Case Study: SEEK Choice Architecture
Research + Design + Delivery (2018)
The way options are displayed can have a huge impact on the choices people make. When it comes to your product offering, it’s imperative you design a choice architecture that will serve your business objectives while still providing a clear, honest experience that’s easy for your users to navigate.
In late 2018 I lead the redesign of our product choice architecture at SEEK, an enterprise B2B in the employment sector.
The existing design
Known internally as “ad selection”, employers posting a job ad on SEEK would see the following 3 ad type options:
A little further down the page, employers would have the option to upgrade their ad type to Premium:
The goals of the business were simple in theory, but challenging in practice; they wanted to increase adoption of Premium ads. There was pressure from the business to make a short-term, tactical play by placing our new preferred product (Premium) in the middle slot.
- The UI sat within the primary revenue conversion funnel and was highly sensitive. As such, it was imperative any new design protect the existing revenue
- The business was very wedded to the existing “3-choice” paradigm and had virtually no appetite to consider alternative approaches
- Due to the strategic importance to the business, many senior stakeholders were involved
- Legacy technology made the UI difficult and time consuming to work with (quick experiments and A/B test were not an option)
I kicked things off by reviewing any previous internal research I could find on ad selection. Given the sensitivity of the UI, there were many failed experiments and A/B tests to look over.
I was not content to consider only the tactical options — I wanted to explore longer term, strategic options in parallel. I completed a landscape review to explore what alternative approaches were being used by a variety of digital products and services. I socialised these with the cross-functional team and broader stakeholder group (samples below).
A previous team had already tested out a “4-choice” option for SEEK, but this had performed poorly in an A/B test.
Likewise, the same team had also tested an updated card design, which was more modern than the existing table view, but this too had performed poorly.
Additional design considerations
- Previous research showed employers didn’t like the “Recommended” label. Consideration would need to be given around how we could remove this but still highlight our preferred product.
- Historically SEEK relied on “forced-choice” approach at ad selection–Employers were required to make a choice to proceed. But was there an opportunity to increase overall conversion by employing a “default-effect” approach? (Pre-selecting an ad product by default)
Designing was highly collaborative and iterative. I ran sketch-ups with our cross-functional team and selective stakeholders, and socialised concepts in our design team pin-ups. To maintain transparency throughout the process, I created a UX wall within our team space and held informal weekly showcases. I provided design updates to the broader stakeholder group in our fortnightly steer co check-ins.
I created a tactical, 3-choice design as explicitly requested by the business.
In addition to the tactical play, I also designed a comprehensive strategic option, taking inspiration from Apple’s “add-on” (shown below).
I was keen to run a usability study on both designs to capture qualitative feedback. I was met with some resistance from stakeholders, who wanted to release the tactical design immediately with a pilot group and monitor the results. I was able to explain that the study would compliment the release, and that we should do both. I facilitated and analysed the usability study in our lab with customers, and recruited a number of stakeholders to observe and take notes. Insights from the research informed further design iterations.
Decision from the business
Ultimately, the short-term tactical play was given priority. It could be rolled out and potentially scaled within a couple of months, whereas the strategic play would require a year (or more) to release. Disappointing as it was at the time, I could understand the rationale. It’s important to be pragmatic about design. What we need to watch out for is when a business consistently chooses tactical design plays over strategic, leading to a degraded overall experience. We can’t win every time, but we absolutely need to win sometimes.
Outcomes + Value Delivered
The tactical design was released to a small pilot in market for monitoring. I worked closely with our customer service team to monitor feedback. Strategic product adoption doubled and the new design was scaled to 100% of users.
Whenever you have many stakeholders involved, design by committee is a risk (or worse, the dreaded HiPPO effect). A great way to work through this challenge is by booking regular, recurring research with users to drive the design decisions.