
ITSM for SMEs: three processes, and the rest of ITIL can wait
ITSM sounds like something for organisations with an in-house IT department of fifty people. It is simpler than that: agreements on how a request comes in, how a change goes ahead and what equipment you have. A small or mid-sized business needs three of them, one page each, and no certificate.
ITSM is how you deliver IT as a service: how incidents, changes and equipment are handled. ITIL is a framework that describes it, and an ITSM tool is software that supports it; they are three different things.
A small or mid-sized organisation genuinely needs three processes: incidents and requests, changes, and assets. Those three cover most of what goes wrong in practice.
Certification, a change advisory board and a full configuration database can wait until a customer or a regulation asks for them. Start with the process, and only then choose the tool.
What ITSM is, and what it is not
ITSM stands for IT service management: the agreements that let you deliver IT as a service to the rest of the organisation rather than as a collection of one-off jobs. Who reports a problem, where does that report land, who decides that an update can go in tonight, and who knows which laptop belongs to which employee.
Three terms get mixed up here, and that costs time. ITSM is the way of working. ITIL is the best-known framework describing that way of working, with dozens of components it calls “practices”. An ITSM tool is software with tickets, an asset list and workflows. You can do ITSM without following ITIL, and you can buy an ITSM tool without a single agreement changing.
For an organisation with twenty to a few hundred users, the question is therefore not “should we implement ITIL?” but: which agreements are we missing, and where does that make things go wrong?
The three ITSM processes you actually need
Look back at what went wrong with IT over the past year and it almost always falls into one of three buckets: a ticket that sat untouched, a change that broke something else, or a device or account nobody knew was still there. Those are exactly the three processes to start with.
- Incidents and requests: how a report comes in, who picks it up and when it may be closed.
- Changes: how an adjustment to systems is announced, approved and rolled back if needed.
- Assets: which devices, licences and accounts exist, who they belong to and what state they are in.
The three are linked. An incident concerns an asset, a change touches assets, and a change that goes wrong becomes an incident. That is why they only work well when they sit in the same system, or at least refer to each other.
Incidents and requests: one way in, one priority
Agree on one way in. Reports by email, phone, a chat message to the colleague who is good with computers and a note on the desk cannot be counted or tracked. Choose one channel where everything ends up as a ticket; a phone call is fine, as long as someone turns it into a ticket.
Separate faults from requests. “The printer is not working” is an incident: something worked and no longer does. “I need a second monitor” is a request: nothing is broken, something needs delivering. They get different timescales, and if you count them together you see a service desk that looks slow when it is mostly processing orders.
Priority comes from two questions: how many people or processes are affected (impact), and how quickly does it get worse (urgency). A three-by-three table is enough. What it prevents is priority being set by whoever shouts loudest.
Finally, agree when a ticket may be closed: only once the person who reported it confirms it works, or after a fixed period without a response. A ticket the engineer closes without asking produces good numbers and unhappy colleagues.
Change management without a change board
ITIL describes a change advisory board that meets every week. For a small or mid-sized business that is too heavy, but the idea behind it is worth a great deal: nobody changes a system other people depend on without a second person knowing.
- Standard changes are pre-approved because they happen often and carry little risk: creating a user, the monthly update in the agreed window. They sit on a list and go ahead without discussion.
- Normal changes get a short request: what changes, when, who will notice, and how do you reverse it. One person other than the one doing the work approves it.
- Emergency changes may go first and are recorded afterwards, within a day, with the reason.
The question that pays off most is the last one on the form: how do we roll it back? A change with no way back is a gamble, and you would rather know that beforehand.
Assets: knowing what you have and who has it
Without a list of devices, licences and accounts you cannot link an incident to a device, cannot see which laptops no longer receive updates, and cannot confirm that a leaver has handed everything back. It is the least visible of the three processes and the one the other two rely on.
