IT service desk software is sold as one category and bought in four tiers of capability. Most teams under 500 people need the first two, pay for three, and configure one. Working out your tier first makes the vendor comparison short.
TL;DR
- The four tiers are a ticket queue, request fulfilment, asset linkage, and change management.
- Teams under 500 people almost always need the first two and rarely configure the fourth.
- Buying for a tier you will not configure is the main source of overspend in this category.
- Pricing is per agent almost everywhere, so your resolver count matters more than your tier choice.
- Two vendors publish full ladders and the heavier platforms quote, which shapes how long your evaluation takes.
- If most of your inbound is questions rather than work, no tier of this category reduces it.
The four tiers of IT service desk software
Tier one, a ticket queue. Work arrives, gets assigned, gets tracked, gets closed. Everything in this market does this competently and it is the only tier every team genuinely needs.
Tier two, request fulfilment. A catalogue of things people can ask for, with approval routing attached. New laptop, software licence, access to a system. This is where most of the value sits for a team under 500 people, because it converts ad hoc messages into a repeatable path.
Tier three, asset linkage. Tickets connected to the specific device or licence they concern, so history travels with the asset. Useful once you have a few hundred devices and somebody is asking which machine keeps breaking.
Tier four, change management. Change requests, advisory boards, scheduled windows, configuration relationships. Genuinely necessary in regulated environments at scale and almost never configured below a thousand employees.
The pattern worth noticing is that vendors price by tier and teams buy aspirationally. A platform bought for tier four and used at tier two is the most common overspend in this market, and it is invisible because the capability is genuinely there.
Working out your tier in an afternoon
What to do: take a fortnight of inbound and answer four questions about it.
- How much of it is work rather than questions? Questions need a different product entirely.
- How many distinct repeatable request types are there? Under about eight and a simple queue covers you.
- Does anybody ever ask which device a ticket concerned? If never, tier three is speculative.
- Has anybody outside IT asked for a change record? If no, tier four is not your requirement.
Most teams who do this land on tier two and are surprised, because the shortlist they had assembled was built from tier-four feature comparisons. The exercise takes an afternoon and routinely removes two vendors from consideration.
What the options publish
Every figure below was read from the vendor’s own page on 15 October 2026. The division between vendors who publish and vendors who quote matters because it sets how long your evaluation takes.
| Option | Published pricing | Strongest tier |
|---|---|---|
| Jira Service Management | Free to 3 agents, Standard $21.50, Premium $55.28 per agent per month | One and two |
| Freshservice | Starter $19, Growth $49, Pro $99 per agent per month, billed annually | Two and three |
| Zendesk | Support Team $19, Suite Team $55, Suite Professional $115 per agent per month, paid yearly | One |
| ServiceNow ITSM | No readable figure published | Four |
| Matram | $29 Basic, $69 Standard, $199 Pro per month | None of the four |
That last row is deliberate rather than dismissive. An answering layer is not IT service desk software at any tier, and it appears here because teams searching this category frequently have a question problem rather than a ticket problem.
Why the agent count beats the tier choice
Almost everything here is priced per agent, which means your resolver count multiplies whatever tier you pick. Getting the count wrong costs more than picking the wrong tier.
Jira Service Management Standard at 12 agents is $258 a month, and at 20 agents it is $430 a month.
Freshservice Growth with its AI add-on works out at $78 an agent, so 12 agents is $936 a month and 20 is $1,560 a month.
Checklist for counting agents: everybody who closes tickets, the office manager handling hardware, whoever approves access, regional cover in other timezones, and anybody who needs to see the queue. Teams underestimate that list by about a third.
The capability you cannot buy
No tier of IT service desk software reduces the number of questions people ask. It files them, routes them and reports on them, all of which is useful and none of which is fewer.
That matters because a meaningful share of teams buy a platform expecting volume to fall, and volume does not fall. Tickets that were previously messages become tickets, which looks like an increase, and the platform gets blamed for a problem it was never going to solve.
Matram publishes $29 Basic, $69 Standard and $199 Pro per month, flat, with unlimited seats and a 30-day trial that takes no card. Disclosure that should shape how you read this: Matram is owned by the same people who publish this site.
Where it struggles, stated plainly: there is no free tier; it answers questions but does not action tickets, so nothing is provisioned or closed; and it is not an ITSM platform, so there is no asset register and no change management. It belongs beside a service desk rather than instead of one.
Verify the published figures yourself, since pricing here moves within weeks: Jira Service Management, Freshservice and Matram all publish openly.
Final thoughts
The shortest route through an IT service desk software evaluation is to establish your tier before you look at a single vendor page. Four questions about a fortnight of inbound will do it, and the answer is tier two far more often than the shortlists suggest.
Then count your resolvers properly, because that number multiplies everything. And if the inbound turns out to be mostly questions, the honest conclusion is that this category is not your purchase, however good the products are at what they do.
What is IT service desk software?
It is the system your internal IT function uses to receive, track and close work from colleagues, and in practice it sells four tiers of capability: a ticket queue, request fulfilment with approvals, asset linkage, and change management. Vendors price by tier and teams tend to buy aspirationally, which is why a platform bought for the top tier and used at the second is the characteristic overspend here.
Which tier do we actually need?
Answer four questions about a fortnight of real inbound: how much is work rather than questions, how many repeatable request types exist, whether anybody ever asks which device a ticket concerned, and whether anybody outside IT has asked for a change record. Most teams under 500 people land on tier two, meaning a queue plus request fulfilment, and never configure the heavier capabilities.
How much should we budget?
Almost everything in this category is priced per agent, so start from your resolver count rather than from the tier you think you need. The worked sums for a 12-agent and a 20-agent team sit in the comparison above, with each vendor’s own figures attached. The headline point is that at identical agent counts the spread between products is wide enough to decide the purchase on its own, which is why the count comes first.
Will it reduce our ticket volume?
No, and expecting it to is the most common disappointment here. A service desk files, routes and reports on work; it does not make there be less of it. Volume often appears to rise after implementation, because messages that were previously informal become tracked tickets. If reducing repeat questions is your goal, that is a different product category entirely.
Do we need asset management in the same tool?
Only if somebody actually asks which device a ticket concerned, which in practice starts mattering around a few hundred devices. Below that, linking tickets to assets is capability you will pay for and not populate, and an unpopulated asset register is worse than none because people trust it. Check whether anybody has asked that question in the last six months before paying for the tier.
Should we consider the enterprise platforms?
Above a few thousand employees with genuine change management and compliance requirements, yes, and nothing lighter competes there. Below that, the implementation effort is disproportionate and the pricing is not published, so every evaluation begins with a sales conversation. For a team under 500 people the two vendors publishing full ladders are the practical starting point.
