AI in Physiotherapy: Filling Cancellations From a Waiting List

A physio clinic can fill last-minute cancellations by keeping an accurate waiting list, matching each open slot to patients who requested that appointment type and time, and sending an approved offer through its booking system. AI can draft the matching rules and messages; staff or the configured system must confirm one booking, withdraw other offers and log the outcome.

The critical distinction is between an offer and a reservation. If three people receive a message, all three may reply. The clinic needs a clear rule for who can claim the slot, how long it is held and when other messages stop. A friendly AI-generated text does not prevent double booking or make a patient clinically suitable for an appointment.

Follow me on Instagram@sagnikteaches

Record the waiting-list fields that make a match possible

Collect only information needed to offer the right appointment. For each request, record the requested appointment type, acceptable practitioners, duration, days and time windows, earliest acceptable date, contact channel, date added and whether the person still wants an earlier slot. Keep clinical notes in the proper record, not in an AI matching sheet. The list should be reviewed when a patient books elsewhere, changes availability or declines repeated offers.

Connect on LinkedInSagnik Bhattacharya

For an illustrative physio clinic, a 30-minute follow-up cancellation at 14:00 cannot automatically go to someone waiting for a 60-minute first assessment. It may fit a patient who requested a follow-up with that practitioner after 13:00. It may not fit a patient who can attend only in the morning. These simple constraints matter more than a model's guess about who sounds "urgent". Clinical priority, if used, must be set by an authorised clinician under the clinic's own process.

Subscribe on YouTube@sagnikteaches

Cliniko's wait list guidance records the appointment type a patient is waiting for, the days they can come in and the practitioners they would like to see, with filters for practitioner, business and appointment type. When staff choose "select from wait list" while booking, Cliniko shows the patients who are available that day and want that practitioner. Jane lets patients add their preferred dates and times to its wait list from the clinic's online booking site. Check the exact fields and permissions in your current account; a spreadsheet copied out of the system can become stale within an hour.

A second illustrative case is a patient who needs an accessible room. The scheduling record should encode the operational requirement in the approved system so staff can filter appropriately. Do not paste the patient's medical explanation into a generic AI prompt just to ask whether a slot fits. The offer decision needs the approved room requirement and availability, not a diagnosis. If the room is unavailable, the slot is ineligible even though the duration matches.

Write an offer order the receptionist can explain

Choose a rule before cancellations arrive. A small clinic might filter by appointment type, practitioner, duration, location and availability, then offer to the earliest waitlist request among eligible patients. Another clinic may use a clinically approved priority band first. Either can work if the rule is documented, consistently applied and reviewed for unfair effects. Do not let AI infer urgency from the length or emotional tone of a message. A clinician, not a language model, should set any clinical priority.

Decide between sequential and simultaneous offers. Sequential contact gives one person a defined window to reply, then moves to the next. It is easy to explain but can leave a very short-notice slot idle. A simultaneous system may offer the same opening to several eligible patients and award it to the first completed booking; that needs a system that handles the race and tells others the slot has gone. If the clinic sends manual texts, a controlled sequential offer is usually easier to supervise.

For an illustrative policy, a cancellation more than 24 hours ahead goes to the first eligible person for 30 minutes. If unclaimed, staff move to the next. A cancellation tomorrow morning might have a 10-minute phone-first window during staffed hours. These are example thresholds to test, not recommended universal settings. Ask whether patients can reasonably see the message, travel and prepare. Set a final cutoff after which no offer is sent, even if the slot stays empty.

Write the rules on one page with an exception column. If a patient has declined two offers this week, do you pause them, or do their availability preferences need updating? If the appointment requires a referral or a particular assessment, who checks it? If a slot comes from a practitioner illness rather than a patient cancellation, confirm that it is actually available before messaging. A documented exception avoids a receptionist improvising under time pressure.

