SAI CHAVA

Impact
50% less
less time spent onboarding a new client. Clients start receiving care sooner, and staff have increased flexibility
The Challenge
When I first heard all their documentation was analogue, I thought to myself this is a great opportunity but a massive challenge. After interviewing care supervisors I found that between 35 and 40% of the questions in these reports were repeating across the pre-assessment, initial assessment, risk assessment and care plan. This paperwork takes up around 60% of the time, time that care staff would rather spend on care.
“
In my experience, care workers prefer to find their own ways around tasks, like writing on the corner of their blouse sleeves.
Dr Nicholas Hine, Senior Lecturer in Computing
30% less
less repetition in documentation.Â
Lower cost to serve each client, and fewer chances for data entry errors
53 pages
Into one flow. Increased operational efficiency through digital transformation, expanded B2B scalability
Approach
I was the lead UX designer on the project, working with MyHelpa’s CEO and with care supervisors and managers at Tamar Homecare. They stayed involved through regular feedback cycles, so the design was shaped by the people who would use it.


Learnings that shaped the project
Care workers want technology that’s no fuss
Many care staff are over 45 and would rather spend their time on care than learning new software. Basic tasks had to be easy on a tablet, without new features to learn every week.
Paper is still the norm
Rostering and scheduling are often digital, but onboarding and risk assessments are still mostly done on paper, which is where most of the time goes.
The repetition in assessment questions couldn't simply be cut
Each section asks similar questions from a different standpoint, so removing them wasn’t an option. The answer had to be making them faster to complete.
Voice notes were the most appreciated feature in testing. Care managers said this is a feature they didn't know they needed.Â
Result
People in care often work long tiresome hours. Hearing the person behind the profile brings empathy back into a long, administrative task.
Why
Added voice notes from the patient/client to their profile, helping staff members identify patients that are new, or circulating.Â
What I did
Reflecting patients as humans, not numbers. Â


03
Care Plan page showcasing the patient profile with their voice notes
Voice notes variationsÂ
What I did
Added a status bar showing how complete each form is, with navigation that mirrors the paper sections supervisors already know.
Busy, often older users needed to see where they were at a glance and recognise options rather than remember them.
The status bar was the second most appreciated feature in testing.
Why
Result
Packing all the sections back into a familiar, heuristic UI


02
Care Plan page showcasing a main navigation on the left, and section list on the right
6 component states of the navigation bar
Sampled an AI-assisted fill built on a semantic network that links answers to the sections they affect. If a client’s main need is old age and isolation, sections such as mental capacity, behavioural risk and mobility switch on automatically, and the care plan reflects them.
The repeated questions all served a purpose, so the best way to save time was to stop supervisors filling in sections that didn’t apply.
About 30% less repetition, and up to half the time saved.
What I did
Why
Result

Unpacking the mess: Breaking down all the four assessments into sections and understanding inter-relatability.
01
I challenged the hierarchy of the documentation that hid multiple sections inside the umbrella term 'assessment'. Each patient profile loads a customised base template that contains sections based on their pre-assessment inputs (a patient with diabetes would have the fall risk section auto-enabled). The rest of the sections can still be accessed manually, giving care supervisors control and flexibility.






Process Gallery



Design decisions
​
Testing and Iteration
I validated the Figma prototype containing 20+ components with care supervisors through remote, task-based usability testing using UX Tweak.Â

Client set-up and navigation between assessments performed strongly. The care plan emerged as the main source of friction: participants valued its clarity and detail, but its layout slowed task completion. That finding set the priority for the next iteration.
Outcome
The proposed design reduced the administrative burden of onboarding, giving care supervisors more of each visit with clients/patients.Â
“
I think you have done an amazing job with the detail you have and how quick you have grasped the sector. Well done!
Tracy Greenan, CEO, MyHelpa
