MDM Pricing: A Practical Guide to What Vendors Publish and What They Do Not

TL;DR

• Half the category does not publish a price. Mosyle, Iru and Esper quote. Jamf, Hexnode, Addigy, Scalefusion and NinjaOne publish, and that split matters more than the figures.
• The cheapest real option is zero. Apple Business replaced Apple Business Manager on 14 April 2026 and includes Apple's own built-in MDM at no cost.
• You may already own management: Intune Plan 1 capability comes with Microsoft 365 E3 and E5, and Google endpoint management is included across essentially every Workspace edition.
• The published range is wide: from $1.50 per device per month at volume to $16 per Mac for a security bundle. Most of that spread is scope, not margin.
• Watch the unit. Per device and per user are different numbers once phones and tablets outnumber people, and one vendor's price gets worse as you get smaller.
• The real cost is rarely the licence. Migration, unused capability and the administrator you did not hire all cost more than the per-device figure.

Why This Comparison Is Harder Than It Should Be

A finance director asks a reasonable question: what does device management cost per head. The IT lead opens six vendor sites and comes back with three numbers, two quote forms and one page that lists prices for bundles that do not map onto each other.

That is not evasion on anybody's part, mostly. It is that the category sells four different things under one name, and the vendors who publish prices are pricing different scopes. One figure covers management alone. Another covers management plus identity plus endpoint security. A third is per user rather than per device. A fourth gets cheaper the bigger you are, and a fifth gets more expensive the smaller you are.

So a table of headline figures is close to useless on its own, and most roundups stop there. What follows is the published numbers, verified against each vendor's own page on 3 October 2026, and then the harder part: what the figures include, what the unit means, and which costs sit outside the licence entirely.

Recommended Tools

Best tools for Device Management

B BambooHR HR management system L Lattice Performance management D Deel Global payroll & EOR

View all recommended tools

One note on method, because it matters in a category this stale. Every figure below came from the vendor's own pricing page on that date. Where a vendor does not publish, this says so rather than estimating, and where a figure covers a bundle rather than a single product, that is stated. Vendor pricing in this category moves and rebrands faster than most published advice tracks, and a number without a date attached is not information.

What Each Vendor Actually Publishes

Three things in that table deserve more than a glance, and all three are ways a reader gets the comparison wrong.

Three Traps in the Published Numbers

The bundle trap

Jamf for Mac at $12.50 per macOS device per month is the highest general-purpose figure in the table, and it is not comparable to the figures above it. That price bundles Jamf Pro for management with Jamf Connect for identity and Jamf Protect for endpoint security. It does not price management on its own.

Set against a management-only licence at $2.20, it looks like five times the cost. Set against a management-only licence plus a separate endpoint security product plus an identity bridge, the gap narrows considerably and sometimes closes.

The same applies to Addigy, which publishes a management figure from $8.25 per Mac and a Security Suite figure from $16, making the scope difference explicit in a way most vendors do not.

So the exercise is not comparing headline figures. It is writing down what each bundle contains, striking out what you already buy elsewhere, and comparing what remains. A team already running an endpoint detection product and an identity provider is being quoted for two things it does not need. A team with neither is getting three products for one price.

The unit trap

Per device and per user are not interchangeable, and the gap grows as an organisation matures.

A company of 100 people might have 100 laptops at year one and 100 laptops plus 70 phones plus 30 tablets at year four. Per-user pricing is flat across that change. Per-device pricing has doubled.

Neither model is dishonest. They suit different fleets, and the direction of travel matters: device counts per person tend to rise over time rather than fall. Model both at your projected size rather than today's, and if you are comparing a per-user figure with a per-device figure, convert them both to annual cost at your own numbers before deciding anything.

The volume trap

Almost all per-device pricing improves with scale, which is why most teams read the lowest published number as their likely rate.

NinjaOne publishes the opposite shape and states it plainly: as low as $1.50 per device per month at 10,000 endpoints, rising to $3.75 at 50 or fewer endpoints. A team of forty budgeting from the headline figure will be surprised by a quote that is two and a half times higher.

The general lesson applies beyond one vendor. Find your own device count on each vendor's own curve rather than reading the most attractive figure on the page, and where a vendor publishes a range, assume you are at the wrong end of it until something tells you otherwise.

Why Half the Category Does Not Publish

Worth understanding, because the usual assumption is that quote-based means expensive and that is not reliably true.