Draft messages that say exactly what can be claimed

Give AI a policy with the approved appointment types, offer windows, response route and neutral tone. Ask it to produce a manual message, a phone script and an "already taken" response using placeholders only. Never put a patient's name, phone number, condition or appointment history into a general chat to personalise the text. The clinic's practice system should populate contact details after eligibility and consent checks.

Draft three short waiting-list messages for a physio clinic:
one time-limited offer, one confirmation, one slot-no-longer-available.
Use [first name], [appointment type], [day], [time], [reply route]
and [offer expiry] as merge fields.
Say that the offer is not a booking until confirmed.
Do not mention a diagnosis, imply clinical urgency or invent a link.
Use this approved offer rule: [paste rule without patient data].

An illustrative offer could read: "Hello [first name], a [appointment type] slot has opened on [day] at [time]. If you would like it, please use [approved reply route] by [offer expiry]. The appointment is confirmed only when you receive a booking confirmation. If this time no longer suits you, you can update your waiting-list preferences through [approved route]." The manager should make it shorter for a text channel if needed and check that the response route is actually monitored.

A poor AI draft might say, "Your urgent treatment has been reserved; reply YES now." It invents urgency and a reservation. Correct it to the actual appointment type and claim rule. A second poor draft includes the physiotherapist's private mobile number because it appeared in a rough policy note. Remove private contact details from the source material and use the clinic's approved channel. The message library should have a version date so staff do not copy an older text with a dead booking link.

Follow one week of cancellations through the diary

Here is an illustrative week for a two-physiotherapist clinic. The waiting list has 18 active requests. Four cancellations create 150 minutes of apparent capacity. The clinic uses its approved matching rule and a receptionist during opening hours; AI helped write the rule and messages but does not see patient data or make bookings. The numbers below illustrate workflow decisions, not a claimed fill rate.

The Monday slot is straightforward. The receptionist checks that the calendar opening is real, the appointment type matches and the first patient still wants an earlier visit. The offer expires at a stated time. The patient accepts through the approved route, staff book inside the practice system and send the system's confirmation. The next two eligible patients receive nothing because the first claim succeeded. The outcome log records the opening, eligible count, offer time, acceptance time and booking reference, without exporting clinical details.

On Tuesday, the only eligible assessment patient declines. The clinic could release the slot to normal online booking or use an approved public availability channel; it should not offer a 60-minute assessment to a 30-minute follow-up request just to fill the diary. On Thursday, the first patient's 30-minute window expires. Staff verify there was no late response or duplicate booking before contacting the second patient. If the first patient replies after expiry, the receptionist uses the approved "slot no longer available" message and keeps the person on the list if they wish.

The Friday opening appears 35 minutes before the start, inside the clinic's illustrative 60-minute cutoff. Staff do not send a speculative text that a patient cannot reasonably act on. The week ends with two of four openings filled from the list, one released and one left unoffered by policy. That count is an observation for this fictional week, not evidence of typical performance. It gives the owner a way to assess whether the cutoff or list quality needs changing.

Fit the rules to Cliniko or Jane rather than duplicating them

Cliniko lets staff add patients to a wait list and filter by practitioner, business and appointment type, then select from the list while creating an appointment. It also offers cancellation alerts for practitioners: off by default, switched on per practitioner under My info, sent by SMS, email or both with a chosen notice window, and triggered when another user or a patient presses Cancel. The wait list help describes staff picking patients from the list rather than automatic offers to patients, so plan a staff-led process unless your account shows otherwise. Check your account settings and staff permissions before designing an external alert or automation. The physiotherapy software comparison helps with a broader platform decision.

Jane's waitlist notification guide says wait lists are available on all plans, while Wait List Notifications are only on the Thrive plan and two legacy plans. Notifications go out automatically or when staff press Send Notifications Now. After an admin buffer (15 minutes when the appointment starts within 24 hours, two hours otherwise), an Exclusive Access period holds the spot for wait-list patients before Jane releases it to anyone. The clinic should verify the current plan and settings. If Jane already handles matching and booking, AI's useful job may be drafting the clinic's policy and reviewing message clarity, not recreating a notification engine outside Jane.

