbers. Count your users, count your devices, work out the ratio and pull your ticket history. Thirty minutes of preparation converts a sales conversation into a procurement exercise. It also makes it obvious very quickly which providers understand your business and which are reciting a standard package.

Insist on itemisation

Every component of a quote should have a name, a unit and a rate. Bundled numbers are not simplicity; they are a defence against comparison. If a provider will not itemise, that is information about how the relationship will run once the contract is signed.

Choose the model your business will still fit next year

The best of the managed IT pricing models is the one that still describes your business after you grow, shrink, acquire or consolidate. Pick for the shape you are becoming rather than the shape you are today, and revisit the choice at every renewal rather than only when something has gone wrong.