A managed service provider sells response time. The SLA clock starts when a ticket lands, not when the new analyst finds the right queue. So an engineer's first week in a service desk, NOC or data hall is about removing delay. Login, tools, who to ask and what not to touch.
Access is the first deliverable
Before day 1, ask the team lead for the access list. A typical list runs as follows. The ITSM login and the SNMP monitoring console. The RMM agent console and a Jira or Freshdesk board. A VPN profile with MFA token. The knowledge base, the password vault in a read-only role and the shift roster. If any is missing at 10 am on day 1, log it as a ticket against yourself.
- Request accounts in the joiner's own name; shared logins break the audit trail.
- Give least-privilege roles for week one, then widen after the day 5 review.
- Record each grant with date, approver and system in one register.
- Withdraw a temporary grant on the date written next to it, not whenever someone remembers.
The wider hiring picture for this group is on the hub page for IT infrastructure and managed services. The recruiting side for support firms is in recruitment for IT support firms.
The week as a sheet
Name: ________ Tower: service desk / network / cloud / data centre Shift: ________
DAY 1: SET-UP (team lead and HR)
Laptop built from the golden image, asset tag recorded, BitLocker key escrowed, Intune enrolment confirmed.
Accounts live: ITSM board, Zabbix or Grafana dashboards, VPN with MFA, vault (read only), Active Directory group, mail, chat.
Shift pattern, weekly off, week-off swap rule, night-shift cab rule and on-call allowance basis stated in writing.
Client list for the tower, with the SLA targets and the account manager's name.
DAY 2: READ AND WATCH (buddy engineer)
Reads the top ten runbooks aloud: DNS and DHCP faults, VLAN mismatch, firewall rule request, disk full on a hypervisor host.
Sits beside the buddy through one P1 or P2 incident from alert to closure note, including the bridge-call etiquette and the customer-update cadence.
Shown the ticket hygiene rules: category, priority, root cause note, customer update time.
DAY 3: SUPERVISED QUEUE (shift lead)
Takes L1 tickets with the buddy doing a line-by-line review of every closure.
Practises a handover note: open tickets, pending vendor RMAs, maintenance windows tonight.
DAY 4: CHANGE AND SECURITY (tower lead)
Sits through a change window, the rollback plan and the CAB summary.
Reads the data handling rules for client logs and screenshots.
Signs the non-disclosure undertaking and the acceptable-use note.
DAY 5: REVIEW (shift lead and joiner)
Joiner explains three tickets he handled and one he escalated, with a hand-over note for each.
Lead records: ready for queue duty / needs a second week / better fit in another tower.
Next fortnight's pager exposure agreed in writing.
Signed: joiner ____ lead ____ HR ____Who ticks which line
- 1
Team lead owns access and the verdict
If the lead cannot spend ten minutes on day 5, move the review rather than skip it.
- 2
Buddy owns the learning lines
Choose a buddy who writes clear root-cause lines in the closure note, not just the fastest closer. Speed copied from a careless closer is the usual source of reopened tickets.
- 3
HR owns the paper lines
Offer letter terms, confidentiality undertaking, allowance basis and the wage date. They are short but they are the lines that become disputes.
- 4
Joiner owns the questions
Tell him he is expected to list at least five things nobody told him. Answers feed the next joiner's sheet.
Where week one usually goes wrong
Three faults repeat. First, the engineer sits idle behind an unactioned access request, then is blamed for slow ticket pickup. Second, the buddy is a night-shift engineer, so the joiner never sees a daytime change window or a client call. Third, nobody tells him the difference between priority and severity, and he marks a printer jam as P1 because the user was senior.
- Fix the first with an access ticket opened a week before the joining date.
- Fix the second by matching buddy shift to joiner shift for at least four days.
- Fix the third with a one-page priority matrix pinned beside the ITSM screen. Use real examples: BGP session down at a branch, mail relay queue stuck, locked Active Directory account, failed patch on a hypervisor host, a syslog flood from one switch.
Notice what these have in common. None needs a new tool. Each needs one person to own a date. That is why the sheet names a person on every group of lines and not a department.
Client-site and data centre variants
An engineer posted to a client site needs three extra lines. The client's visitor and badge process. The site contact who signs his attendance. A named person at your office who phones him daily through the first week. Cut off in week one, he will form his view of the employer in the client's canteen.
In a data hall, add the rack unit and patch panel labelling rules. Add the cold-aisle and hot-aisle walk and the PDU phase check. Add the no-food rule and the cross-connect request form. Add the escalation for a UPS or chiller alarm, plus a hands-on session with the SFP and fibre cleaning kit. Do not allow tailgating through the mantrap even once, in training or otherwise.
Compare with attendance and shift planning for networking firms once the new engineer reaches the roster. A follow-up for month two is in the employee onboarding process page.
