Key takeaways
- Patients now move between phone, website and NHS App within a single request, so friction usually sits at the hand-offs between channels.
- Map one journey at a time, from the patient's first contact to the follow-up, using real cases rather than the process manual.
- A 90-minute workshop with reception, a clinician, a manager and a patient is enough to produce a usable map.
- Fix the points that cause repeat contact first: they waste staff time and frustrate patients in equal measure.
Why journeys need mapping now
The way patients reach practices has changed quickly. In the 2026 GP Patient Survey, 30.8% of patients said they last tried to contact their practice online, up from 22.4% in 2025 and 16.8% in 2024[1]. Online requests per 1,000 patients rose 67% between March 2025 and March 2026, after practices were required to keep online tools open in core hours[2]. Those hours run from 8am to 6.30pm[3].
Source: Ipsos (GP Patient Survey provider for NHS England), 2026[1]
Show the data
| Period | Value |
|---|---|
| 2024 | 16.8% |
| 2025 | 22.4% |
| 2026 | 30.8% |
The NHS App adds another route. It reached 25.7 million users in 2025/26, and 76% of GP practices could be contacted through it, with around 2.4 million online requests starting there each month[4]. A single patient might read the website, start a request in the app, get a text and then ring to check. Each step is a chance to lose them.
Ease differs by channel
No channel is clearly easier than the others at national level, which is why local mapping matters. Survey results also show that experience depends on which route people used: patients who used the NHS App were the most likely to report a good overall contact experience (75.7%), and those using a different website or app the least likely (67.5%)[1]. The ease ratings below come from the same survey[1].
| Channel | Patients finding it easy, 2026 | Typical hand-off to watch |
|---|---|---|
| Practice website | 57.6% | Form submitted, then no clear acknowledgement |
| Phone | 56.9% | Told to go online, then calls back when stuck |
| NHS App | 54.3% | Request starts in the app, reply arrives by text |
Choose one journey and define its edges
Mapping everything at once produces a wall of sticky notes and no decisions. Choose a single journey that is common and causes complaints, and agree where it starts and ends.
- GP: a working-age adult with a new, non-urgent problem, from first contact to the outcome of the appointment.
- GP: a repeat prescription that needs a medication review.
- Dental: a new patient from enquiry to completed first course of treatment.
- Private clinic: a self-pay enquiry from first call to follow-up after the first consultation.
Define the end point as the patient sees it. A journey is not finished when the appointment is booked; it is finished when the patient knows what happens next and it happens.
If the team cannot agree which journey to start with, count complaints and repeat contacts by type over the last quarter and pick the biggest. Starting where the pain is greatest wins support for the second and third rounds, because staff see the benefit quickly.
Running the workshop
Before: gather three real cases
Pick three recent patients who took this journey, one smooth and two difficult. Pull call logs, online requests and messages with identifying details removed.
First 15 minutes: set the frame
Agree the journey, the start and end points and the rule that the map shows what actually happens, not what the policy says.
Next 30 minutes: walk the cases
Lay out each step on a long sheet: what the patient did, which channel, who or what responded and how long it took. Use a different colour for waits.
Next 25 minutes: mark the friction
Everyone marks points where the patient had to repeat themselves, chase, switch channel or wait without knowing why.
Last 20 minutes: choose fixes
Score each friction point for impact and effort, pick no more than three to fix, and name an owner and a date for each.
Invite a receptionist, a care navigator or administrator, a clinician, the practice manager and, ideally, a member of the patient participation group. If an AI Employee such as Ava answers some calls, show it on the map as its own channel with its hand-off points to reception.
The map template
| Stage | What the patient does | What they need to know | Common friction | Evidence to check |
|---|---|---|---|---|
| Find out how | Searches, reads website or calls | Which route to use and when | Website and phone message disagree | Website, greeting, posters |
| Make contact | Calls, fills in form or uses app | That the request was received | No acknowledgement; form times out | Call logs, form analytics |
| Wait for response | Waits, sometimes chases | When they will hear back | No stated response time | Repeat calls about same request |
| Get the outcome | Receives slot, advice or signposting | What to do and when | Message unclear or arrives on a channel they did not use | Message templates |
| Attend | Comes in or takes a call | Time, place, who they will see | Late running unannounced | Clinic delay records |
| Follow up | Waits for results or next steps | Who will contact them, how and when | Patient has to chase results | Results-related calls |
Fill the template row by row with the three real cases, writing in the patient's words where you have them. Leave a cell blank rather than guessing; blanks show where you lack evidence and give someone a task before the next session. Timings matter: write down how long each wait lasted, because a two-hour silence and a two-day silence are very different problems.
