← All customer results

Rural dispensing practice · 12,000+ patients

Deddington Health Centre

Immediate, queue-free patient access at a busy rural practice

Reporting period 2 June – 1 July 2026 · Pilot May 2026 · Live from June 2026

  • 12,000+ Registered patients
  • Good CQC overall rating
  • NORA Primary Care Network
  • Live After a May 2026 pilot

Key results at a glance

2,191 Calls handled
1,142 Resolved end-to-end (52%)
54.2h Reception time returned
2.8 min Average resolution time

Executive summary

Deddington Health Centre introduced Jackie as the first point of contact on inbound patient calls. Jackie answers immediately with no queue, captures appointment, sick-note and routine-care requests through structured questioning, books eligible blood-test and cervical-screening appointments directly into the practice’s booking system, flags potential urgent symptoms for the practice team, and routes everything else on with a structured summary attached. All clinical decisions remain with practice staff.

Following a successful pilot in May 2026, Deddington took Jackie live. Over the 30-day analytics period Jackie handled 2,191 inbound calls, resolved 1,142 of them end-to-end with no reception involvement, and reclaimed 54.2 hours of reception call-handling time.

The decision to go live followed the verdict of the team that works with Jackie every day: surveyed at the end of the pilot, 88% of the practice’s patient services team said they wanted to continue.

The challenge

The practice telephone is the front door of general practice, and when it becomes congested the whole system feels it. Deddington Health Centre — a rural dispensing practice serving more than 12,000 patients across Deddington and the surrounding north Oxfordshire villages — faced the operational pressures familiar to busy practices, sharpened by a list that has grown steadily in recent years. In the 2025 GP Patient Survey, only 44% of the practice’s patients said it was easy to get through on the phone. Improving that access, without adding staff the practice does not have, was the goal.

Concentrated peaks of demand

A human receptionist can take only one call at a time. Demand is heavily front-loaded: the busiest hour in the period was Monday 08:00–09:00, with weekday mornings consistently the pressure point. During these windows patients queue, redial, or give up.

High-volume routine work

A large share of reception time goes on routine, repeatable requests — appointment bookings, blood-test monitoring, medication reviews and sick notes — that consume capacity skilled staff could spend on patients who need a person.

Access as a safety issue

Long phone queues are not only a source of dissatisfaction; they can delay the surfacing of urgent need. Any solution therefore had to improve access without compromising clinical safety, and without removing the option to speak to a person.

The solution

Jackie is a digital receptionist designed to work alongside the reception team, not replace it. At Deddington Health Centre, Jackie is the first point of contact on inbound calls, handling high-volume routine workflows end-to-end while preserving full human oversight and a route to a person at any time.

  • Immediate answer, no queue Every call is picked up at once, eliminating hold time.
  • Concurrent capacity Jackie handles multiple simultaneous calls — capacity a single receptionist cannot match.
  • Structured capture Patients are guided through relevant questions, producing a consistent, accurate summary for the practice team.
  • Direct booking Jackie books eligible blood-test and cervical-screening appointments straight into the practice’s booking system, so patients leave the call with a confirmed appointment rather than waiting for a callback.
  • Safety indicator detection Where a patient describes potentially urgent symptoms, Jackie advises the caller it sounds urgent and escalates to reception.
  • Routing with context Calls Jackie cannot or should not complete are passed to the team with a structured summary already attached, reducing manual re-keying.

Quantifiable outcomes

Across 2,191 inbound calls in the period, Jackie carried the routine load while routing everything that needed a person cleanly, with a structured summary attached.

Call outcomes for Deddington Health Centre over the reporting period
OutcomeCallsShare
Resolved end-to-end by Jackie 1,142 52%
Routed to the practice team 805 37%
Abandoned by caller 222 10.1%
72.8%
Fully handled by Jackie

Jackie’s fully-handled rate for the period — the share of calls she carried to a complete outcome once out-of-scope requests and early caller abandonment are set aside. It measures how well Jackie completes the work she is actually configured to do, and is a different metric from the 52% resolved end-to-end above.

Of the calls Jackie resolved, the largest categories were general enquiries (41%) and appointment requests (36%), followed by routine care (19%), with safety-indicator escalations (4%) and sick notes (1%) making up the remainder.

Where a person was needed, this was most often because the patient asked for one (49%) or the request fell outside Jackie’s configured scope (44%).

Speed and capacity

Reception call-handling time reclaimed
54.2 hours
Average resolution time (Jackie-handled)
2.8 minutes
Average handling time (human-supported)
7.3 minutes
Peak simultaneous calls
6

Clinical signal

What the month showed clinically

Reclaimed hours are the means, not the result. At Deddington the clinically interesting part is who reached the practice and what was captured when they did — on a rural list where a phone queue and a web form exclude different people.

  • Potentially urgent symptoms surfaced, not sorted Safety-indicator escalations accounted for 4% of the calls Jackie resolved. In each, Jackie told the caller it sounded urgent and passed it to the practice team. She does not decide urgency or prioritise — that judgement stayed with practice staff on every call.
  • Monitoring booked on the call, not queued for one Blood-test monitoring was the single most common routine-care request at 31%, and Jackie books those appointments directly. Cervical screening is booked the same way. Work that usually waits for a call-back was completed while the patient was still on the line.
  • The phone stayed a full channel Every request was captured in the same structure whoever called and whenever they called — no smartphone, app or web form required. On a rural list that is the difference between a patient being heard and being asked to try another way.