Jane sends each notification by email, SMS or both, as the patient chose when joining the list. The email links to book the spot online or prompts the patient to call, and anyone who clicks after the slot has gone sees a message that it is no longer available. Because every eligible patient is notified at once, Jane's route works as a simultaneous offer: the owner should decide whether that suits the clinic, how long the exclusive hold should be and what happens when a patient cannot book online. The guide also says notifications cannot currently be sent to patients waiting for a class or group appointment, so a clinic running classes needs a manual route for those. A technical feature does not eliminate the need for staff to resolve exceptions.

An illustrative setting error: a clinic switches on automatic notifications and assumes same-day cancellations will now fill themselves. Jane's troubleshooting guide explains that a cancellation falling inside the clinic's Do Not Allow Booking Within period (under Settings > Online Booking) triggers no automatic notification, so a late-afternoon gap can sit empty while staff believe the system is handling it. Decide who watches for those short-notice openings and sends offers by hand. Test a fictional cancellation close to the appointment time, with a staff member acting as the patient, before going live. If the booking is contact-only, the message must point to a staffed route and the clinic must be able to answer promptly.

Stop messages when the slot or patient state changes

Write explicit stop conditions into the workflow. Stop all offers for a slot when it is booked, blocked, moved, no longer clinically appropriate or within the cutoff. Stop offers to a patient when they withdraw, opt out of the channel, book the same visit elsewhere, or no longer meet the appointment criteria. Do not assume the AI remembers these events; read them from the live practice system immediately before each send and again before confirming a booking.

For example, a patient accepts a Thursday offer by text at 16:12 while another patient books the slot online at 16:11. The receptionist should not promise the appointment based on the text timestamp alone. Check the diary, explain that the opening has been taken, offer to keep the patient on the list and document the exchange. If the clinic's system offers an exclusive hold, use and test it; if it does not, message wording must make the conditional nature of the offer clear.

A different edge case is practitioner continuity. A patient waiting for a follow-up with one physiotherapist may not want a slot with another, even though both can deliver that appointment type. Ask and record acceptable practitioners when adding the waitlist request. If the clinic changes the request without consent, a rapid cancellation offer can feel like a switch. The receptionist should check the selected practitioner and room before confirming, particularly when several diaries show open time.

Clinical exceptions belong with clinicians. If a patient says symptoms have worsened while asking for an earlier slot, the receptionist should follow the clinic's triage and escalation policy, not score the message with AI or treat the waitlist as a substitute for urgent assessment. The physio receptionist guide covers the broader call and booking hand-off. Here, the waiting-list process stays within approved scheduling criteria.

Review whether the list is current enough to help

Track openings, eligible matches, offers sent, accepted, expired, declined, already-booked conflicts and minutes left unused. Add a reason when there was no match: wrong appointment type, incompatible time, channel opt-out or list stale. A low fill count may mean the clinic's schedule is unusual, not that AI wording is poor. Review a week of cases with staff before changing thresholds or sending more messages. The missed-appointments tutorial addresses an upstream cause of unused capacity.

Ask patients who decline whether they still want to remain on the list and update preferences in the practice system. Keep a routine clean-up date so a request from months ago is not treated as active indefinitely. For an illustrative list of 18, the receptionist may find three patients who have already booked and two whose available days changed. Removing those five stale requests makes future matching faster without any new model or paid automation.

Test the complete journey after every settings change: create a fictional waitlist request, cancel a fictional appointment, check eligibility, send or trigger an offer, claim it, confirm that other routes stop, and inspect the log. Do the test through the same channels patients use. A template can look correct in an editor yet include a broken link in a text message. The owner should sign off the whole booking path, not merely the AI-written copy.