Vendors withhold pricing for a handful of reasons, and only one of them is about the number being high. Some sell through resellers and cannot publish a figure the channel then has to honour. Some price by scope rather than by seat, so there is no single number to publish. Some use the sales conversation to qualify, which is a commercial choice rather than a pricing one. And some are simply mid-change and would rather not put a figure in writing.

What quote-based reliably costs you is calendar time. A sales cycle adds days or weeks between deciding to evaluate and knowing whether you can afford it, and on a decision with a deadline that delay is itself a cost. If you are trying to settle this inside a quarter, the vendors who publish let you eliminate options in an afternoon.

It also costs you the ability to compare honestly. A quote reflects what the vendor thinks you will pay, which is not necessarily what the next organisation pays for the same thing. Published pricing is checkable and a quote is not, and for a buyer that difference is worth something independent of the figure itself.

There is one genuine advantage to quote-based pricing that is worth acknowledging rather than dismissing. A quote can be negotiated and a published rate card usually cannot, so an organisation with real volume or a compelling reference value may do better from a conversation than from a list price. That advantage is real at scale and close to worthless below about a hundred devices, where you are taking the standard rate anyway and have simply spent two weeks to learn it.

A useful tactic when dealing with quote-based vendors: ask for the figure at your projected device count as well as today's, in the same conversation and in writing. Growth pricing is where surprises live in this category, and a vendor quoting a competitive rate at your current size has no obligation to hold it when phones and tablets have accumulated faster than headcount.

One practical note for anybody building a business case: do not let a colleague fill a quote-based vendor's row with an estimate from a roundup. Estimated prices for vendors who decline to publish their own are how a business case acquires a number nobody can defend, and it falls apart the moment somebody asks where the figure came from.

What You May Already Be Paying For

The single largest saving available in this category is not a cheaper vendor. It is discovering you already hold the capability.

Microsoft 365 E3 and E5 include Intune Plan 1 capability. That covers device management across Windows, macOS, iOS and Android, configuration profiles, compliance policies, application deployment and the conditional access tie-in through Entra ID. If your organisation holds either licence, the marginal cost of managing devices is nothing.

Google Workspace includes endpoint management across essentially every edition, including Business Starter, Standard and Plus, Enterprise Standard and Plus, Frontline tiers, Essentials and Cloud Identity. Basic mobile management is applied automatically with no user setup and no additional cost. Advanced mobile management asks users to install an app and adds remote wipe, stronger password enforcement and Android work profiles.

Apple Business is free. Since 14 April 2026 Apple has retired Apple Business Manager, Apple Business Essentials and Apple Business Connect, replacing all three with one free platform carrying Apple's first built-in mobile device management. You need it for automated enrollment regardless of what else you run, so there is no scenario where trying it costs anything.

Check all three before evaluating a paid platform. This takes ten minutes, it requires no vendor conversation, and it regularly removes the need for the rest of the exercise.

The Costs That Are Not the Licence

The per-device figure is usually the smallest number in the real total, which is why a comparison built only on licence cost tends to produce the wrong answer.

The administrator. The largest hidden cost by a wide margin. A platform with a scripting layer and deep configuration assumes somebody who will use it. If that person does not exist, you are paying for capability that will go unused, and the honest comparison is between a cheaper platform and the salary of the person who would justify the expensive one.

The migration. Every device unenrolled from the old platform and enrolled into the new one. In an office that is a trolley and an afternoon. Across a distributed fleet it means asking each person to act on their own machine, the stragglers take weeks, and you run two consoles until the last one completes. On a 400-device fleet this is a programme, and it frequently exceeds the first year of licence saving.

The unused tier. Buying the plan you expect to grow into, then running basic policies on it for three years. This is the same mistake at every scale and it is invisible in a price comparison because the licence was never the problem.

The procurement channel. Automated enrollment requires buying through Apple, an enrolled reseller, or a zero-touch reseller for Android. If your organisation buys hardware locally in emergencies, some devices will sit permanently outside automated enrollment regardless of what you pay, and the cost shows up as manual work forever.

The second tool. A free or cheap option covering one platform is not cheap for a mixed fleet. It is cheap for part of one, and the rest needs a plan, which is often a second licence plus the reconciliation work between two consoles.

The sales cycle. Days or weeks of calendar time on quote-based vendors, which is a real cost when a decision has a date attached.

The exit. Rarely costed and occasionally expensive. When you leave a platform, its record of what was configured, when and by whom leaves with it unless somebody exports first, and that history is sometimes the only documentation of a decision made years earlier. Budget an afternoon for the export before the licence lapses rather than discovering the gap during an incident review, because the alternative is reconstructing from memory what a departed administrator did and why.

