How many recall attempts are reasonable before stopping?

The exact figure matters less than the written rule. A common and sustainable threshold is three attempts across at least two different channels, spread over a couple of weeks, with the last one explicitly stating the clinical consequences of skipping the check. What makes the rule useful is not the number but the fact that it was decided in advance and that after the final attempt somebody records how the case was closed.

The owner always says yes and then does not come back. What do I change?

In most cases the yes is sincere and falls at the next handover, namely when the appointment is not booked on the spot. Taking the date before the person leaves the room is the intervention with the highest return. If the return still does not happen, check two concrete things: whether the expected cost was stated openly, and whether the available slots are compatible with a working day.

How do I handle checks for animals also followed by another practice?

With a written division of tasks communicated to both sides. The typical risk is the opposite of the feared one: not duplication but a gap, because each practice assumes the other is following the parameter. It helps to note in the record who owns which problem and at what interval, and to hand the owner the same scheme in readable form.

Is it really worth recording unsuccessful attempts?

Yes, for three reasons. The first is clinical: telling an unreachable patient from a never contacted one changes the week's priorities. The second is organisational: without a record every shift restarts from the first attempt and the same person receives four identical calls. The third concerns reconstructing events when a decision is challenged, where a documented attempt carries as much weight as the consultation itself.

Veterinary practice8 min readUpdated on August 5, 2026

The follow up that gets lost: where it happens and how to measure it

The points where continuity between two visits breaks, the indicators that make the loss visible and the recalls that genuinely bring patients back.

Audience: Veterinarians
Species: All species
Scope: Valid everywhere, European Union, Italy

No owner decides to interrupt their animal's course of care. The course interrupts itself, at predictable and almost always organisational points: a follow up date nobody set, an external report that arrives and sits in an inbox, a chronic treatment that continues by inertia with no reassessment, a recall call nobody owns. The loss appears in no register because its shape is absence: an animal that does not come back generates no line. As a result continuity of care gets judged by impression, and impressions are systematically optimistic. This article lists the real breaking points, proposes indicators that make the loss visible using data every practice already holds, and describes how to build a recall that produces a return rather than an irritation.

Where the chain breaks

Continuity between two visits crosses five handovers, and each has a typical way of failing. Lining them up helps you stop attributing the loss to client motivation, which is the most comfortable explanation and the least actionable.

Handovers in continuity, breaking point and observable signal
Handover	How it fails	Observable signal
Closing the consultation	The follow up stays a spoken intention rather than a date	The plan ends with the word review and no day attached
Handing over instructions	The discharge document repeats the clinical note instead of saying what to do	Phone calls asking for information already written down
Test sent outside	Nobody checks that the sample left and that the report came back	Requests with no outcome weeks later
Communicating the result	The report arrives and waits to be read and explained	Results recorded and never shared with the owner
Active recall	The task has no owner with a name attached	A recall list that grows and never shrinks

One recurring detail deserves attention: four failures out of five happen inside the practice, not in the client's home. That is good news, because it means most of the loss is fixed by internal decisions that depend on nobody's cooperation.

A follow up with no owner does not happen

The single most useful rule is also the dullest: every task due after the consultation ends needs a name, a date and a closing criterion. Miss one of the three and the task becomes a hope. The name prevents responsibility from diffusing across three people who each assume somebody else has it. The date turns the task into something that can be late. The closing criterion says when to stop chasing.

Name: the person who takes it on, not the generic role. A task assigned to reception is assigned to nobody.
Date: the day the task becomes late, not a vague window such as next week.
Closing criterion: what counts as done, for example owner reached and appointment booked, or three documented attempts.
Recorded outcome: even the failed attempt gets written down, otherwise next time somebody restarts from the first attempt.
Exit rule: after how many attempts you stop, and what goes into the record when you do.

That last item is what makes the system sustainable. Without an exit rule the recall list grows without limit, whoever runs it stops trusting it, and within a few months the whole list gets ignored. A list that empties is more effective than a complete list nobody opens.

The three categories lost most often

Not every loss carries the same clinical weight. Three categories concentrate almost all the real harm, and they share one feature: at that moment the animal looks well, so nobody perceives urgency.

The late postoperative check, the one about recovered function rather than about sutures. It is skipped most often because the wound has healed and the problem looks closed.
Stable chronic treatment, continuing through automatic renewals while nobody revisits dose, monitoring tests and side effects. Apparent stability is exactly why it never gets reassessed.
The test awaiting an external result, where the task leaves the practice and returns only if somebody is actively waiting for it.

