The goal

To provide an appointment booking system—Self-Book—that allows practices to invite patients to book an appointment without manual phone calls or letter posts.

Invited patients can also book an appointment with Self-Book which would be directly saved in the practice's electronic medical records (EMR).

The problem

The current process of booking an appointment is very process-heavy and requires a lot of manual phone calls to schedule a time. Also, when a practice tries to schedule a time with patients, many of them either don’t pick up the phone or miss the call, because they are not told when they will be called, which increases a lot of back and forth.

The process

This project has been broken down into 3 phases:

Telephone appointment booking
Face-to-face appointment booking
Batch face-to-face appointment booking

Telephone appointment booking

Practice invite flow

The practice inviting flow for telephone appointments is very straightforward. The practice would need to choose a slot type from their EMR so that patients could book into the chosen slot type for their telephone appointments. Once the slot type is chosen, the practice can start sending out a text message with a unique booking link to a patient.

A diagram of the practice invite flow — The whole user flow when a practice send out a Self-Book invite.

A demonstration of how a practice user send out a Self-Book invite to a patient.

Patient booking flow

When the patient receives the message, they can click the link and book a telephone appointment by choosing a date and a time. Once the patient confirms a time, we would check with the practice EMR to make sure the booking is registered.

5 screens of the patient flow. From left to right: Enter date of birth, Select a time of the day, Review the details, Loading screen and Confirmation. — Patient flow. From the left to right: Enter date of birth, Select a time of the day, Review the details, Loading screen and Confirmation.

Animated loading screen

While the checking process is usually quick, sometimes it can still take up to more than 30 seconds. In order to make the waiting process less frustrating and keep the patient stayed. I created a loading animation that:

reduces the perception of waiting time
explains why they need to wait
provides a pleasant user experience

The calendar animation on the loading screen.

Because the patient flow contains many moving parts, I mapped out the flow with the mockups in the Figma file to explain the logic of the flow. I also added notes to explain some additional screens for edge cases. This gave the engineers and stakeholders a clear picture of how the feature should work and made sure we didn’t miss any use cases.

A Figma file that I organised and mapped out all the designs for telephone booking flow. — I like to map out all the screens and put notes in the Figma file to explain the flows and behaviours.

After a round of usability testing, a few months of piloting with multiple practices as well as iterations. In March 2022, we rolled out this feature nationally. So far (29 October 2022), there are:

The stats of telephone appointment booking via Self-Book: 571 active practices, 66,489 Self-Book links sent, 72% of success patient booking rate.

"It's a lot easier..it has streamlined how we book in. It has saved time… certainly helped us minimise how much we have to do in reception when patients can do it themselves"
— Elouise, a receptionist from Wilmslow Health Centre

Face-to-face appointment booking

We started focusing on face-to-face appointment booking when the engineers were working on rolling out the first phase nationally.

The practice invite flow is very similar to the telephone’s, so we re-used most of the design components and patterns. This means that it would be easy for the users to pick up as they were already familiar with the design.

But for the patient booking flow, a practice can have more than one site in different locations and capacities. I designed an additional screen for patients to select the site first before choosing a time.

A screengrab of 2 practice staffs went through the prototype in the user testing sessions.. — We tested the face-to-face appointment invite design with 5 practice users.

2 practice staffs went through the prototype in the testing session. — Katie and I tested the design the second time with 5 memebers of staff at a practice in London.

3 participants tried to booking an appointment with the prototype in the testing sessions. — We tested the patient booking flow with 5 participants with a range of ages.

After a couple of rounds of user testing and piloting, we rolled out face-to-face appointment booking nationally in June 2022. Since then, we’ve been receiving feedback from them:

“That is incredible guys - GAME CHANGER - it’s what everyone has been waiting for”
— Prash Thurairatnam, a Managing Partner from Tudor Lodge Health Centre

“Oh great, I no longer have to keep ringing the patient now…We’re happy bunnies”
— Sarah Botherway, a Practice Manager, at Tadcaster Medical Centre

Batch face-to-face appointment booking

We were so excited about the final phase as it focuses on the biggest frustration of the appointment booking—recall. Booking an appointment has been frustrating for GP practices, particularly when providing ongoing care for patients with long-term conditions (LTC). In England, about 26 million people have at least one LTC, and 10 million have two or more LTCs. This leads to about 78% of all GPs’ appointments being with patients with long-term conditions.

Batch appointment booking allows practices to send out booking links to a group of patients at once without any phone calls or letter posts.

2 participants looked at the prototype in the 2 separate research calls. — Katie led the 1st round of research.

In the beginning, I worked with Katie, a researcher, to run the 1st round of the research. This aimed to validate our assumptions and figure out what are the key components to inviting a group of patients for recall.

Then, I created a few potential user flows and mockups and shared them with the team to gather thoughts and questions.

Those comments were really helpful as they pointed out different use cases and flows for different types of users that I hadn't thought about.

A screengrab of the board that the team left comments and question on the 3 potential user flows for Batch Self-Book. — I ran a session where we looked at 3 possible user flows with the team and gathered thoughts and questions from them.

A screengrab of me posted a recap and the next steps on Slack after the session. — I like to post a Slack message after the session to share the recap and the next step with the team. Not only this does make sure that everyone is on the same page but also allow people who can't make the session to understand the outcome.

Once I mocked up all the screens for the main user flow, I ran another crit session with wider stakeholders to share the designs, explain the rationale and gather feedback.

A FigJam board that contains many post-it notes on the design of Batch Self-Book MVP. — All the post-it notes after the design crit session with stakeholders.

A Figma file that I organised and mapped out all the designs for Batch face-to-face booking flow. — All the screens for the first Batch Self-Book MVP.

National rollout

On 5 September 2022, we nationally rolled out Batch Self-Book in England and Wales.

Over the 5 weeks, we have:

The stats of Batch face-to-face booking via Self-Book: 1,620 active practices, 5 million Self-Book links sent, 750,000 patients booked.

“I can’t make any of the times on your calendar but I think Accurx Batch Self-Book is the most amazing thing ever... It’s been amazing for our flu clinics, it has saved us many hours work plus paper ink and postage...It’s easy to use, I can’t say anything negative about it – I love it ."

“Everyone at the practice is raving about Self-Book its what they've all been waiting for, and patients love it!”

Next steps

We are now working on optimising user experience, patient uptake (currently 57%) and product scalability so that we can drive more adoption (currently 25%).

Enter password