Key takeaways
- Plan the rota from demand by half hour, not from habit: calls, online requests and desk visitors each need their own line.
- Convert calls to staff with a simple formula: calls per hour multiplied by average handling time, divided by 3,600, then allow for breaks and time between calls.
- Online requests are now a core workload: practices received 9,617,628 in March 2026, 69% more than in August 2025[1].
- Expect Mondays, the days around bank holidays and the start of each month to break the pattern, because call volumes follow the calendar[2].
- Test the rota for four weeks, compare forecast with actual demand and adjust one thing at a time.
Why rotas built on habit fail
Most reception rotas grow by accretion. Someone always starts at 8, someone always does the late, and the middle of the day is filled with whoever is available. The result is too few people at the peak and too many in the quiet afternoon. With 45% of GP appointments in the year to August 2026 booked for the same day[3], demand is concentrated early, and a rota that ignores the shape of the day makes the first hour harder for everyone.
A demand-led rota starts from what actually arrives: calls by half hour, online requests by hour, visitors at the desk, plus fixed tasks such as post, scanning and prescriptions. It takes a few hours to build the first time and much less to review.
The data you need
- Calls offered per half hour for four typical weeks, from your telephony reports, including abandoned calls.
- Average handling time per call, including after-call work such as booking or messaging a clinician.
- Online requests received per hour, and the average time to process one.
- Desk visitors per half hour, which a simple tally sheet will give you over two weeks.
- Fixed tasks and their usual duration: opening up, post, scanning, prescriptions, closing.
- Planned absences and known pattern breakers: bank holidays, flu clinics, extended access sessions.
Use the median week rather than the busiest, then plan separately for predictable peaks. Planning every week for the worst Monday wastes hours; planning for an average Tuesday leaves Mondays short.
Converting demand into staff
Calculate phone workload
For each half hour, multiply calls by average handling time in seconds and divide by 1800 (the seconds in half an hour). The result is the number of people fully occupied on the phones for that half hour.
Allow for time between calls
Nobody can take calls back to back all day. Divide by about 0.8 to allow for short gaps, toilet breaks and the odd long call.
Add online request workload
Multiply requests arriving in the period by the minutes each takes and divide by the minutes in the period.
Add desk and fixed tasks
One person on the desk during opening hours is usually a minimum; add a second when visitors exceed what one can handle.
Round up and build shifts
Round each half hour up to whole people, then fit shift patterns that cover the shape with as little waste as possible.
Example: a fictional practice receives 120 calls between 8.00 and 8.30 on a typical Tuesday, averaging 150 seconds each. That is 120 × 150 ÷ 1,800 = 10 people fully occupied, or about 12.5 allowing for gaps. Even a large team cannot staff that, which is the signal to spread demand, through online access, call-back and releasing appointments in waves, rather than to rota more people for half an hour.
A worked half-hour plan
| Time | Calls (example) | Phone staff | Online and admin | Desk | Total |
|---|---|---|---|---|---|
| 08:00-08:30 | 50 | 6 | 0 | 1 | 7 |
| 08:30-09:00 | 35 | 4 | 1 | 1 | 6 |
| 09:00-10:00 | 55 per hour | 3 | 1 | 1 | 5 |
| 10:00-12:00 | 35 per hour | 2 | 2 | 1 | 5 |
| 12:00-14:00 | 25 per hour | 2 | 1 | 1 | 4 |
| 14:00-16:00 | 30 per hour | 2 | 2 | 1 | 5 |
| 16:00-18:30 | 18 per hour | 1 | 1 | 1 | 3 |
Notice the online and admin column. In this example, two people move to online requests and admin once the morning phone peak passes, rather than going on break together. Breaks are then staggered across late morning and early afternoon, where the plan has slack.
Sizing the online request queue
Online requests are no longer a side task. In March 2026 practices received 153 online requests per 1,000 patients[4]. For a practice of 10,000 patients that is roughly 1,530 a month, or about 70 to 75 each working day, before you count requests that arrive overnight and at weekends.
Source: Care Quality Commission, 2026[1]
Show the data
| Item | Value |
|---|---|
| August 2025 | 5,690,082 |
| March 2026 | 9,617,628 |
Example: if each request takes four minutes to read, code and route, 75 requests is five hours of work a day. That is most of one person's shift, and it peaks in the morning alongside the phones. Rota a named person for the online queue at the busiest times, and set a target such as every request reviewed within an hour of arrival during core hours.
Shift patterns that fit the shape of the day
Once you know how many people each half hour needs, design shifts to match. A demand curve that peaks at 8am and tails off after 4pm suits a mix of long and short shifts rather than everyone working the same eight hours.