Check whether the offer order is fair in practice

A written first-eligible rule can still behave unfairly if the data behind it is poor. Patients who cannot answer a daytime text within ten minutes may repeatedly lose openings to those who can. Review the offer and acceptance log by time window and channel, without exporting clinical information to a general AI tool. Ask whether the clinic's chosen window gives people a reasonable chance, and whether the rule for clinical priority was applied by an authorised person. If a patient asks why they were passed over, staff should be able to explain the non-clinical matching criteria without revealing another patient's details.

For an illustrative comparison, the clinic sends twelve sequential offers in two weeks. Six go out during staffed daytime hours and six close to the end of the day. Five daytime offers are accepted, while none of the late offers are seen before their 15-minute expiry. That small sample does not prove a general pattern, but it gives the owner a reason to test a longer late-day window or a phone call for very short-notice slots. The revised rule should be trialled and measured, not quietly applied only to selected patients.

Another fairness issue appears when staff keep adding only people who telephone repeatedly. A patient who submitted an online waitlist request may be eligible but invisible if those requests are not reviewed. Schedule a daily check of new requests, confirm contact preferences and apply the same matching order. If a person needs an accessible communication route, record the operational preference in the approved system and train staff to use it. AI may help rewrite the policy in plain language, but the clinic must make the list complete.

Keep a simple reason code for every unfilled opening. "No eligible patient" is different from "eligible patient did not reply" and "system booking rule blocked claim". In an illustrative month with 20 cancellations, six no-match cases may suggest that the list needs better appointment-type data, while four booking-rule failures point to settings. Those figures are diagnostic examples, not a benchmark. Correct the actual bottleneck before buying another tool or increasing message volume.

Waiting-list offer questions

Should every waiting patient receive the same cancellation text?

No. Match the appointment type, duration, practitioner, location, availability and contact preference first. Use a short approved message that contains only the details needed to accept the slot. A patient who asked for a different type of visit or opted out of that channel should not receive the offer.

Can AI book the first person who replies?

Only if an approved booking system and rule can prevent simultaneous claims and the clinic has tested the workflow. For a small clinic, staff or the practice software should make the final booking and confirmation. AI can help draft the rule and message, but a draft response is not a reserved appointment.

What if a cancellation is too close to the appointment time?

Set a stop time that leaves the patient enough travel or preparation time and staff enough time to confirm the booking. A last-minute clinical slot may need a phone call rather than a mass message. The clinic should decide this threshold from its own opening hours and appointment types.

Further reads

• AI Tools for Physiotherapy Clinics: What Each One Actually Does — Place scheduling among other physio AI tasks.
• AI for Physiotherapists: Employer and Workplace Contract Proposals — Plan a separate route for workplace visits.
• AI in Speech and Language Therapy Clinics: Booking and Waiting Lists — Compare another therapy clinic's booking constraints.
• AI Chatbot for Physio Clinics: First-Visit Questions and Booking Hand-Offs — Handle first-visit questions before a booking.
• How a Small Clinic Can Use AI at the Front Desk Safely — Set wider front-desk message and hand-off rules.
• Osteopath Software: Which Packages Add AI and What to Check — A seven-part checklist for osteopaths comparing practice software that advertises AI, with verified vendor prices and a scoring sheet.
• Chartered Society of Physiotherapy and AI: A Clinic Checklist — Turn the CSP's AI principles into a working checklist: what each asks of a clinic, who owns it, what evidence to keep and when to review.
• AI Agents for Healthcare: Six Safe First Jobs for a Small Practice — Six admin-only first jobs for an AI agent in a small practice, the clinical work that stays human, and the contract checks each tool needs.
• AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.

Sources: Cliniko Help, Using the wait list and Appointment cancellation notifications for practitioners; Jane Guide, Setting Up Your Wait List Notifications, Wait List Notifications: Client Experience, and Why didn't my patient or client receive a wait list notification? Check current plan availability and settings before relying on automation.