Measuring the loss instead of estimating it

Every practice already holds the data needed to measure continuity: four counts suffice, all obtainable from the diary and the records. The absolute value matters little; what matters is the trend over time and the comparison between patient categories.

Four continuity indicators and how to calculate them
Indicator	How to calculate it	What it is for
Actual returns	Follow ups booked over follow ups planned in the record	Measures the distance between plan and reality
Median delay	Days between the planned date and the actual return	Separates the lost patient from the slow one
Results not communicated	Results recorded and not yet explained to the owner	Makes visible the queue that builds up silently
Chronic treatments without review	Active plans past their scheduled review date	Finds the renewals that replaced a clinical decision

A practical note on cadence: look at these counts monthly, not yearly. Once a year the number becomes a verdict on the organisation and nobody wants to read it; once a month it stays a list of four things to do this week, and it gets used.

The recall that works

A useful recall looks very little like a marketing reminder. It carries a clinical reason, a single action and an easy way to reply. Without the clinical reason the message reads as advertising, and the response rate collapses whatever channel you use.

Set the date before the owner leaves
The moment with the highest uptake is the one where the person is still in the room. An appointment booked immediately, even a movable one, beats any later recall.
Write the reason, not just the due date
The check exists to verify a specific parameter and to decide whether a dose is still appropriate. One sentence with the clinical reason changes how the message is read.
Ask for one action only
Confirm, move or call back. A message offering three tests, a supplement and a promotion gets zero replies on every item.
Document the attempts
Date, channel and outcome of each attempt stay in the record. It is the only way to know whether the patient is unreachable or whether nobody actually tried.
Close the case explicitly
Once the exit rule is met, write in the record that the pathway was interrupted and why. A badly closed case occupies mental space for years.

What the tools make available

No tool replaces the organisational rule, but some lower the cost of applying it. The features available in the veterinary workspace and on the owner side are these, with their real placement.

Automatic recalls in the clinic workspace, built on declared references such as international vaccination guidelines, outside the free plan.
Clinic agenda and availability, online booking on the owner side, reception queue with clinical triage.
Treatment plans with interaction checking and clinical records with templates, where the review date can be set alongside the prescription.
Laboratory reports shared with the owner, which move result communication from a phone call to a document that can be consulted.
Clinic statistics, useful for deriving the counts described above instead of estimating them.
On the owner side: reminder centre, recurrences, treatment adherence and push notifications, with predictive reminders outside the free plan.

The rest deserves saying too. The product sends no text messages, runs no bulk email campaigns, integrates no telephone system and receives no reports directly from external laboratories: results arrive because somebody uploads them. Anyone designing a recall system therefore has to decide which channel to use outside the application, while still keeping the trace of the attempts inside the record.

Frequently asked questions

How many recall attempts are reasonable before stopping?: The exact figure matters less than the written rule. A common and sustainable threshold is three attempts across at least two different channels, spread over a couple of weeks, with the last one explicitly stating the clinical consequences of skipping the check. What makes the rule useful is not the number but the fact that it was decided in advance and that after the final attempt somebody records how the case was closed.
The owner always says yes and then does not come back. What do I change?: In most cases the yes is sincere and falls at the next handover, namely when the appointment is not booked on the spot. Taking the date before the person leaves the room is the intervention with the highest return. If the return still does not happen, check two concrete things: whether the expected cost was stated openly, and whether the available slots are compatible with a working day.
How do I handle checks for animals also followed by another practice?: With a written division of tasks communicated to both sides. The typical risk is the opposite of the feared one: not duplication but a gap, because each practice assumes the other is following the parameter. It helps to note in the record who owns which problem and at what interval, and to hand the owner the same scheme in readable form.
Is it really worth recording unsuccessful attempts?: Yes, for three reasons. The first is clinical: telling an unreachable patient from a never contacted one changes the week's priorities. The second is organisational: without a record every shift restarts from the first attempt and the same person receives four identical calls. The third concerns reconstructing events when a decision is challenged, where a documented attempt carries as much weight as the consultation itself.

What to do next

Pick one of the five breaking points and close it this week: the best return comes from setting the follow up date before the owner leaves the room, with the clinical reason written beside it. Then count two numbers at the end of the month, actual returns over planned ones and results recorded but not yet communicated. Once those two numbers start moving in the right direction, add the next point.