Where friction usually hides
- Channel switches the patient did not choose, such as a phone caller sent to a form they cannot complete.
- Silence after submission: no message saying the request arrived and when to expect a reply.
- Repeat questions, where the patient gives the same details to the form, the receptionist and the clinician.
- Dead ends after closing time, when the website says to call and the phone says to go online.
- Follow-up gaps, where results or referrals rely on the patient to chase.
The quickest way to spot these is to count repeat contacts: two or more contacts from the same patient about the same request within a few days. Every repeat contact on your map is a friction point with a measurable cost.
Pay particular attention to patients who find it hardest to use any channel. Office for National Statistics data show adults renting from a council were 46% more likely, and those renting from another social landlord 51% more likely, to report difficulty contacting their GP practice than outright homeowners[5]. If your three cases all come from confident, digitally active patients, add a fourth from someone who struggled, such as an older patient without a smartphone or a carer calling on someone's behalf. Their path through the system is usually where the sharpest friction sits.
Deciding what to fix first
A good workshop finds more friction than any practice can fix in one go. Use a short set of criteria so the choice is not simply whoever argues hardest.
- Safety: could this point delay care for someone who needs it urgently? Fix these first, whatever the effort.
- Volume: how many patients pass through this step each week?
- Repeat contact: does the friction cause patients to call or write again?
- Effort: can the fix be made by changing wording, settings or rotas rather than buying something new?
- Ownership: is there a named person who can make the change within a month?
Plot the shortlisted points on a simple grid of impact against effort. High-impact, low-effort fixes, such as aligning the phone greeting with the website, should be done within days. High-impact, high-effort items become a project with a plan and a date for review.
Example: one journey, three fixes
Example (fictional): a practice mapped the journey for working-age adults with a new problem. The cases showed patients submitting an online request at lunchtime, receiving no acknowledgement and phoning at 8am the next day to check. The team agreed three fixes: an automatic acknowledgement stating when the patient would hear back, a matching line in the phone greeting and a daily check that every request had an outcome by 6pm.
The practice tracked repeat contacts for the same request before and after. Within six weeks the morning queue was noticeably shorter, and reception reported fewer calls that started with 'I sent a form yesterday'.
After the workshop
- Type up the map within two days while memories are fresh, and share it with everyone who took part.
- Track one measure per fix, such as repeat contacts or time to acknowledgement.
- Check the fixes with a few mystery patient contacts after a month.
- Map the next journey once the first fixes are in place.
- Review the map whenever a channel changes, such as a new phone system or online tool.
Journey maps also show where new technology belongs. If you are considering automation for calls, our article on how patients feel about AI answering the phone covers the design choices that affect trust.
Frequently asked questions
What is patient journey mapping?
It is a structured way of laying out every step a patient takes, across every channel, from first contact to follow-up, so a team can see where people get stuck, repeat themselves or wait without information.
How long does a journey mapping workshop take?
About 90 minutes for one journey if you bring real cases prepared in advance. Larger redesigns need several sessions.
Should patients take part?
Yes, ideally. A member of the patient participation group spots assumptions staff miss, especially about wording and timing.
Which channels should be on the map?
Every route patients actually use: phone, website forms, the NHS App, text messages, email and the front desk. In 2026 almost a third of patients last contacted their practice online[1].
References
Ipsos (GP Patient Survey provider for NHS England) (2026). 2026 GP Patient Survey results released (opens in a new tab) ↑
Care Quality Commission (2026). State of Care 2025/26: primary and community services (opens in a new tab) ↑
British Medical Association (2025). Pressures in general practice data analysis (opens in a new tab) ↑
NHS England (via GOV.UK) (2026). NHS England annual report and accounts 2025 to 2026 (opens in a new tab) ↑
Office for National Statistics (2026). Experiences of NHS healthcare services in England: September 2026 (opens in a new tab) ↑