Five quick wins to improve clinic workflow this quarter
- Ray Delany
- 11 hours ago
- 4 min read
A practice manager I know keeps a spreadsheet she has never told anyone about. Every morning she opens it, checks who is on, cross-references it against the appointment book, and works out by hand where the day is going to fall over. She built it because the systems she was given don't talk to each other, and she has kept it running for four years. When I asked her why she'd never asked for it to be fixed, she said, "Because this is faster than explaining it to anyone."
That's the thing about clinic workflow. The people closest to it have usually already solved the problem - quietly, off to the side, in a way that survives on their own memory and goodwill. The waste isn't hidden because nobody noticed. It's hidden because somebody competent absorbed it.
So this is not a post about buying something. It's about five changes you can make this quarter without a business case, a vendor, or a single line item. Every one of them works for the same reason: it makes the actual work visible before anyone tries to speed it up.
Start with the workflow that's frequent, frustrating and worth fixing
Don't start with the biggest problem. Start with the one that meets all three tests: it happens often, it annoys people, and getting it wrong actually matters. A monthly reconciliation that everyone hates but only takes twenty minutes fails the frequency test. The daily scramble over who's covering phones passes all three.
Pick one. Resist the urge to fix everything, because the point of a first win is that people see it land.
Win one: a short daily huddle
Ten minutes, standing up, before the doors open. Who's on, who's short, what's already looking difficult about today, and who needs to know. That's it.
The value isn't the information, most of it is knowable already. The value is that it's said out loud, in the same room, once, instead of discovered in fragments across the morning. My spreadsheet-keeping friend ran her first huddle reluctantly. Within a fortnight two other people knew what she knew, and the day stopped depending on her being at her desk by 7:45am.
Win two: triage the phone before it triages you
In a lot of NZ practices the phone is the least-managed, highest-pressure part of the whole operation. Every call arrives as if it's urgent, gets treated as if it's urgent, and eats an appointment slot as if it's urgent, when a good number could be handled by a nurse, a text back, or a booking the caller could have made themselves.
You don't need a new phone system to change this. You need an agreed set of rules for what gets booked, what gets a callback, and what a receptionist is allowed to resolve without asking. Write the rules down. Put them where the phones are. The tool can come later, once you know what you'd want it to do.
Win three: stop asking for the same information twice
Watch a new patient's details get keyed in. Then watch where those details go next, and how often someone types them again into a second system that should already have them or asks the patient a question the form on the wall just answered.
Duplicate entry is the most boring waste in healthcare and one of the most expensive, because it's paid for in the time of people you can't easily replace. You can't always eliminate it without changing systems. But you can nearly always reduce it by mapping where a single piece of information travels and cutting the steps that add nothing. Do that on paper, with the people who do the keying. They know exactly where the double handling is.
Win four: give one clerical task back to the system it belongs to
There's a category of work where a skilled person does something a machine did for someone else years ago. End-of-day reconciliation done by hand. A roster rebuilt in a spreadsheet every week. A recall list maintained from memory. Not because a better way doesn't exist, but because the manual version was quicker to set up and never got revisited.
Find one of these. Ask whether the capability to do it properly already exists somewhere you've already paid for - inside your PMS, inside a module nobody switched on. Often it does. This isn't buying anything. It's using what you have on purpose.
Win five: make the result visible
This is the one people skip, and it's the one that decides whether there's a win two, three, and four.
When you fix something, show it. Put the before and after on the wall. "Phone triage freed up nine appointment slots last week." "Time from enrolment to first booking halved." Make it concrete, make it recent, and attach it to the people who did it. Momentum in a clinic doesn't come from a strategy. It comes from a team seeing that the last small thing actually worked, and believing the next one will too.
Why these work - and why a tool alone wouldn't
None of these is Lean with a capital L. There's no kaizen board, no belt, no consultant. But they're all the same discipline underneath: understand how the work happens before you try to make it faster.
That order matters, and it's the whole argument. A digital enrolment form that halves time-to-booking is a genuinely good change, but only because someone first understood why booking was slow. Bolt the same form onto a process nobody has looked at and you'll automate the confusion. The tool will do faithfully, at speed, whatever mess it was pointed at.
That practice manager didn't need software. She needed ten minutes, permission, and someone to notice the spreadsheet. Most clinics are sitting on wins exactly like hers right now - not behind a purchase order, but behind the quiet competence of the person who's been absorbing the problem so nobody else has to.
The question worth asking this quarter isn't "what should we buy?" It's simpler than that. What have your best people already worked around and what would it take to fix it out loud?
CIO Studio helps New Zealand health and community organisations get the process right before they invest in digital and AI. For the full argument - and readiness self-check - download our eBook, Before You Invest in Digital and AI.
