Imagine your first Monday as the new systems coordinator at a hospital group. Someone hands you a laptop, a badge and a spreadsheet that has seven thousand rows. Each row is a weakness in some piece of software that the organisation runs, and every one of them has a number next to it. Your manager says, kindly, “Start with the highest numbers.” By Thursday you will probably suspect that this advice, while friendly, cannot be the whole plan. This article explains why, and how a way of working called VulnerabilityOps gives a newcomer something firmer to stand on.
No security degree is assumed. If you can follow a hospital discharge checklist, you can follow this. The focus is on non-clinical operations, the paperwork and routines that keep computers dependable, and nothing below is medical guidance.
What VulnerabilityOps means for someone new to the job
Picture a conveyor belt next to a heap, and you have the core of it. A heap is what you get from a periodic scan: a pile of findings that grows until someone declares a clean-up week. A conveyor belt is what you get when each finding is sorted on arrival, handed to a named person with a due date, confirmed as truly fixed, and counted. The Ops Atlas page that introduces VulnerabilityOps describes it in similar terms, as an operating discipline for the whole stream of findings rather than a product you buy, and it points out that the Atlas does not claim any official standard behind the name.
Why should this matter for your career? Because the people who get trusted with more responsibility in hospital IT are rarely the ones with the biggest dashboards. They are the ones who can say, calmly, who owns each open item, when it is due and what proof will close it.
Step one: sort before you chase
Back to that spreadsheet. The Atlas argues that a severity number tells you how bad something could be in theory, and that this is a different question from whether it is being exploited right now. It suggests reading three kinds of signal in a fixed order. Evidence that attackers are actively using a flaw comes first. A probability estimate comes second. The severity score comes last. Think of it like a triage desk: a patient already in distress is seen before a patient who has a worrying history on paper.
For you, this means adding two columns to the sheet. One records whether the system faces the public internet, like a patient-booking page or a staff login portal. The other records whether the flaw appears on the public catalogue of actively used weaknesses. Those two columns alone will reorder the sheet more usefully than any score could.
Step two: a name and a due date for every row
An orphaned finding is like a ward bed with no assigned nurse. Everyone assumes someone else has it. The Atlas calls for a dated owner queue, which means a list where each item names a team and carries a deadline that the team agreed to. As a newcomer you will spend much of your first month finding out who really manages each system, and that detective work is valuable.
One rule from the Atlas is worth memorising because it protects you from well-meaning shortcuts: analysts may raise an item to a more urgent tier but should not lower one. If you suspect something is less urgent than it looks, you raise a question with the owner, you do not quietly demote it.
A practical habit helps here: write the hand-off in one short paragraph. Name the system, say why it matters to the business in a single line, state the due date, and describe what proof you will accept. Owners rarely push back on a request that is that specific, and the paragraph doubles as your record when someone asks months later why a particular machine was handled the way it was. Keep a copy beside the queue so that nothing depends on anybody’s memory of a corridor conversation.
Step three: closed means proven
This is the part most beginners get wrong, and it is easy to see why. A colleague writes “done” on the ticket and the row turns green. The Atlas is blunt that ticket status is not evidence. Closure should come from a fresh scan that logged in properly, a test of the setting, or a rebuilt image with a new fingerprint. Another route to closure is a dispute that was properly reviewed and accepted.
There is a quiet trap here. If a scanner loses its login to a machine, it may stop reporting problems on that machine, and the chart improves. Nothing was fixed; the scanner simply went blind. The Atlas describes this as green on the dashboard that comes from credential failures. A coordinator who checks that every system was actually inspected before celebrating is worth a great deal.
- Ask which assets the last scan could not log in to.
- Ask what evidence closed each item, not who clicked the button.
- Ask whether a rescan could reopen the same item with a fresh clock, which would make late fixes look punctual.
A first quarter you can actually follow
The Atlas includes a thirty, sixty and ninety day outline that suits a new hire well. In the first month the goals are modest: find the gaps in your asset list, tag which systems face outward, start a daily check against the public catalogue of exploited flaws, and keep a baseline to compare against later. Month two is when you draft a first policy, agree a hand-off record so an owner knows exactly what is being asked, switch closure over to rescans, and open a door for exceptions. In the third you pick one high-volume class of findings to automate, hold the first monthly review of how the flow is behaving, and set an alert for coverage slipping.
Notice what is absent: heroics. Nobody is asked to read seven thousand rows in one sitting.
Reading the numbers like a colleague, not a spectator
Once the queue exists you will want to know whether it is healthy. The Atlas suggests a simple ratio for each tier: new arrivals divided by confirmed closures over the same period. If the figure stays below one, you are draining the queue. If it sits above one, the queue is growing, whatever the headline chart says. A small worked example in the Atlas shows a middle tier running just under one and a low tier running above it, which meant the low-tier backlog would take longer to clear than its sixty-day promise allowed. A newcomer who can explain that in a team meeting will be remembered.
There is a companion warning. The average age of old items can fall even as the pile grows, because lots of new items dilute the old. So quote the age and the ratio together, never the age alone.
More reading from this blog
Should this is your first brush with operations vocabulary, a general overview of building dependable business systems with technology services gives useful background, and a short guide on why teams care about security within software work covers the neighbouring territory that VulnerabilityOps deliberately leaves to others.
Questions a new coordinator can ask without embarrassment
Newcomers often hold back because every question seems to expose how little they know. These are safe, and each one shows seniors that you are thinking about flow rather than panic: which systems are untagged, and what happens to an untagged system by default; who approved the last exception and when does it expire; how many items were reopened after being marked done last quarter. Good teams enjoy these questions because they surface drift early.
The career angle, in one paragraph
Hospitals will always need people who can keep patient-facing and back-office systems steady. As the volume of public findings keeps climbing, the useful skill is less “I know every flaw” and more “I can run a calm, evidence-based queue”. That skill travels well between employers. If you build it in your first year, you will be the person others ask when a headline-grabbing weakness appears and a manager wants to know, within the day, which of our systems carry it.
Start small. Add the two columns. Name the owners. Ask what proof closes each row. By Friday your Monday spreadsheet will look different, and you will know why.
