Predictive Patient Tracking
Cancellation rate reduced by 31% within 90 days; treatment room utilisation improved from 67% to 79%.
The challenge
Appointment cancellations were eroding revenue and staff capacity without any predictive mechanism to intervene before the slot was lost.
The group's cancellation rate was running at approximately 22% across locations, with same-day cancellations and no-shows accounting for nearly half of that figure. The revenue cost was quantifiable — roughly £18,000 per month in lost treatment revenue at average treatment values — but the operational cost was harder to measure: staff time spent on reactive rescheduling, practitioner downtime, and the compounding effect on patient lifetime value when cancellation behaviour became habitual. The existing approach — reminder SMS sent 24 hours before appointment — was uniform and unresponsive to individual patient behaviour patterns. There was no mechanism to identify which patients were likely to cancel, no tiered intervention logic, and no way to fill vacated slots efficiently.
The system
Decision system built
We designed and implemented a predictive patient tracking system that scored individual appointment risk ahead of time and triggered differentiated retention interventions based on that score. The system operated within the group's existing practice management infrastructure, using historical behaviour data already being collected but not actioned.
System components
Cancellation risk scoring model: trained on 18 months of historical appointment data, incorporating lead time, patient tenure, treatment type, booking channel, and previous cancellation history
Tiered intervention logic: three-tier communication sequence mapped to risk score — standard reminder for low-risk, personalised check-in for medium-risk, direct outreach from front-of-house for high-risk
Slot recovery workflow: automated identification of waitlisted patients matched to vacated slots by treatment type and location, with templated outreach reducing fill time from hours to minutes
Utilisation dashboard: real-time and rolling 30-day view of room utilisation by location, practitioner, and treatment category to identify structural capacity issues versus behavioural ones
Feedback loop: weekly model recalibration incorporating new cancellation and attendance outcomes to maintain scoring accuracy over time
How we worked
Engagement scope
Predictive cancellation risk modelling, tiered intervention design, slot recovery workflow, and utilisation reporting across a 6-location aesthetics group.
Timeline
8-week build and integration, 4-week calibration period, ongoing model refresh cadence.
Operating model
Embedded delivery with the group's operations manager as primary owner. Training provided to all front-of-house leads on intervention tiers and escalation logic. Governance checkpoint at 30 and 90 days post-launch.
Outcomes
Business impact & measurable results
Cancellation rate reduced by 31% within 90 days; treatment room utilisation improved from 67% to 79%.
Cancellation rate reduced from 22% to 15.2% within 90 days — a 31% relative reduction — with same-day cancellations falling disproportionately from 11% to 6.4%
Treatment room utilisation improved from 67% to 79% across the group, recovering approximately £14,000 in monthly revenue that had previously been lost to vacant slots
Slot recovery workflow reduced average time-to-refill from 4.2 hours to under 45 minutes on high-demand days
Front-of-house staff reported a meaningful reduction in reactive rescheduling workload, with structured intervention protocols replacing ad hoc decision-making
Governance
Trust, collaboration & governance
Model logic explained to clinical and front-of-house teams — no black-box scoring that staff couldn't interrogate
Human-in-the-loop preserved for high-risk interventions: the system flagged, staff decided how to act
Patient communication templates reviewed by the clinical team for tone and compliance with consent frameworks
Weekly accuracy reporting to allow rapid recalibration if behavioural patterns shifted seasonally
Reframe
The signals were already in the system, unread — the intervention was reading them in time.
Across every engagement, the goal is the same: engineer a system that makes better decisions — faster, more consistently, and at scale — than the process it replaces.
Start a discovery
Most engagements begin with a conversation about context.
We do not send a proposal before we understand the problem. Start by telling us about your decision context — we will identify the highest-leverage intervention areas before any scope is agreed.