Anyone who coordinates cross-border care logistics depends on technology that rarely gets inspected: enquiry forms, document portals, appointment calendars, messaging tools and payment pages. When one of these quietly fails, the consequence is a missed message or a lost attachment, not just an inconvenience. This checklist applies ITOps, the discipline of keeping IT services running, as an internal audit for a medical-tourism coordination team. It covers technology operations only and says nothing about clinical quality or any provider.
How to use this audit
Score each item red, amber or green and note an owner and a date. Do not try to fix everything in a week. The structure below follows the scope described in the Atlas entry on ITOps, which lists the service desk, incidents, requests, problems, change enablement, monitoring, the configuration register and service levels as the working core. Eight sections, eight scores.
1. Service desk: one front door
- Is there a single channel where internal staff report technology faults, and does each report get a reference number?
- Is there a named person or rota behind it on weekends and across time zones?
- Do shadow spreadsheets exist? The Atlas lists these private lists as a warning sign that the official process is too awkward.
2. Incidents versus requests
- Are breakages and routine requests recorded as different types?
- Is there a written priority rule, with impact and urgency defined in plain words?
- Can you show the last five incidents, each with start time, fix time and owner?
3. Problems: repeat causes
- Do you review tickets monthly for lookalikes?
- Is there at least one open problem record with an owner and a due date?
- When a workaround is used, is it labelled as temporary?
4. Change enablement
The Atlas distinguishes three change types. Use them as an audit lens.
- Is there a short list of pre-approved, low-risk changes? If every edit to a web page needs a meeting, the process will be bypassed.
- Are higher-risk changes assessed with a written score, and is there one calendar that shows overlaps?
- Are urgent fixes allowed but reviewed afterwards, with the review on record?
- Are records created by automated deployment tools, but approved by people rather than by the tool itself?
For teams with European operations, the Atlas also notes that the EU digital resilience law, applicable since January 2025, wants approvers kept independent of change authors and wants urgent changes ratified once they are in place. Simplified-framework firms have lighter duties. Check with your compliance adviser which regime applies; this checklist does not give legal advice.
5. Monitoring and events
- Which services are watched: the enquiry form, the document upload, the calendar, the payment page?
- Who receives alerts, and who is the backup?
- Is the alerting itself tested on a schedule? The Atlas warns that treating the event layer as non-production lets silent failures hide outages.
- Are repeated alerts grouped, so staff see one incident rather than thirteen messages?
That last question echoes an Atlas exercise in which a stack of thirteen raw alerts ended up as only two incidents inside one time window. Grouping is a design decision with an owner, not an accident.
6. The configuration register
- Do you list every system that holds enquiry or document data, who administers it and where it is hosted?
- Are entries reviewed? The Atlas health-check example flags records not updated for more than sixty days, orphans and duplicate serial numbers.
- Is the register limited to what decisions need rather than every object?
Our companion piece on connected IT operations frameworks and learning paths gives wider context for how these registers connect to other disciplines.
7. Service levels and clocks
- Are response targets written down for each priority level?
- Does each target state its clock: round the clock, or business hours with pauses for weekends and holidays?
- Are internal handoff times and supplier commitments added up before a figure is promised to users?
The Atlas sample splits one eight-hour promise across a desk, a network team and a carrier, plus a buffer. If your promise has no buffer, your real compliance will disappoint.
8. Fit of process to team size
- Is the process proportionate? Heavyweight machinery imposed on a team running one service is a named anti-pattern.
- Does it avoid adding new steps after every failure, a pitfall the Atlas links to research on change delivery?
- Is a certification actually required by a customer, or is a free, lightweight standard sufficient?
Keep proof for each colour
A score means little without something to back it. For every section, attach one artefact: a screenshot of the ticket queue, the change calendar for the last month, the alert routing list, the register export. Keep them in a single folder with the date in its name. When the next review comes round, you can compare like with like, and anyone joining the team can see how the previous scores were reached without asking around.
Reading your scores
Count reds. Three or more in sections 1 to 3 suggest the basics are missing, so begin there. Reds concentrated in 4 and 5 mean you can recover from incidents but are creating them. Mostly amber with scattered green is normal for a small team, and it is a fair place to start from.
A useful habit is to repeat the audit every quarter and compare, much like checking a fleet of machines. Our overview of how robotic fleets are looked after day to day shows the same routine inspection mindset in another setting.
- Pick the two lowest-scoring sections.
- Assign an owner and a date to one action in each.
- Re-score in three months and record the change.
The goal of the audit is a dull, reliable layer underneath your coordination work, so that people planning care journeys can focus on the people rather than the plumbing.