ITSM · IT Help Desk
IT Service Management and IT Help Desk
What is ITSM?
- Start by routing every request through one front door
- Separate incidents, requests, problems, and changes, each with its own procedure
- The goal is to eliminate repeat tickets
What changes with ITSM is that asking 'someone you know' over messenger goes away. When requests like a slow laptop, a locked account, or a new hire's email setup travel through personal messages and hallway conversations, nothing records who agreed to do what by when. Once requests are gathered at a single help desk, intake, handling, and completion times are recorded, backlogs become visible, and repeat requests show up too. The essence of ITSM is that IT's work is redefined from 'odd jobs' to 'service'.
ITSM sorts requests by type and handles each differently. Incidents are about restoring something that has stopped, fast. Service requests, such as account provisioning or equipment issuance, follow a set procedure. Problem management finds the root cause of recurring incidents. Change management reviews and approves system changes. The best-known framework for these distinctions is ITIL, which originated in the UK and now continues as ITIL 4. The basic help desk metrics are ticket volume, first response time, resolution time, and first-contact resolution rate. An SLA that commits service levels between departments is what makes 'by when' clear.
Training and HR teams are major help desk customers. New-hire account and equipment requests, training room equipment setup, LMS login problems, and year-end review system errors all come through this channel. Define these requests in advance as service catalog items such as “New-hire setup” and “Training equipment loan” and you no longer have to explain them each time, and each gets a set turnaround time. The core skill of running a help desk is routing repeat questions to an FAQ and self-service guides to cut ticket volume.
The most common failure in practice is adopting a ticket system while still taking requests over chat. Once people bypass the front door because something is urgent, the system holds only half the picture, the statistics become unreliable, and eventually nobody looks at them. The opposite happens too: every request demands approvals and forms, so a password reset takes two days. Another common case is outsourcing the help desk, then hearing nothing but “let me check with headquarters” because the agent doesn’t know the company. The goal is not fewer tickets but making sure the same ticket doesn’t keep coming back.
Seolyeong ParkGenerative AI and AX training expert with 17 years of hands-on experience and 200+ sessions, making AI usable right away even for non-specialists
Seungtae ParkPractical AX instructor who teaches by building AI business development and marketing automation hands-onIt began with ITIL (IT Infrastructure Library), a set of IT operations best practices compiled by a UK government agency in the late 1980s. ITIL's framework for handling incidents, requests, problems, and changes separately spread to the private sector and became the field known as 'IT service management.' It was revised as ITIL 4 in 2019 and is still in use.
Consider a hypothetical case. Company A, a construction materials firm with 540 employees, had a 4-person IT team that took requests by messenger and phone. Last January, account and laptop setup for 23 new hires was only half done by their first day, and the HR lead filed a formal complaint. When the year-end performance review system went down, no one had even recorded who reported it first.
- When introducing the ticket system, the IT lead held one rule for three months: chat requests get a ticket created for them, but are handled in ticket order. Of 412 tickets received in the first month, 38% had bypassed the system through chat.
- In consultation with the HR lead, the team added two items to the service catalog, 'New-hire setup (submit 5 business days before the start date, completed by the first day)' and 'Training room equipment loan (submit 3 business days in advance),' and fixed those deadlines as SLAs.
- After classifying three months of tickets, the team found that password resets made up 27% of the total. Kim, an IT team associate, said, 'This isn't something a person should be doing,' and built a self-service reset page and announced it to staff.
- When the performance review system failed twice in November, the team grouped the failures into a problem-management ticket and traced the cause. Simultaneous logins on the review deadline turned out to be the culprit, so HR changed the schedule to stagger deadlines by two days per department.
- Six months later, messenger workarounds fell from 38% to 6% and monthly tickets dropped from 412 to 290. This January, setup for 31 new hires was 100% complete by their first day.