Designing career profiles

Challenge

The Skills Funding Agency is a part of the government that offers careers information, advice and guidance to anyone in the UK. Our team's purpose was to create a careers service that met the Gov.uk Government Digital Service (GDS) standards. As a user researcher, I helped the team to make decisions grounded in evidence so we can deliver a useful, inclusive and usable service.

Our service was at an alpha stage and we reached a point where, we wanted to understand what topics of information users will need to fully understand a career they're interested in. This then help us inform the design of a career profiles of our service.

Approach

Understanding what mattered

To answer this, I conducted a card sort with 20 users. We wanted to know how people's circumstances influenced information priority and so with the help of a recruitment agency, we spoke to people from varying circumstances: students, parents returning to work, recently made redundant, long term unemployed and people wanting a career change.

We created an initial set of 22 cards where each card had a specific aspect about a career. For example, entry requirements, typical working hours, career progression etc.

Activity 1:

Participants were asked to sort these cards into two initial categories - topics that are useful and those that aren't. To further understand the reason, participants described the label on each of the cards, the kind of information they would expect to find under it and how they'd use it to their benefit.

Cards that participants consistently categorised as 'not useful'.

Activity 2:

Participants then ordered the card headings in descending order of importance - starting with the cards they felt were most important and useful to them to the cards they felt were not needed and not important.

Cards ordered in descending order of importance

As a result…

22 cards reduced down to 10 cards

5 of the 22 cards were repeatedly kept aside by participants as they didn't find it useful at all. The remaining 17 merged into 10 due to overlaps and similarities in what they meant to them.

Priority of headings

A consistent order of headings was found which informed the order of headings in the first version of the prototype.

Labelling was refined to be clear and descriptive

E.g. Skills, interests and qualities required to Skills you'll need.

Demographic based statistics were not helpful and often misleading

During this research, participants consistently discarded these cards and ordered them right at the bottom. 13 participants didn't care about it and 7 misunderstood what they meant.

"What I am going to do knowing that 43% of this role are women?"

"Knowing that there are lot of people unemployed in this role is making me worry."

"Oh, so if it's hard to fill, may be it is a tough role that I shouldn't go for…"

Creating the design

With what we learnt, I collaborated with a content designer and a careers advisor to create an initial set of career profile pages.

Our service was informational and not transactional. So, we needed to make sure that the page's structure could help users easily find content that met their need.

To start with, we picked 8 different careers that hand ranging amount of advice. At a high level, a career profile page had:

• A career summary at the top to give readers a snapshot
• and headings in a specific order (informed by the card sorting) with content under it.

After creating the pages we immediately realised that users would need to scroll a lot to get the section of the page they're interested. They also might spend alot of time reading before realising that this might not be right for them.

So, we added list of headings at the top of the page that also served as anchor links. Users could scan the headings to get an idea of the content on the page as well as jump to that part of the page quickly.

I was able to create a HTML prototype using the GDS prototyping kit. This meant the design patterns used on the page have been tested to be usable or are being tested (to which our service could contribute to).

23% of overall GOV.UK traffic was through mobile devices. To meet user needs through smaller devices, the pages were designed to be responsive. The two-third/ one-third layout changes so that the one-third column moves to the bottom of the page. The service main menu at the top collapses into a drop-down menu and the page spacing is slightly tightened.

Testing

We tested the prototype with 10 users in a variety of career situations. One of our service principles is to create inclusive services. So we regularly include people with different disabilities in user research. For this specific round, out of the 10, 5 were people with visual impairments.

We conducted task based evaluations to learn about:

• How they'd find out about our service
• Learn about a specific career they're interested in
• The next steps they'd take

Here are some of the key insights:

Next steps

We plan to iterate on the designs specifically increasing the contrast and/or changing the grey, questioning the two-third one-third column layout and exploring the option use video content for richer stories.

Reflection

Stakeholders and demographics: Organisations run into situations where stakeholders are often attached to demographics. Sometimes, you change minds this by raising awareness, discussions with rationale and sometimes you need to debunk it by proving it. This was a key moment in our wider service team as we gained their trust in user research. Doing the hard work to make it inclusive: Watching people with disabilities trying to use our service was one of the best moments in my career. It helped me see their experience of using a service and what we needed to do improve it.