- Early peak shifts, such as 7.45 to 12.15, cover the morning surge and suit staff who prefer to finish by lunchtime.
- Full-day shifts, such as 8.00 to 4.30, provide continuity and someone who knows what happened in the morning.
- Late shifts, such as 10.30 to 6.45, cover the afternoon and closing without making anyone work twelve hours.
- Short school-hours shifts, such as 9.15 to 2.45, fill the middle of the day and widen the pool of people who can work for you.
- Overlap of 15 minutes at handover, so messages and unfinished tasks are passed on properly.
Check every pattern against the half-hour plan. Where the rota gives you more people than needed, move a start time or turn a full day into a peak shift. Where it gives you fewer, look first at whether demand can be moved before adding hours.
Days that break the pattern
NHS England notes that call volumes are affected by the number of days in the month, public holidays and weekend days[2]. Practices see the same thing at local level: Mondays are heavier, the day after a bank holiday is heavier still, and the weeks around Christmas are distorted. In England and Wales the 2026 Christmas period includes a four-day closure from Friday 25 to Monday 28 December[5].
- Build a separate Monday pattern with an extra person for the first two hours.
- Plan the day after each bank holiday as an extra Monday.
- Plan the Christmas and New Year weeks individually, at least six weeks ahead, and protect leave fairly across years.
- Staff enhanced access sessions, typically 6.30pm to 8pm on weekdays and 9am to 5pm on Saturdays where your PCN runs them, with their own rota line[6].
- Add capacity for flu clinic booking periods and other seasonal campaigns.
Planning for absence
A rota that only works when everyone is present will fail most weeks. Build in planned cover for annual leave from the start, agree how many people can be off at once in each part of the year, and keep a short list of staff willing to pick up extra hours at short notice. Cross-train at least two people on every fixed task, such as scanning, prescriptions or registrations, so a single absence does not stop the work. Review sickness patterns quarterly; a cluster of Monday absences often points to a workload problem rather than an attendance one.
Reducing peak demand instead of chasing it
Every call moved out of the first hour reduces the staff you need at the worst time. The most effective levers are on the practice side: release appointments in waves through the day, keep the online route open from 8am and say so early in the queue message, and offer a call-back option. NHS England trials found moving practices to high-quality digital telephony raised patients' ease-of-getting-through scores by 30 percentage points[7].
Routine questions about opening hours, test results processes or how to order prescriptions do not need a receptionist. An AI Employee such as Ava can answer practice information calls and cover peaks, passing anything patient-specific to the team, which leaves your people for the calls that need them.
Test, compare and adjust
- Run the new rota for four weeks without major changes.
- Each week, compare forecast and actual calls, abandoned calls and online request response times by half hour.
- Ask the team where they felt overstretched or idle, and note it against the data.
- Change one thing at a time, such as moving a start time by 30 minutes.
- Review the whole pattern each quarter and before winter.
Related guides: planning for National Living Wage rises, bank holiday demand and the 30-day receptionist induction checklist.
Frequently asked questions
How do you work out how many receptionists you need on the phones?
Multiply calls in a half hour by average handling time in seconds, divide by 1,800, then divide by about 0.8 to allow for gaps between calls. Round up to whole people.
How many online requests should a practice plan for?
Practices received 153 online requests per 1,000 patients in March 2026, so a 10,000-patient practice should plan for about 1,500 a month and check its own figures[4].
Why are Mondays and the days after bank holidays so busy?
Demand that builds up while the practice is closed arrives on the next open day. NHS England notes call volumes are affected by public holidays and weekends[2].
How often should we review the reception rota?
Compare forecast and actual demand weekly for the first month, then review the pattern quarterly and before winter.
References
Care Quality Commission (2026). State of Care 2025/26: capacity in primary care (opens in a new tab) ↑
NHS England (2026). Statistical Note: Integrated Urgent Care Aggregate Data Collection (IUC ADC) July 2026 (opens in a new tab) ↑
British Medical Association (2026). Pressures in general practice data analysis (opens in a new tab) ↑
Care Quality Commission (2026). State of Care 2025/26: capacity in primary care (opens in a new tab) ↑
GOV.UK (2026). UK bank holidays (opens in a new tab) ↑
NHS England (2026). Network Contract DES 2026/27 Part A Guidance: Clinical and support services (Section 8) (opens in a new tab) ↑
NHS England and DHSC (2023). Delivery plan for recovering access to primary care (opens in a new tab) ↑