Modelling Your Own Number

Five steps, none of which needs a vendor, and together they produce a figure you can defend.

One. Count devices and people separately. Both numbers, today and projected two years out. This is what lets you compare per-device and per-user pricing honestly rather than by feel.

Two. Write down your platform split. How many Apple, how many Windows, how many Android, and how many are dedicated or single-purpose. A single figure for the fleet hides the fact that different parts of it may need different tools.

Three. List what you already pay for that overlaps. Endpoint security, identity, patch management, monitoring. This is what turns a bundle price from expensive into competitive, or confirms it is genuinely expensive for you.

Four. Check both entitlements. Microsoft 365 and Google Workspace. Before anything else, because the answer may be zero.

Five. Add the non-licence costs. Migration effort in person-weeks, and honestly whether an administrator exists for a platform that needs one.

Then and only then compare annual totals at your own device count, with each vendor's own volume curve applied. The number that comes out of that exercise bears very little resemblance to the lowest figure on anybody's pricing page, and it is the only one worth taking into a budget conversation.

A Worked Example

Abstract per-device figures are hard to reason about, so here is the same organisation priced three ways. A company of 150 people with 150 laptops, 60 phones and 20 tablets, so 230 devices. Of the laptops, 90 are Macs and 60 are Windows.

The point of this exercise is not the totals. It is that the three columns differ by more than the per-device rates suggest, because the unit and the scope do most of the work.

Three observations, and they are the reason a pricing page cannot answer this question for you.

The first column is free at the margin, and for a great many organisations that is the correct answer rather than a compromise. The second column is a real number but an incomplete one, because endpoint security and identity are not in it and will be bought somewhere. The third column is the largest and also the most complete, since it includes two products the second column has not yet paid for.

Which means the honest comparison between columns two and three is not $506 against $1,585. It is $506 plus whatever endpoint security and identity cost for 230 devices, against $1,585. That gap is much narrower than the headline, and on some quotes it closes entirely.

The figures above are arithmetic on published list prices rather than quotes, and no vendor has agreed to them. Volume terms, bundle discounts and minimums will move all three columns. Use the shape of the comparison rather than the totals.

The Full Cost, Not the Licence

Licence cost is one line of six, and the five below it are where comparisons usually go wrong.

The last row deserves a note because it is the only one no vendor can fix. Devices must be bought through Apple, an enrolled reseller, or a zero-touch reseller for Android to appear in automated enrollment. Every device bought outside those channels needs manual handling for the rest of its life. If your organisation buys hardware locally during emergencies, that manual cost is already in your operation and no licence tier removes it.

The Decision Table

The first three rows cost nothing and cover a large share of readers. Work through them before the rest of the table, because the cheapest outcome of a pricing exercise is discovering you do not need to spend.

Where Teams Get MDM Pricing Wrong

Comparing bundles with single products. The most common error in every roundup of this category. Jamf at $12.50 includes management, identity and security. Compare like with like or the comparison is meaningless.

Reading the lowest published figure as their rate. Published ranges exist because pricing varies by tier and volume. Assume the wrong end of the range until a quote says otherwise, and check which direction the volume curve runs.

Mixing per-user and per-device figures. Convert both to annual cost at your own counts before comparing. The two models diverge as device counts per person rise, which they reliably do.

Estimating prices for vendors who do not publish. It produces a business case with an indefensible number in it. Write quote-based and get the quote.

Forgetting the entitlement check. Buying device management while holding a licence that already includes it. Ten minutes to rule out, and it happens constantly.

Costing the licence and not the migration. On a distributed fleet the migration regularly exceeds the first year of savings, which can invert the entire comparison.

Buying the tier you hope to need. Then running basic policies on it for three years. The licence was never the problem and a price comparison will never reveal this one.

Trusting a price without a date. Vendor pricing in this category moves and rebrands quickly. A figure in an undated article is not information, and the same applies to the figures here once enough time has passed.

Two illustrations of how fast that happens, both from the twelve months before this was written. Kandji rebranded to Iru in October 2025, so any comparison naming Kandji as a current vendor is at least a year stale. Apple retired Apple Business Manager in April 2026, so advice describing it as a separate thing you set up predates a change that affects every Apple deployment. And Jamf, which for years declined to publish a rate card, now publishes one, meaning articles asserting that Jamf pricing is quote-only are both out of date and understating what a buyer can find out for themselves. Three material changes in a category whose published advice tends to be refreshed annually at best, which is the argument for checking a vendor's own page rather than trusting any roundup, including this one.

