Key takeaways
- An embargoed appointment is a slot held back from booking until a set time, usually to protect same-day or short-notice capacity.
- Releasing every embargoed slot at 8am recreates the 8am scramble; releasing in waves spreads demand across the day.
- Online requests now arrive throughout core hours, so release schedules must treat phone and online patients equally.
- Use a small number of embargo types with clear names, and record when each one opens in the slot itself.
- Measure when slots are booked and how fast they go, then adjust the waves, not just the number of slots.
What 'embargoed' means
An embargoed appointment is a slot that exists in the appointment book but cannot be booked until a set time. Most clinical systems let you mark slots as embargoed until a particular hour on the day, or until a number of days before the appointment. Practices use embargoes to stop routine requests using up capacity that is needed for patients who become unwell later, and to stop slots being booked weeks ahead and then wasted through DNAs.
Embargoes are powerful because they shape when patients try to book. If every same-day slot opens at 8am, patients learn to call or log in at 8am, and the practice gets its whole day's demand in the first half hour. If slots open in stages, demand spreads out and the phone queue shortens.
Why release timing matters now
Same-day demand is the largest single block of GP work: 45% of appointments in England were booked for the same day in the year to August 2026[1]. The government has made ending the '8am scramble' an explicit aim of the 10 Year Health Plan[2], noting that booking an NHS appointment still often means queueing on the telephone[3]. Release timing is one of the few levers a practice controls directly.
The other change is online demand. Since practices had to keep online consultation tools open through core hours from October 2025, online requests per 1,000 patients rose 67% between March 2025 and March 2026[4]. By March 2026 there were 153 online requests per 1,000 patients a month[5], and 9.6 million online requests in that month alone, against 5.7 million in August 2025[6]. A release plan built only around the phone line no longer fits how patients ask for help.
Source: Care Quality Commission, 2026[6]
Show the data
| Item | Value |
|---|---|
| August 2025 | 5,690,082 |
| March 2026 | 9,617,628 |
Common embargo patterns
| Pattern | How it works | Main risk |
|---|---|---|
| Single 8am release | All same-day slots open at 8am | Recreates the morning rush; slots gone by 8.30am |
| Waves through the day | Same-day slots open at, say, 8am, 10.30am and 1.30pm | Patients must know the waves exist |
| Triage-only | Same-day slots never open for direct booking; a clinician or care navigator books them after triage | Triage capacity becomes the bottleneck |
| 48-hour embargo | Slots open two days before the appointment | Creates a second peak at the release time |
| Rolling horizon | Routine slots always open up to a fixed number of days ahead | Long horizons increase DNAs |
| Clinician-held | Slots kept for follow-ups the clinician books | Unused slots must be released in time to be filled |
Most practices combine two or three of these: triage-only or waves for same-day, a short embargo for next-day slots, and a rolling horizon for routine and review appointments.
Design a release schedule that spreads demand
Map when requests arrive
From your telephony and online consultation data, plot requests by 30-minute interval for four weeks. Most practices see a sharp 8am peak by phone and a flatter curve online.
Map when slots are booked
From the appointment system, plot when same-day slots are booked. If most are gone by 9am, your release is concentrating demand.
Split same-day capacity into waves
Hold back a share of same-day slots for later in the day. A common starting point is half at the start of the day, a quarter mid-morning and a quarter early afternoon.
Release to all channels at once
When a wave opens, it should be equally available to phone, online and walk-in patients, so no group gains an advantage.
Keep a duty buffer
Keep a few slots outside the waves for the duty clinician to use for urgent cases that come in at any time.
Release unused clinician-held slots
Agree a time, such as noon the day before, when clinician-held follow-up slots that have not been used are released to the general pool.
Fairness between phone and online
The fastest way to undermine trust is for phone callers to find that online users took every slot, or the reverse. Decide the principle and apply it in the system: either a single queue where requests are dealt with in order of clinical need regardless of channel, or a fixed share of each wave for each channel. A single queue based on need is usually fairer, because it does not reward whoever is fastest on the phone or the keyboard.
Patients who cannot go online, including many older people and people in insecure housing, must not lose out. Make sure the phone line can offer the same slots as the online form at the same time, and that reception can submit an online request on a caller's behalf.
