In this article
Short answer: In an occupational health service, administration rarely sits in the clinical step itself. It sits in the coordination around it: the order, the rebooking, the reminder, the sick-leave notice and the month-end compilation. We deliberately publish no general time savings, because the figure depends entirely on how your operation looks. But it can be measured in a week with touch counting, and that measurement decides what is worth automating in your case.
Administration eats profitability step by step: a form keyed in by hand, a rebooking that takes three phone calls, a follow-up nobody is watching. Together it is a cost few people measure and everyone carries. This article is about measuring it before you buy anything, and about what automation actually removes and what it merely moves.
Why we will not give you a benchmark
Numbers circulate in this industry: "save 20 minutes per case", "free up a third of a full-time position". They sound concrete and are in practice invented. The spread between two occupational health services of the same size is larger than the claimed saving, because it depends on contract complexity, how much is already digital and how many exceptions are handled manually.
A number you have not calculated yourselves will not survive a budget discussion either. Count it yourselves instead. It takes less time than reading a quote.
Touch counting: measure it in a week
A touch is every time a person on your side does something with a case: reads an e-mail, makes a call, keys in a detail, looks up a contract, updates a status.
Do this for one week:
- Pick three flows: a typical order, a rebooking and a sick-leave notice.
- Have staff make a tally mark at every touch, on paper. Not timing, only counts. Timing stresses people and makes the data unreliable.
- Take the median per flow, not the average. A single catastrophic case otherwise ruins the picture.
- Multiply by the yearly volume, and set an average cost per minute for a touch. If you are unsure, time ten touches with a clock and use the median.
The result will be rough, and that is the point. What you are looking for is not an exact cost but which flow has the most touches. That is where automation pays off first, and it is rarely the flow you guessed.
Where touches tend to gather
The order
The classic: an e-mail from a manager who does not know what the contract includes, a reply from you, a clarification, a confirmation. Four touches for something that could have been zero. A contract-driven ordering portal removes them by letting the client see its own offering, but only if the catalogue really mirrors the contract. If it does not, you swap the e-mails for rejected orders.
The rebooking
A new booking is cheap, a rebooking is expensive. A cancellation by phone the day before requires someone to release the slot, find a new one, contact the person and update the calendar. The fewer steps it takes to free the slot, the more touches are saved, and the sooner someone else can take it.
Reminders and no-shows
A no-show costs the whole slot plus the work of rebooking. Automatic reminders are the single cheapest intervention, because they require no change in how you work.
Sick-leave and wellness notices
The touches here are spread out and therefore easy to underestimate: the notice is received, the manager has to be informed, the day-8 certificate has to be tracked, the day-30 plan has to be followed up. No single item takes long, but they recur per case and per day. We go through the flow in sick-leave reporting that works.
The month-end close
If the month-end close means someone compiles orders from several sources, that is a flow that should be close to zero touches. Portway does not produce invoices or an export for them, so this step stays in your own system; what you can ask of any tool is that orders and bookings are registered once, in one place.
What automation removes, and what it only moves
The uncomfortable part of the calculation: an automated process does not remove the work, it removes the normal case. The exceptions remain, and they become your new working day.
That means two things when you set requirements:
- Exception handling is a main function, not a detail. What happens when the client orders something that is not included, when a person does not show up, when a contract ends in the middle of a case? A system with only one happy path makes the exceptions more expensive than before.
- Count touches in the new flow too. An automation that sends an e-mail someone still has to read and acknowledge has moved the work, not removed it.
Never automate something you cannot prove happened
This is the lesson we ourselves have paid most for, and it is worth more than all the time savings above.
When a step is manual, it is noticed immediately if it is not done: someone is standing there with a case. When the same step is automated, that signal disappears. If the sending breaks, everything keeps looking normal: no error messages, no complaints, just reminders that never arrive. A notification function can look entirely correct in the code and in practice never send a single message. Nobody notices, because nobody misses an e-mail they never knew was supposed to arrive.
The conclusion is simple to state and easy to skip: every automated message should leave a trace. A timestamp per case showing that the reminder went out, to whom and when. Then "has the client received the reminder?" becomes a question someone can answer in the system rather than a guess, and a broken automation becomes visible the same day instead of at the next quarterly meeting.
Ask every vendor you evaluate: where do I see that this reminder was actually sent? If they cannot show it in the interface, the automation is a hope.
The order that tends to pay off
- Reminders: cheapest, require no change in working method, effective immediately.
- Faster rebooking: fewer steps in the most expensive single step.
- Contract-driven ordering: biggest effect, but requires the contracts to be structured first, and that is where the work lies.
- Month-end compilation: easier to compile once step 3 is in place, because orders are registered once.
Rehabilitation tracking and report packages come last, not because they lack value but because they are hardest to get right and easiest to get wrong.
How we do it in Portway
Portway brings orders, booking and sick and recovery reporting into one place. Each client company has its own catalogue and places its own orders, and every order has a status (New, In progress, Done or Cancelled) that both you and the client can see. Reminders before booked visits go out by e-mail 24 hours and 1 hour before, and in Sjuk & Frisk the follow-ups are created automatically, so day 30 does not depend on someone remembering it.
If you want to know where your touches sit, you can book a walkthrough; bring the week's tally sheet.
Frequently asked questions
How much time do you save by digitalising occupational health administration?
It depends on how much is already digital and how many exceptions you handle by hand. We deliberately publish no general figure. A week of touch counting gives an answer that applies to your operation and holds up in a budget discussion.
What should be automated first?
Reminders before visits. They require no change in working method, reduce no-shows and can be measured immediately.
Doesn't it become impersonal when the client does it themselves?
What disappears is the coordination calls, not the clinical ones. The experience is that time shifts from booking and rebooking to answering real questions, which is the contact the client actually values.
We already have a system but still do a lot manually. Do we need to switch?
Not necessarily. Count the touches first. If they sit in exceptions the system does not support, it is often configuration; if they sit in the normal flow, it is the system. Five signs you have outgrown your systems helps tell the two apart, and the switching checklist takes over if the answer is to switch.