< Back to folio


UX design and research
for integrated self-service


Yarra Valley Water


Objective:
• Evaluate the current experience of using the self-sevice portal
• Understand the customer value proposition of YVW
• Evaluate design prototypes against the value proposition
(job stories / jobs to be done)

Deliverables:
• Customer value proposition
• User stories / jobs to be done / user requirements
• Establish Design principles
• Wireframes and User-tested digital prototype

Research methods:

Evaluate & compare
Heuristic evaluation
Landscape review
Desktop research
Output: Benchmark current state of self-service

Determine value drivers
Interviews (cross section of customers)
Card sort for jobs to be done (JTBD)
Quantitative survey
Output: Archetypes, CVP, JTBD

Reimagine users' experience
UX design (mid fidelity wireframes)
IA card sort
Single dot, first click test
Output: Future state prototype

Test with users
Prototype testing
Usability testing
Output: Validation of research and design

UX testing:

Single dot, first click testing
We developed multiple design options for various features and jobs to be done. A survey was conducted to understand most popular designs. 20 frames were tested with 22 participants. Optimalsort was used for this testing.

One on one interviews
The purpose of these interviews was to gather feedback on mid-fidelity wireframes for the self-service portal. Research findings, key jobs to be done and landscape review informed initial designs and concepts.

Qualitative research (usability testing interviews) was conducted with 10 participants across 2 rounds of testing, 1.5 weeks apart (with 5 participants each round) to inform and test design iterations. Interviews were one hour long, moderated remote face-to-face.

Participants included a mix of male and female, aged between 25 and 62. A mixture of Yarra Valley Water customers and non-customers. They were from varying cultural backgrounds and represented a range of technological preferences and familiarity.

What we tested
User interaction (what they do) and their thinking (why they do do it) around self service functions such as make a payment, download bills, change contact details, digital metering onboarding

How we went about UX testing remotely
• Recruit via Askable
• Task based usability testing via video call, where customers can share their screen and we can see them face to face (visuals of them doing the tasks)
• Activity called Build your own dashboard to see which features and details are important to users (the order of jobs to be done)

Build your own dashboard activity
We used Miro for this activity - drag and drop

What we were testing
What information is most important to users
The order of information tiles
The display of usage
Which tiles should be present

Discovery
Quick links with alerts were popular and were often used as a form of navigation.
The circle/dial was a winner with the option to toggle between Litres and Dollars.
Water saving tips was considered occasionally but was also thought to go on the usage page.
The usage circle dial was preferred first - our original assumption was "Billing". As this tile combines Billing and Usage in the one place, it makes sense.

UX prototype

Below are a selection of scenarios tested with users.

Overview & single property scenario
As a customer who has multiple properties, all with Yarra Valley Water, find the balances of your services with multiple properties, which one requires payment - how would you go about it?

Overview discovery
How much information is the right amount of information on each property tile.
What actions do you expect to be able to do from this page.
What would you expect this page to be called.
Would you like to label the tiles yourself.
• Address as the identifier
• Relationship to property
• Account balance
• Due date
• Payment method

My account discovery
Users understood the information in the usage graph and liked the ability to toggle between usage and dollars.
The circle was understood as a position in time.
This format suits both the Set and Forget users where the dashboard displays just enough information and with the ability to drill down to more detail, this caters to the Relational users.

Usage Scenario
As a customer you would like to add another person to your household to manage your target realistically. General observations about the Usage page were also captured.

Discovery
For users who like a lot of detail, this is exactly what they expected to see, including hourly usage and comparisons.
The option to toggle between usage in Litres and usage in Dollars was very well received and easily found. Default would be to not have the service charges breakdown showing but the option to be there for easy consideration with access to more detailed information.
The breakdown list by month at the bottom of the screen was also well received, a quick comparison view.
Users who felt comfortable with their water usage weren’t overly interested in comparisons with neighbours, users who weren’t quite sure but keen to do better were interested in this feature. Recommendation, this wouldn’t be the default setting, have these options be optional/clickable.
Users found it easy to update their settings from this page.
The word Behaviour had mixed responses. Most users understood what would be under the tab but didn’t like the word. Recommendation is for us to change this.

Information architecture card sort
To understand which jobs and tasks should be grouped together to help users navigate the self-service portal.
The outcome: The grouping of jobs to be done informed the menu system for user testing sessions.

Single dot, first click testing
To understand which variation of the functional design elements users prefer and why. Follow up questions to understand their selection.
The outcome: The preferred designs were used to inform functional elements of prototypes used during user testing.

Usability testing interviews
To test if users understand how the system works, what it does and how they interact with it. This was done with an interactive prototype where users were presented with scenarios and tasks to complete. Two rounds of usability testing were completed.
The outcome: Validated UX design grounded in user testing and feedback.