What No Price Covers

Every platform in that table knows about devices it manages, and no tier of any of them changes what follows.

None can tell you about a machine that was never enrolled, because none has seen it. None can tell you where a laptop physically is, only where it last checked in from. None can tell you whether the machine issued to somebody who left in March came back, only that it stopped reporting. And none can tell you what you own, which is the question an auditor actually asks.

That is an asset management gap rather than a shortcoming of any product, and it is the one cost in this article that cannot be addressed by moving up a pricing tier. The fix is a register alongside the MDM. Platforms built for that side of the problem, such as RemoAsset, handle procurement, delivery and physical retrieval, with enrollment happening on delivery. RemoAsset is not an MDM and does not replace anything in the table above.

The diagnostic is one question. Ask what went wrong last quarter. A misconfigured, unpatched or unencrypted machine is an MDM problem and this pricing comparison is the right one to be running. A laptop nobody can account for is not, and the budget belongs in a different line.

What to Establish Before You Commit

Devices and people, counted separately, today and projected. The two numbers that make per-device and per-user pricing comparable.

Your platform split, including dedicated devices. Because one figure for the fleet hides the parts that may need different tooling.

Both entitlements, confirmed. Microsoft 365 and Google Workspace, checked rather than assumed.

What overlapping products you already buy. Endpoint security, identity, monitoring, patch management. This decides whether a bundle is good value for you specifically.

Whether an administrator exists. For any platform whose value depends on depth, this is the question that determines whether the licence is an investment or a donation.

The migration cost in person-weeks. Including how many remote staff must act on their own machines.

A date on every figure you collect. Yours and anybody else's.

The figure at your projected size, in writing. Not just today's, because the number you will actually pay is the second one.

None of those eight requires a vendor conversation except the last, and that one should be requested in the same call rather than discovered at renewal.

None of those seven requires a vendor conversation. They are answerable inside your own organisation in about a day, and they turn a list of headline prices into a number you can defend.

What Getting This Wrong Costs

The first cost is paying for what you already own, and it is the most common. A device management licence running alongside a Microsoft 365 or Workspace entitlement that already covered it, invisible until a renewal review asks what it is for. Recoverable, embarrassing, entirely avoidable in ten minutes.

The second is the unused tier, and it is the largest in absolute terms. An organisation buys for capability, nobody is staffed to operate it, and three years later it runs policies a free option would have enforced. The comparison that justified the purchase was not wrong. It was simply about capability nobody was hired to use, and no amount of price research catches this because the price was never the issue.

The third is the migration paid twice. Choosing on headline price, discovering the cheaper platform cannot do something load-bearing, and migrating again. Every device unenrolled and re-enrolled, twice, which is why the modelling exercise above is worth an afternoon.

The fourth is the business case that collapses. A finance conversation built on estimated figures for vendors who do not publish, which holds until somebody asks where the numbers came from. Quote-based is a defensible answer in a business case. An invented figure is not, and being caught with one costs credibility that is slow to recover.

When You Are Ready to Decide

Check the entitlements first. Microsoft 365 E3 or E5, any Google Workspace edition, and Apple Business if your fleet is Apple. For a great many organisations the answer is that management is already paid for or free, and that check costs ten minutes against a comparison that costs weeks.

Then count devices and people separately, write down the platform split, and list what overlapping products you already buy. Those three facts convert a page of headline prices into a comparison that means something, because they tell you which bundles are good value for you specifically rather than in general.

Then eliminate on published pricing where you can. The vendors who publish let you rule options in or out in an afternoon, which is worth real money when a decision has a date on it. Request quotes from the remainder, and write quote-based in your business case rather than an estimate.

And add the costs that are not the licence before signing anything: the migration in person-weeks, and whether the administrator a deep platform assumes actually exists. Those two numbers change more decisions than any per-device figure in this article.

A closing word on what a pricing exercise is actually for. The purpose is not to find the lowest number, because the lowest number in this category is zero and it is available to a large share of readers through an entitlement they already hold or a free platform from the company that made their hardware. The purpose is to establish what you are buying that you do not currently have, and what it would cost to operate rather than merely to license. Done properly, that exercise ends one of three ways: you discover management is already paid for, you discover the cheapest adequate option is adequate, or you discover the expensive option is worth it for a reason you can name in one sentence. Any other outcome means the exercise is not finished.

Frequently Asked Questions