Cliniko Already Knows Who to Ring. Should a Machine Do the Ringing?

The cheapest appointment you will ever book is with someone who has already been to your clinic. They know where you are, they have paid you before, and you do not have to buy a click to reach them. Cliniko is sitting on that entire list right now, sorted by exactly the thing that matters, which is who has not been in for a while. Almost nobody works it. The interesting question is not how to find them. It is who does the phoning.

Getting the list out

This part is genuinely ten minutes and needs no software beyond what you pay for.

In Cliniko, use the patient list and filter on the date last seen. Take everyone whose last appointment was more than nine months ago. Nine is a deliberate choice for routine care: it is comfortably past a six-weekly rhythm and past an annual check, so you are not chasing someone who is simply due next month, but it is recent enough that they still think of your clinic as theirs.

Then sort by most recent first. This is the step people skip and it matters more than the filter. Someone who lapsed ten months ago remembers you. Someone who lapsed four years ago has moved, changed clinic, or forgotten you exist. Working the list newest-first means your best conversations happen first, which also means you find out quickly whether this is worth continuing.

What to strip out before you ring anyone. Patients discharged deliberately, anyone with a note explaining why they stopped, and anyone who has asked not to be contacted. This is not optional. A recall list is a marketing list, and the rules about consent and about contacting people on health matters apply to it fully.

The real question: who makes the call?

Here the advice you will read online goes badly wrong. The obvious move, once you have a list and access to automation, is to have the machine do all of it: text everyone, follow up automatically, book them straight into the diary, hands off. It is the version that demos best. It is not the version we would recommend for a clinic, and we sell the automation.

Option one: do the whole thing by hand. Print the list, ring twenty people a week yourself or have reception do it between patients. Where this genuinely wins: it is free, it starts this afternoon, and the conversations are excellent because the person ringing knows the patient. Where it falls down: it is the first thing dropped in a busy week. Most clinics do it once, get a good result, feel encouraged, and never do it again. A recall that happens twice a year is barely a system at all.

Option two: automate the whole sequence. A tool texts every lapsed patient, chases the non-repliers, and books the ones who respond into the diary without anyone touching it. Where this genuinely wins: it is consistent, it scales to a list of thousands, and for a clinic with several practitioners and a long back catalogue it will surface bookings that no human was ever going to chase. Where it falls down, and this is the important part: a lapsed patient with a foot problem is a clinical conversation, not a retail one. "Are you still having trouble with that toe?" is a question with a medical answer, and the machine cannot hear the hesitation that tells your practitioner this person needs to be seen in a fortnight. There is a tone problem too. Chasing a health issue with automated persistence reads as pushy in a way that chasing an abandoned shopping basket does not.

Option three: the machine builds the list, a human makes the call. Automation does what it is genuinely better at, which is remembering. It runs the filter every week, applies your exclusions, ranks by who is most worth ringing, and puts a short list in front of a person. Twelve names, with the last appointment and treatment beside each. A human rings them. Where it falls down: it still needs someone to actually make the calls, so it does not solve a clinic with no time at all, and building it is real work.

The recommendation, plainly: option three. The value of automation here is not that it can talk to patients. It is that it never forgets to look. The part that keeps failing in real clinics is not the phoning, it is that nobody remembers to pull the list. Fix the remembering with software and leave the talking to a person who can respond to what they hear. The £2 you save by automating a call is not worth the regular patient who feels processed.

Where we would change that advice: multi-practitioner clinics with thousands of dormant records genuinely cannot ring everyone, and there a well-written first text that simply opens the door is better than silence. Even then, we would automate the opening and hand every reply to a human.

What to actually say

Keep it short, make it about them, and give them a reason that is true. Something close to: "Hello, it's the clinic. You were in last October about your nails and we've a few appointments free this week, so I thought I'd check whether you're due a look."

That works because it proves you know who they are, gives a genuine reason for ringing today, and asks a question that is easy to answer either way. Do not open with an offer or a discount. It makes a routine health check sound like a sale, and it teaches your patients to wait for the next one.

Do this today, for nothing

Pull the filter now: last seen more than nine months ago, sorted newest first. Look at how long the list is, because that number alone is usually a surprise worth having.

Then ring the top twenty this week. Not a campaign, not a system, just twenty calls. You will book some appointments, and more usefully you will learn what your lapsed patients actually say, which is the thing any future system has to be built around.

Do the twenty, then bring us what happened and how long the full list was. Twenty minutes and we will work out whether a weekly list in front of a human is worth building for your clinic, or whether ringing them yourself each month is genuinely the better answer.