Demand insights for resource planning

Beyond handling calls, Jackie turns call data into a planning asset, giving practice leadership a granular view of why patients are getting in touch.

Appointment requests by presenting category 409 requests

  • Follow-ups, discussion about tests 21%
  • Dermatological 17%
  • Other 17%
  • Musculoskeletal 13%
  • Respiratory 7%
  • Urinary 7%
  • Sensory 7%
  • Abdominal 5%
  • Mental health 4%
  • Neurological 3%

This lets leadership anticipate clinical demand — for example, planning ahead of seasonal respiratory spikes.

Routine care requests 211 requests

  • Blood-test monitoring 31%
  • Medication review 19%
  • Vaccination 15%
  • Other 13%
  • Long-term condition review 13%
  • Postnatal check 3%
  • Cervical screening 2%
  • NHS health check 2%
  • Pre-travel 1%

Jackie books blood-test and cervical-screening appointments directly — including blood-test monitoring, the most common routine-care request in the period — so these are completed on the call rather than passed on. Sick-note requests were low in volume (6 in total), most either extensions or cases Jackie identified as not eligible and resolved with appropriate guidance.

Patient and staff experience

Patient perception is decisive for any change to how people access their practice. Feedback is collected automatically: after any interaction with Jackie lasting longer than 90 seconds, the patient is sent an SMS inviting them to rate their experience. Over the period, 1,830 of these messages were sent and 128 patients responded — a response rate of roughly 7%, in line with typical SMS survey benchmarks.

59.4% Positive or neutral sentiment
128 Patient responses
88% Of the patient services team wanted to continue

What patients valued

  • “The first AI I have ever been happy with!”
  • “Very easy and straightforward — I am happy and impressed.”
  • “It certainly felt like talking to a human being.”
  • “I was very impressed by the system, and it really helped me get assistance this morning when I was really feeling unwell.”
  • “I didn’t have to repeat anything — surprisingly straight to the point, and got me what I required.”

Addressing friction points

The themes below are drawn from the free-text comments left by patients who rated their experience neutral or negative, and translate directly into a configuration and model-improvement roadmap.

Patient feedback themes and the actions taken
Theme Action
Preference for a human receptionist Clearer up-front opt-out, with a set phrase to reach a person at any point
Speech understanding (accents, names, medical terms) Speech-model tuning for local accents and terms; removing repeat loops
Callback expectation vs. resolution Clear outcome expectations and consistent, communicated callback timeframes
Older or less digitally confident patients Slower pacing option and plain-English guidance on how to speak to Jackie
Out-of-scope requests (e.g. prescriptions, results) Clearer signposting and, where agreed, scope expansion or direct routing

This feedback loop — capturing patient sentiment at scale and translating it directly into configuration and model improvements — is central to how Jackie matures within each practice. The critical minority here is a roadmap, not a verdict: each theme maps to a specific, already-identified change.

Access

Jackie answers immediately, so patients who engage never wait in a queue. Across the period the overall call-abandonment rate was 10.1% (222 calls) — most of these early hang-ups before Jackie engaged, with the remainder in-call. Abandonment concentrates at the same weekday-morning peaks where human lines are historically most stretched, which is precisely where Jackie’s always-on, concurrent answering adds the most value.

Conclusion

  • Immediate access and capacity: zero queue, and 54.2 hours of reception time reclaimed in the period.
  • End-to-end resolution: 1,142 calls completed by Jackie with no reception involvement, at an average of 2.8 minutes.
  • Net-positive patient reception: a majority rating their experience four or five out of five, with candid critical feedback driving a concrete improvement roadmap.
  • Staff endorsement: surveyed at the end of the May pilot, 88% of the practice’s patient services team wanted to continue — and the practice went live.
  • Clinical capture: potentially urgent symptoms escalated to the practice team, and monitoring and screening appointments booked during the call rather than queued for a call-back.
  • Insight for planning: granular demand data across appointments, routine care and sick notes to support resource planning.

About Deddington Health Centre

Deddington Health Centre is an NHS GP practice based at Earls Lane in the rural market town of Deddington, north Oxfordshire, serving more than 12,000 registered patients across Deddington and the surrounding villages. It is a dispensing practice rated Good by the Care Quality Commission, and a member of the North Oxfordshire Rural Alliance (NORA) Primary Care Network — one of five practices covering approximately 47,000 patients. Jackie was piloted at the practice in May 2026 and went live in June 2026.

Operational figures are drawn from Jackie’s digital reception analytics for 2 June – 1 July 2026 and from patient feedback collected 15 May – 1 July 2026. Practice profile (list size, CQC rating, PCN membership) is from published NHS, CQC and Oxfordshire County Council sources.

Take routine patient contact off your team’s day.

See how an Auxilis worker could handle a defined patient workflow, complete approved actions and bring in your team only when they are needed.

Meet the workers