Online Claim Application

The brief

Over half (51%) of injured workers who attempt to make a new claim for worker’s compensation fail to successfully complete WorkCover Queensland’s online application form.

This results in multiple online attempts, or deferring to another channel to register the claim, such as phone, fax or mail.

Of the online applications received, the data provided is often low quality and not to the right level of detail WorkCover needs to register the claim, requiring follow-up phone calls to gather the correct information.

Incorrectly filled claim forms and claims made through analog channels have a significantly higher administrative burden on the customer support staff, add delays to registration and decision timeframes and may require the injured worker to tell their story multiple times.

Existing claim application form

The claim form is structured over four sections (pages) with page level validation, it is not accessible to WCAG standards and it is not device responsive. The upfront guidance does not encourage workers to complete the form alone, despite this being an option. As such, the guidance isn't inclusive of workers with an employer who does not support the claim.

Research activities included: an end-to-end user test and Microsoft Desirability test with six Queensland workers, analysing the usability metrics through Google Analytics, gathering existing customer insights from customer feedback initiatives, and workshops and interviews with WorkCover customer support staff.

The claim form is shown below, highlighting the negative attributes identified during desirability testing and customer verbatim from user testing.

Customer verbatim
"Not polished... it looks more like a back-end system"

Customer verbatim
"Not as bright and shiny as I'm used to with other insurers like NRMA"

Customer verbatim
"I don't know why some of these options are here"

Customer verbatim
"This is not common language"

Customer verbatim
"Ughhh, I've gotta call my boss and find out. I'm annoyed. If this question was optional it would be okay, but it is mandatory"

Problems identified

Through the research activities the following problem areas were identified:

Users find the language cold and too technical with industry terms and jargon
Users want guidance to understand what they need to do now and what happens next
There are unfair questions that workers cannot complete on their own without help from their employer

There are expectations of modern interface and technical features that weren't being met
If users don't understand a question they will guess what to put in, or they will call WorkCover
Workers don't want to be made to tell their story more than once

Problem statement
How might we empower injured workers to successfully complete an online claim application improving the data collection and user experience?

Design approach

Improving experience through applying principles and patterns to reduce cognitive effort

1. Adopt a beginners mindset
No two people and no two claims are the same. Everyone's level of understanding of the WorkCover process, their digital literacy levels and their cognitive abilities are different. By assuming everyone is starting a new claim as a novice we can be more inclusive.

2. Clear and informed
Workers are guided through the application with upfront information before starting the form, detailing what is required to make a claim and the estimated time to complete, providing a brief overview of the sections in the application form and what happens after submission.

Once submitted the claimant receives next steps guidance, providing an overview of what is happening now and what needs to happen before WorkCover can make a decision on whether to accept or deny the claim.

There are some frequently asked questions included here, providing answers to the questions worker's commonly have while waiting for an application outcome.

3. Lead the way
Users are guided step-by-step through the form, one question at a time, with conditional logic applied to the questions. Additional circumstantial questions are only revealed when triggered. By applying this logic the form could be short and straight-forward for simpler injuries and more detailed for the complex injuries that require more information.

4. A visible record of progress
I added an interactive stepper, showing a visible record of progress throughout the application, giving a clear indication of what they have completed and what is still to come.

5. Choice and control
Users can go back and make changes to their answers whilst completing the form. After completing the form users will see a summary of all the information provided, they can go back and make changes if desired before submitting.

Improving data quality

I changed the language and tone to an active voice and applied the Kincaid-Flesch Grade 8 reading level. I added supplementary guidance and instructional content with each question to clearly explain what the user needs to do and provide definitions for industry terms.

These changes aim to remove the ambiguity in the questions and provide clear guidance so workers can provide better quality data at the level of detail WorkCover needs.

Removing barriers preventing completion

I removed the unfair questions about the employer's policy which were a barrier for an injured worker to complete the application without help from their employer.

This empowers injured workers to complete the application on their own and ensures WorkCover is asking the right person for the right information at the right time.

By collecting just the employer's name and contact information WorkCover has enough information to either link the claim to an existing policy and/or contact the employer for any outstanding information required to link the claim to the policy.

Modernise the product

The solution design was aligned to WorkCover's target state architecture, including a React front-end and an AWS cloud back-end.

As part of the front-end uplift we were also developing a new design system for WorkCover, which was implemented as part of this project.

The user interface is accessible to WCAG 2.1 Level AA.

The design took a mobile first approach and is completely device responsive.

Test and iterate

I tested the design changes with six Queensland workers, replicating the original usability test with an end-to-end run through of the application form prototype. Once again, I asked the respondents to use the talk out loud method, taking note of any issues.

At the end I ran another Microsoft Desirability Test. The results found a 100% positive desirability sentiment, compared to the baseline measure which was 60% positive and 40% negative.

From the testing I identified some issues with the flow that could be simplified and some button actions that could be clearer, from there I made some small tweaks to the design and the prototype was very well received.

Customer verbatim
"The whole process overall was a lot easier and simpler than I thought... when you think of dealing with insurances, a lot of people put that off"

Customer verbatim
"The whole process wasn't overwhelming"

Customer verbatim
"Very logical and straightforward"

Customer verbatim
"I didn't expect it to be so simple"

Customer verbatim
"The wording is very good, it's not corporate. It's simple, easy to understand"

Customer verbatim
"I think it's a shift from what it used to be, I think of Government as factual and by the rules, whereas this relates back to you and more like we're here for you kind of thing"