Talk to a person who has staffed the reception side of a health service and ask what actually goes wrong on a rough day, and the answer is rarely dramatic. A login that fails for the entire morning shift. An appointment calendar that shows yesterday’s slots. A printer queue that swallowed the day’s paperwork. These are administrative and technical hiccups, and they are the territory of ITOps, the quiet discipline that keeps an organisation’s computers and software dependable.
This article is about administration and operations only. It says nothing about clinical care or treatment; it is a plain-language list of eight habits borrowed from the Ops Atlas description of ITOps and translated for the back office of a health-information service.
1. Give people one front door
The first practice in the Atlas list is the service desk. In everyday terms: one place where staff report trouble or ask for something, instead of shouting across the office or messaging whoever happens to look friendly. One door means every request leaves a trace, and traces are what later let you see patterns. A shared mailbox and a simple form are enough to begin with.
2. Treat “fix it now” and “stop it happening” as separate jobs
When the scheduling page stalls, the first goal is to get it working again. That is incident handling. The second goal, often skipped, is asking why it stalled and fixing the reason. The Atlas calls the second job problem management. A small team can keep the two apart with a simple rule: every incident that happens twice in a month earns a short written investigation with an owner.
3. Sort changes by risk instead of by habit
A new colour on the website footer and a new way of sending appointment reminders are not the same sort of change. The Atlas describes three types: pre-agreed routine changes, changes that get an assessment, and urgent changes done at once and reviewed later. If your team runs everything through the same slow path, people will start bypassing it; if everything runs through no path, an unlucky edit will eventually land on a busy morning. Writing down which changes belong in which lane takes about an hour and pays back for years.
4. Keep one calendar of planned work
Few things hurt more than two well-meaning people updating different systems on the same afternoon. A single shared calendar of planned changes, with blackout dates around the busiest administrative periods, such as the start of a month or a registration drive, lets everyone see collisions before they happen. The Atlas lists this forward schedule, with conflict detection, among the mechanics of a healthy change process.
5. Know what you have, but only what matters
Every service depends on a list of systems, accounts, domains and suppliers. The Atlas warns against a record that tries to hold everything, and notes that a sample health check of ten entries found two stale ones, one orphan and one duplicate. For an administrative team the lesson is modest: keep a one-page inventory of the systems whose failure would stop the front desk, name an owner for each and review the page quarterly. When the list is short, it actually gets read.
6. Make sure the alarms have an alarm
Monitoring is the practice of noticing trouble before users do. The less obvious rule from the Atlas is to treat the monitoring layer itself as a production dependency. When an alert system fails quietly, outages hide in plain sight. A monthly test, where someone triggers a harmless alert and confirms a human receives it, takes five minutes and prevents false comfort.
7. Write fix-times with a clock, not a feeling
“We’ll look at it soon” produces arguments. A short table of priorities with targets produces conversations. The Atlas shows a five-level example in which urgent items get four hours around the clock and lower levels get progressively longer, in some cases counting only business hours. It also shows why the clock rules matter: one sample ticket covered 92.5 elapsed hours but only 4.5 counted, because weekends and a holiday paused it. Should the office shut on particular days, put that in the rules.
To see the same write-it-down habit in a household setting, read this walkthrough of splitting household bills with a shared tracker and clear rules. The principle is identical: shared records and agreed rules prevent most disputes before they start.
8. Review what went wrong, calmly
After any significant failure or rushed change, spend fifteen minutes asking three questions. What happened? What did we assume? What one rule will we change? The Atlas recommends reviewing every failed or emergency change, and the habit matters more than the format. Keep the tone factual; the aim is a better process, not a guilty party.
A note on tools
Nothing has to be purchased before you begin. The Atlas makes two cautions here: ticketing software merely holds the notes of the work, and loading a full-weight process onto one small team with one service is a known misstep. If you eventually adopt software, pick something proportionate, and keep the habits above in place regardless.
Where the term comes from
For the complete definition and its neighbouring terms, see the Ops Atlas page that defines ITOps. It keeps three ideas apart: the staffed function, the working practices and the software used, and it notes that the vocabulary traces back to guidance first published in 1989 and to an international standard first issued in 2005.
For people who want to work in this field
Administrative teams sometimes discover that one of their own has a talent for this: the colleague who always knows why the shared drive broke. If that sounds like you, the roadmap for learning technical operations skills online and finding remote work is a reasonable next read, and an ITOps role is a natural stepping stone from a support position.
A one-week starting list
- Create the single mailbox or form for reporting problems.
- Start a log with date, symptom, fix and owner.
- Write the three change lanes on one page.
- Set up the shared calendar with blackout dates.
- List the five systems that would stop the front desk.
- Test one alert.
Eight habits, none of them exotic. Together they turn a reactive office into a steady one, which is exactly what administrative staff and the people they serve both want.