06Practice
IT Consulting & Managed Services
For companies whose software matters but whose core business is not software. We assess what you have, write the plan, then run the parts you would rather not staff internally — applications, cloud, endpoints, vendors and the on-call rota. Named engineers, published response times, monthly reporting in plain English.
- Typical duration
- 4–6 weeks to start · Ongoing
- Engagement models
- Advisory · Managed operations · Transition and exit
- Practice lead
- Named at proposal stage

01The problem
Software you depend on, without a team that owns it
Plenty of good companies reach forty million in revenue with an IT function that is one overloaded manager, two contractors and a filing cabinet of vendor contracts. It works until it does not: a key integration breaks during month-end close, the person who understood it is on leave, and the vendor agreement turns out to cover response but not resolution.
The fix is rarely a bigger tool budget. It is knowing what you run, who owns each piece, what it costs, and what happens at 3am. We start by writing that down — a genuine inventory and an honest assessment — and only then discuss what we should operate on your behalf versus what you should hire for directly. Sometimes the recommendation is to hire.
02What GatiQ delivers
The work itself
Every item below is something you can inspect, run or read at the end of the engagement.
- 01
An assessment that names things plainly
Four to six weeks of interviews, architecture review, spend analysis and vendor scrutiny, ending in a document your board can read. Risks are ranked by likelihood and business impact, each with a cost to fix and a cost of doing nothing.
- 02
A roadmap with sequencing and money attached
Twelve months, quarter by quarter, with dependencies mapped and decision points flagged. We revisit it every quarter against what actually happened, because a roadmap nobody revises is a wish list with dates on it.
- 03
Operations with named people
You get the names of the engineers who hold your pager, not a queue. Sev-1 acknowledged within thirty minutes around the clock, Sev-2 within two business hours, and a status page your internal users can check themselves.
- 04
Maintenance on a calendar
Patch windows, dependency upgrades, certificate renewals, access reviews and backup restore tests scheduled in advance and reported afterwards. The unglamorous work that prevents the incidents nobody notices because they did not happen.
- 05
Reporting a non-technical executive can act on
One page a month: availability against target, incidents and their causes, work completed, spend against budget, and the two or three decisions we need from you. Written in English, not in ticket IDs.
03Engagement model
Three ways to buy it
- Advisory
- Fractional CTO or architecture advisory, typically two to four days a month, for leadership teams making platform decisions without a senior technologist in the room.
- Managed operations
- Twelve-month agreement covering defined applications and infrastructure, priced per service rather than per ticket, so nobody is rewarded for closing things quickly.
- Transition and exit
- Taking over from an incumbent provider, or handing to your new internal team. Documented, rehearsed, time-boxed to sixty days, with knowledge transfer sessions recorded.
04Typical timeline
What the calendar looks like
- 01
Assessment
4–6 weeksInventory, interviews, spend and vendor analysis, ranked risk register.
- 02
Stabilise
4–8 weeksClose the top risks, set up monitoring, document runbooks, agree targets.
- 03
Transition
3–6 weeksShadowing, credential handover, rehearsed escalation, go-live on the pager.
- 04
Run and review
OngoingMonthly service review, quarterly roadmap revision, annual re-assessment.
Ranges, not dates. We narrow them at the end of discovery and tell you the same week if they move.
05Technologies
What we reach for
Defaults, not rules. If your team already runs something else competently, that is usually the better choice.
- Service management
- Jira Service Management, PagerDuty, Statuspage
- Endpoint and identity
- Intune, Jamf, Okta, Tailscale
- Monitoring
- Datadog, Grafana Cloud, Checkly, Uptime Kuma
- Automation
- Terraform, Ansible, GitHub Actions, n8n
- Backup and continuity
- Veeam, AWS Backup, Restic