Discover potential diagnoses through user routines

Discover potential diagnoses through user routines

The Trends feature in the MySense app leverages data science to provide health insights. Helping customer identify potential diagnoses sooner for their loved ones by detecting individuals’ routine changes.

Key contributions — Leading and executing functional designs, team collaboration, research & workshops

image

Main Objectives

1. Reduce at least 20% of customers everyday workload on everyday tasks

2. Reduce unnecessary human errors on data research and analysis

3. Improve product data structure and speedup the loading time by at least 50%

4. Increase the volume of business acquired

Outcomes

~73%

Significant speed up on users everyday tasks

~40%

Reduction on return tickets from users systems (Error or additional inquiry)

~60%

Significant Loading speed improvement

3

New businesses onboarded

What We Are Trying To Solve

User spent significant time reviewing individuals’ health data. Managing the workload and human error became challenging when more individuals onboard.

High Cognitive Load

Users felt overwhelmed by the amount and complexity of data they had to analyse every day.

Product Performance

Users were frustrated when comparing health data because of the platform’s slow loading time.

Human Error

The level of complexity of daily tasks lead users feared of making mistakes.

image

Principles & Strategy That Drive Business Impact

Work Smarter Not Harder

The solution should give users focus and clarity. Delivering the health insight in a straightforward solution.

How it works

Keep It Simple

Unify data presentation and reduce the need for extra processing by users.

How it works

Speed It Up

Improve our data structure to provide a faster and smoother user experience for their daily tasks

How it works

Insights & learning drive our execution

With the learning on the existing health insights, which presented users with a overwhelming amount of data without clear structure. We defined our improvements.

 Inconsistent ❌ Indirect ❌ Improvement Needed

1. Simplify & Unify

16 Data attributes 11 → 3 Unified modules

We simplify the health data presentation from 11 down to 3 unified modules to summarise individuals’ health insights.

image

Counting Countable data that continuously accumulates from zero.

image

Range/ Rate The measurements that fall between two specific values (ranges).

image

Time Displaying time-based data (specific times or periods during the day).

2. Focus & Clarity

image

Exploratory Data vs…

The previous system display data across different charts, where users need to explore and identify unusual health pattern. This approach is time-consuming and increases users workload.

Explanatory Data

Instead of make them search for Wally, we bring them Wally.

We simplify the logic and deliver the pattern change to users directly as Trends. It create focus so users can manage their workload by ignoring unnecessary and distracting data.

3. Reduce Friction

60% Performance Speed improved

With Data Science and Back End Team amazing work, we managed to achieve a significant faster loading speed with the new Trends approach.

image

X-Functional Collaboration

Cross-functional Workshops

Prioritising tasks for different teams based on difficulty and potential technical limitations

Design Execution

Define core function specification based on the design direction

Proof of Concept

Evaluating our decision with fully functional product performance

Qualitative Testing

Measuring the impact against our defined objectives

Delivery & Impact

MySense Trends Reduced friction in data delivery by focusing on the most significance health pattern changes. This drives a lighter workload and faster performance.

User Experience

✅ Simplify & Unify

Using 3 consistent modules to deliver 16 data attributes. Highlighting specific changes in health data.

✅ Focus & Clarity

Applying Trends to users’ daily tasks can simplify their workload by reducing unnecessary and distracting data.

✅ Reduce Friction

Significant faster performance provide better and lighter experience on data delivery.

image

Challenges Data analysis is complex, unexpected scenarios happens, data collection could went wrong…

A/B Testing

We test the Trends along side the original function. With qualitative interview and feedback sections, we measure the impact on both users and business.

Testing Methodology

~73%

Significant speed up on users everyday tasks based on 30 individual tasks.

”The new approach helped me prioritise tasks effectively. I allocate more time to those who genuinely require attention in stead of mapping data on everyone.”

~38%

Reduction on return tickets from users systems (Error or additional inquiry)

“With the new approach, I can work on the require actions based on the Trends insight and take note right away in stead of copy and paste into a work document. It would definitely reduce human errors.”

60%

Significant Loading speed improvement.

“It is so much faster!”

Learnings

We’ve identified a few areas that we need to address in the next version, as per users’ feedback.

There are Improvements to enhance the functionality while also introducing some specific features that we could incorporate into the product.

Data Relation
Data Comparisons
Data Timeframe

*All user data and quantities in this case study have been adjusted in accordance with individuals’ data protection policies and the company's NDA.

More case studies

👩‍✈️ Lufthansa / Cabin System →

👩‍⚕️ NHS / Testing & Feedback →