Example: a three-wave schedule
Example (fictional): a 12,000-patient practice has 70 same-day clinical slots on a typical Tuesday. It releases 35 at 8am, 18 at 10.30am and 12 at 1.30pm, and keeps 5 for the duty GP. Online requests and phone calls feed one triage list. Before the change, all same-day slots were booked by 9.10am and 40 callers a day were told to ring back tomorrow. After four weeks, the last same-day slot goes at around 2.45pm and the number told to ring back falls to about a dozen.
Waves only help if patients know about them. Say on the phone message, website and online form that 'more same-day appointments are released at 10.30am and 1.30pm'. Otherwise everyone still calls at 8am.
Bank holiday weeks and winter
Release schedules that work in an ordinary week can fail in a short one. The day after a bank holiday carries the demand of two days, and the waves fill faster. Plan those days separately rather than relying on the standard settings.
- Increase the first wave on the day after a bank holiday, but keep the later waves so the afternoon is still covered.
- Open a small number of next-day slots on the working day before a long weekend, so patients with non-urgent needs can book ahead of the break.
- From November to February, move some routine capacity into same-day waves, and review the split weekly while respiratory illness is high.
- Tell patients in advance on the website and phone message when the release pattern will differ.
Give reception a one-page sheet showing the release times for each slot type, updated whenever the template changes. If the person answering the phone cannot say when the next appointments open, the waves lose most of their value.
Embargoes for next-day and routine slots
Embargoes are not only for same-day work. A short embargo on next-day slots, opening at noon the day before, gives patients a route that is not the morning rush. For routine GP and nurse appointments, a rolling horizon of two to four weeks keeps the book open without letting slots be booked so far ahead that patients forget them. Review and recall slots can open further ahead, since patients are invited directly, often by self-book links.
- Next-day GP: open at noon the day before, bookable by phone and online.
- Routine GP: rolling two weeks, with a share online.
- Nurse and HCA: rolling four weeks for bloods and routine nursing.
- Reviews and recalls: rolling six weeks, booked by invitation.
- Procedures: booked by clinicians, no embargo.
Pitfalls to avoid
- Too many embargo variants, so reception cannot tell when a slot will open.
- Embargoes that lift at odd times nobody knows about, so slots open and sit unused.
- Releasing next-day slots at 8am, adding to the same-day rush.
- Forgetting to update embargo settings when a clinician's session changes.
- Clinician-held slots never released, so they expire empty.
- Opening every embargo online first, which disadvantages patients who phone.
Measure and adjust
Track three things weekly: the time the last slot in each wave is booked, the number of patients turned away or asked to call back, and the share of embargoed slots that end up unused. If a wave is gone in ten minutes, it is too small or too well known; if it is still half empty at 4pm, it opened too late or is not advertised. Adjust the size and timing of waves before changing the total number of slots.
Embargo rules belong in the wider template; see designing your appointment book template, same-day versus pre-bookable slots and offering online booking.
Frequently asked questions
What does an embargoed appointment mean at a GP surgery?
It is a slot that cannot be booked until a set time, such as 8am on the day or noon the day before. It protects capacity for patients who need care at short notice.
Why are all appointments gone by 8.30am?
Usually because all same-day slots are released at once. Releasing them in waves through the day spreads demand and gives patients who cannot call at 8am a fair chance.
Should embargoed slots open online and by phone at the same time?
Yes. Releasing to all channels together, or managing all requests in one queue by clinical need, keeps access fair for patients who cannot go online.
References
British Medical Association (2026). Pressures in general practice data analysis (opens in a new tab) ↑
GOV.UK (Department of Health and Social Care) (2025). Fit for the future: 10 Year Health Plan for England (opens in a new tab) ↑
GOV.UK (Department of Health and Social Care) (2025). Fit for the future: 10 Year Health Plan for England (opens in a new tab) ↑
Care Quality Commission (2026). State of Care 2025/26: primary and community services (opens in a new tab) ↑
Care Quality Commission (2026). State of Care 2025/26: capacity in primary care (opens in a new tab) ↑
Care Quality Commission (2026). State of Care 2025/26: capacity in primary care (opens in a new tab) ↑