Contact the warmest, most clinically relevant groups first: patients who started a plan and stopped, then patients overdue for a review the doctor asked for, then patients examined and quoted who never started, then recent no-shows. Old enquiries and happy past patients with nothing due come last. Within each group, start with the most recent. Size each group against your team's real calling hours before you begin.
Picture a urology clinic that has just finished cleaning its records. About 7,000 contactable people. The owner's plan is to "message everyone this week". His front desk has two people and one of them is already on the phone most of the day.
Messaging everyone at once is the classic mistake I describe in Good Doctors Go Broke. The same message reaches a man who finished stone treatment last month and an enquiry from four years ago. It helps neither. And when 200 people reply on the same day, nobody can call them back in time, so the warmest replies go cold.
The better question is not "how do we reach everyone?" It is "who do we reach first, and how many can we actually handle?"
What makes one group warmer than another?
Three things decide the order:
- How close they came to treatment. Someone who started a plan is closer than someone who only enquired.
- Whether there is a genuine care reason to call. An overdue review the doctor asked for is a real reason. "We haven't seen you in a while" is not much of one.
- How recent the contact was. A quote from three months ago is warmer than one from three years ago.
That gives you a natural ranking.
How should you rank the groups?
Here is the order I use, with example counts from a clinic like the urology one above. Your counts will be different.
| Rank | Group | Example count | Why this rank | Who contacts |
|---|---|---|---|---|
| 1 | Started a plan and stopped | 180 | Warmest people in the database. Small reasons stop them, and one call often solves it | Coordinator, by phone |
| 2 | Overdue for a review the doctor set | 650 | A real care reason, already agreed with the doctor | Recall owner, WhatsApp then call |
| 3 | Examined and quoted, never started, last 12 months | 420 | Already trusted the doctor enough to be examined | Coordinator, by phone |
| 4 | Booked and did not show, last 6 months | 300 | Showed intent, something got in the way | Front desk, call then WhatsApp |
| 5 | Enquired, never booked, last 12 months | 1,900 | Some intent, but no relationship yet | WhatsApp first, call those who reply |
| 6 | Finished care, nothing due | 2,100 | Happy, but no care reason to call | Occasional useful education only |
| 7 | Enquiries older than 12 months | 1,450 | Coldest, lowest priority | Leave until the rest are done |
Notice that ranks 1 to 4 add up to about 1,550 people. That is where most of the value sits, and it is less than a quarter of the list.
How do you sort inside each group?
Within a group, go by recency first. A decent CRM sorts this in seconds. Patients who stopped a plan two months ago before those who stopped two years ago.
Then by how much care is left. A patient with four sessions remaining or an important review overdue comes before someone with one minor visit left.
Then by any flag the doctor has added. If the doctor has written "please make sure this patient comes back", that patient goes to the top, and may deserve a call from the doctor personally.
How many calls can your team really make?
This is the step most clinics skip, and it is why drives collapse in week two. Do the maths before you start. Example figures:
- One coordinator can give 2 hours a day to reactivation calls.
- A useful call, including dialling, waiting and logging, takes about 6 minutes. So around 20 dial attempts a day.
- Maybe half are answered on the first try. That is about 10 real conversations a day, or 50 a week over five working days.
Now size the top groups against that:
- Unfinished plans, 180 people. First attempts take about 9 working days. With second and third attempts for those who did not answer, call it three weeks.
- Overdue reviews, 650 people. WhatsApp goes first, and calls only go to those who don't reply. If 40% reply to the message, about 390 still need calls. At 50 conversations a week, that is close to two months for one person.
- Quoted not started, 420 people. Roughly another eight weeks of calling.
So the honest plan is: one person needs four to five months for ranks 1 to 3. If that is too slow, add a second person for two hours a day, or bring part-time help for calling only. Do not add more groups until the first ones are finished. When we run reactivation drives with clinics, this sizing step is the one we never skip.
What message fits each of the top groups?
Each group needs its own genuine reason, written the way the doctor would speak:
- Unfinished plans: "You completed two of your four sessions with Dr Rao. We wanted to check how you are, and whether anything made it hard to continue."
- Overdue reviews: "Dr Rao had asked to see you six months after your stone treatment for a check. Shall we find a time?"
- Quoted not started: "You saw Dr Rao in March about your prostate symptoms. Do you have any questions about what he suggested?"
No offers, no discounts and no "limited slots". The book's rule holds: lead with care, not offers. The best reactivation message doesn't sell. It remembers.
When do you contact the lower groups?
Once ranks 1 to 4 are done and the replies are handled, move to recent enquiries with a single WhatsApp, and call only those who reply. Finished patients with nothing due do not need a reactivation call at all. They need a sensible educational message once in a while, and the next review when it falls due.
The short version
- Never message everyone at once. Rank groups by closeness to treatment, care reason and recency.
- Start with unfinished plans, then overdue reviews, then quoted but not started, then recent no-shows.
- Sort inside each group by recency, care remaining and doctor flags.
- Size the calling effort before you start: hours a day, minutes per call, conversations per week.
- Finish the top groups properly before touching cold enquiries.
Frequently asked questions
How many old patients can one receptionist call in a day?
With two protected hours, roughly 20 dial attempts and about 10 real conversations is realistic, because many calls go unanswered and every call needs logging. If the receptionist is also handling the main line, expect half that. Measure your own first week and plan from those numbers, not from hope.
Should I start with WhatsApp or phone calls for old patients?
For unfinished plans and quoted patients, a call works better because the reason they stopped needs a conversation. For overdue reviews and old enquiries, a WhatsApp first message saves time, then call those who do not reply. Always make sure a real person answers replies within minutes.
Is it worth contacting enquiries from three years ago?
Only after warmer groups are finished, and only those whose consent is clear. Expect low replies. One respectful message is enough, and a call only to those who answer. Your team's time is better spent on people who started treatment or are due for a review.
How do I stop the drive from overwhelming my front desk?
Release contacts in batches that match calling capacity, say 50 to 100 a week, instead of messaging thousands at once. Keep one person accountable for replies. A reply nobody answers is worse than no message at all, because the patient raised a hand and was ignored.
On a free growth call, my team traces your last month from enquiry to treatment and shows you your biggest leak in rupees.




