Digital Product Designer



Australia Post is the largest postal service provider in Australia. The help and support portal at provides information to customers as well as modes of contact with the support team.


The majority of customers seeking help and support were using the call centre instead of online forms, resulting in a higher cost to serve. 

The existing help and support platform was inflexible and could not be scaled or optimised. In particular, mandatory fields in the online forms had been changed or were no longer mandatory. 

In order to address these issues, the platform needed to be migrated to Salesforce. I was tasked with redesigning the help and support portal to help customers find the right information and modes of contact.


  • Salesforce developers
  • UX researcher
  • Customer experience manager
  • UX writer
  • Business analyst
  • Aus Post developers


10 weeks for design, build and testing of the "lift and shift."


  • Helping facilitate co-design
  • Sketching and whiteboarding
  • Wireframing
  • Prototyping
  • Helping plan user research
  • Helping facilitate user research
  • Helping synthesise user research results
  • UI design


The existing help and support portal

The existing help and support portal

When I was tasked with the project, we had a kick off meeting with the core team and stakeholders in the business where the customer data and insights from the support team were relayed to us. I collated the findings from this discovery research and used it to identify user needs which could be used to provide a basis for empathising with them. 


  • Most of the customers who were defaulting to the call centre could have had their issues or enquiries self serviced online, however they were dissatisfied with the online experience or couldn't find the solution to their enquiries through the portal.
  • Customers were using incorrect online forms.
  • Online forms had a high drop out rate due to forced authentication.
  • Forms were very dated and had no UX process underpinning them when they were built. Therefore they were not smart or contextual, were hidden, and had too many unnecessary fields.
  • The categorisation of help topics was confusing.
  • The overall experience was inconsistent due to legacy designs being mixed in with the new branding.
  • Customers were at a heightened sense of frustration and disappointment to begin with, and the convoluted and difficult process exacerbated these feelings.

I framed the problem as:

How might we help customers quickly find the appropriate information and contact modes to improve their satisfaction levels and reduce costs to serve?

After the co-design workshop, the core team and I pooled the best ideas together and whiteboarded a solution.

After the co-design workshop, the core team and I pooled the best ideas together and whiteboarded a solution.


Together with the UX researcher, we held co-design workshops with stakeholders around the business to frame the problem and ideate on solutions. These stakeholders included people from the contact centre, people from MyPost (forced authentication), analytics and Salesforce. We asked people to research good examples of help and support portals and explain why they were effective designs. We created a list of the most common user problems and framed them as scenarios that could confront a variety of potential users. People split into groups, chose the scenarios, and then sketched ideas that would help to resolve them. The group then voted on these potential solutions.


  • Design the information architecture (IA) and user journey so that users are funnelled to the correct online forms.
  • Simplify language and add tooltips to explain confusing or difficult form fields.
  • Communicate time frames and expectations to users.
  • Notify customers more frequently
  • Use conditional logic so that form fields are contextual and relevant to users.
  • If forced authentication is required, then communicate its benefits and include login and signup in the same window.

One of the wireframes created in Sketch

One of the wireframes created in Sketch


After our whiteboarding session I did some more sketching in my notebook, and then created various low fidelity wireframes using Sketch. 

Mobile prototype created in Axure. Feel free to interact with the prototype. Please click the arrow in the top left hand corner to close the sitemap.


After more meetings and iterations, I turned my designs into interactive mobile and desktop prototypes created in Axure.

View desktop prototype

(opens in a new tab)


The UX researcher and I conducted user testing on the desktop and mobile prototypes.

  • We held 12 one-to-one interviews, which included moderated usability testing, Treejack testing and an ideal journey exercise.
  • 6 of the participants had used the website or called Help and Support in the last 3 months.
  • User testing included an "ideal" state and a "lift and shift" state, across both mobile and desktop.
  • For Treejack testing, participants were asked to complete 10 tasks based on top Help and Support content areas to see if the new IA and labelling was clear and easy to navigate.

“I don't really know where to go. I would go to an email or LiveChat, or go to
the nearest post office and ask them."
Jillian, 31-40 years
User testing analysis

User testing analysis


Customers could not easily find out how to raise a case online or seek help & support

All participants failed to locate the “Help and Support” option in the global header on the Australia Post homepage. Just over half of the participants were not able to raise a query without being assisted in the sessions.

The forms in the "lift and shift" designs forced customers to create an account before they could log their issue, causing frustration, and /or drop out

The vast majority of participants were annoyed to be forced into creating an account to log an enquiry. Some said they would choose to call rather than creating an account.

There were a few terms that were unclear and caused participants to hesitate

The two phases: “Your item is still on the way, and…” and “Your item has been delivered, and…” were seen to be hard to read or process due to the “and”. Nearly half of the participants were confused between the options of “you think its missing/lost” and “it’s delayed”.

Over half of the participants did not click on the ‘Contact us’ button when asked to raise a query

When questioned, they said they thought “Contact us” would lead to telephone numbers.



Participants were asked to map our their ideal flow for seeking help and support.

This exercise resulted in findings that confirmed the hypothesis uncovered in the co-design workshop that customers needed more notifications, including:

  • Customers want a default notification setting which they can customise easily.
  • Opt in and Opt out capability on a case by case basis is important.
  • Email, SMS and phone are the key ways in which customers want their query resolved.


We gained some valuable insights from the research and testing.


  • Focus on copy and content to ensure forms deliver the best contextual experience.
  • Make the account creation requirement as smooth as possible.
  • Ensuring the service message is correct.
  • Ensuring the design allows for action above the fold, on both desktop and mobile.


Ensure that help text is helpful by both explaining why we ask for the content and giving examples to customers (e.g. description text).


After synthesising the user testing results and prioritising which changes to make, I worked on the UI design in Sketch using the Australia Post branding guidelines and UI global components. I created the illustrations following the Australia Post illustration guidelines. I used Zeplin to hand over my designs to the developers.


Despite having a lot of customer insights from the contact centre, it would have been very beneficial to conduct more detailed discovery research at the beginning. Unfortunately, due to the tight deadline, we did not have an opportunity to do so. 

Looking back, I've since learnt some ways to do lean and quick discovery research, which would have been better than solely relying on insights from the contact centre. Without more detailed discovery research, I had made some assumptions which ended up being caught in user testing, but could have been discovered much earlier on if we had done discovery research, or earlier stage testing with low fidelity prototypes.

I also didn't have an opportunity to work on the IA and user flow as much as I would've liked. Together with the UX writer we'd created a spreadsheet and user flow, but I would have used card sorting and created more detailed user flows to ensure that the IA worked as best as it could.

Form accessibility was another issue that I had the opportunity to learn about through this project. I needed to ensure the forms would follow best practice for accessibility, so I made modifications such as adding borders to radio buttons and adding hint text outside of the form field.

Unfortunately I'd left Australia Post before I had the chance to find out what the results of the new design were.