Start small. Per device: what it is, the serial number, who or which location it was issued to, since when, and whether it is still under management. Per licence: who it is assigned to. How to set up such a list and keep it current is covered in our article on the IT asset register.
What you do not need is a full configuration management database recording every dependency between every server and every application. It is a large part of ITIL, and in a small environment keeping it up to date costs more than it returns.
Which parts of ITIL can wait
ITIL is written for organisations of every size, which means it contains a lot that solves no problem in a small environment. What you can safely postpone:
- Certification. An ITIL certificate is a personal qualification for someone who takes a course, and ISO/IEC 20000-1 is the standard against which an organisation has its IT service management certified. Neither is needed to work well; the second only becomes useful when a customer or tender asks for it.
- Problem management as a separate process. Keep it to one rule: if the same incident comes back three times, someone looks for the cause instead of fixing it again.
- An extensive service catalogue. One page listing what colleagues can request and how long it takes is enough.
- A separate process role for every component. In a small organisation the owner of all three processes is usually the same person. That is fine, as long as there is one name.
If your organisation falls under NIS2, documented incident handling does become an obligation. That does not change what you do, only that you must be able to show that you do it; the three processes above are where that starts. In the UK, the equivalent rules sit under the NIS Regulations, which apply to a narrower group of organisations.
Process first, then the ITSM tool
The most common mistake is starting with the software. An ITSM tool with dozens of modules invites you to configure all of them, and six months later there are forms with fields nobody fills in. Write the three processes down first, one page each, and then see which system makes that sequence easiest.
Three questions that sharpen the choice: do incidents, changes and assets sit in the same system and refer to each other; can you read the reporting without an explanation; and do you get the full history when you eventually switch?
If you outsource the service desk, the question is often already answered: you work in the provider’s system, and what matters is that you can see into it and that the data stays yours. What else to settle when outsourcing is in our article on outsourcing your service desk.
Where our work stops
What we do. We run service desk and support along these three processes: one way in, priority from impact and urgency, changes with a way back, and an asset register that belongs to every ticket. We are ISO 9001 and ISO/IEC 27001 certified, and that is how we work ourselves.
What we do not do. We are not a training provider or an examination body: we do not run ITIL courses and do not award certificates. Nor are we a certification body: we do not certify your organisation against ISO/IEC 20000-1 or any other standard, and we do not declare that you comply with NIS2. That is for an accredited certification body and, on the legal side, your own lawyer or adviser.
Questions we get about this
The five that come up most often.
What is ITSM?
ITSM, IT service management, is how an organisation delivers IT as a service: how incidents are reported and resolved, how changes are approved and how equipment and licences are tracked. It is about agreements and responsibilities, not about a particular software package.
What is the difference between ITSM and ITIL?
ITSM is the way of working; ITIL is a framework describing how to set that way of working up. You can do IT service management well without following ITIL to the letter, and many smaller organisations take only the parts they need.
Which ITSM processes does a small business need?
In practice three: incidents and requests, changes, and assets. They cover most of what goes wrong: tickets left untouched, changes that break something else, and devices or accounts nobody knows exist. Other components can be added as the organisation grows.
Do you need ITIL certification to do ITSM?
No. An ITIL certificate is a personal qualification after a course, and an organisation can be certified against ISO/IEC 20000-1, but neither is needed to work well. Certification only becomes useful when a customer, tender or regulator asks for it.
Can you certify us against ISO/IEC 20000 or ITIL?
No. We do not run ITIL courses, do not hold exams and are not a certification body, so we do not certify anyone against a standard. What we do is deliver service desk and support along documented processes for incidents, changes and assets.
Further reading
Where these processes land in practice.
Shall we put your three processes on one page?
Tell us how incidents, changes and equipment are handled today. We will show you where things get stuck and which agreement to add first, without a certification programme.
Practical IT knowledge in your inbox
New guides on management, security and the workplace, written by the people doing the work. No sales talk, and you can unsubscribe in one click.