Understanding User Needs and Pain Points for Redesign
The task: Understand current user needs for to inform the redesign of the PropertyGuru [PG] mortgage affordability calculator.
My role: Leading the research, with close collaboration the product designer.
Deliverables: Consolidated report for high level stakeholders; First hand learnings to be used in design; Improving designer knowledge and confidence in running user research in the future.
As a launchpad for a Finance offering (in the form of a mortgage marketplace), PG had decided to revamp the most used finance page on the website - the mortgage affordability calculator. This required understanding user needs and pain points with current affordability calculators.
Before and after. Screenshots of the calculator results page before (left; August 2019) and after (right; Dec 2019) the redesign.
Research Design
Research needs. This was no ordinary redesign. The previous calculator had not been touched for 3 years. Additionally, users were not consulted in its design. On a business level, the redesign also represented a first step towards a bigger service - PG Finance. This meant that the stakes for this research were unusually high, and we had very little knowledge of the user needs.
Desk research. The designer and I built up our research scope on previous work and on competitive analysis. As a result, we came in knowing the big picture needs + context of the loan seekers, with some ideas that could be validated, and a largely blank canvas when it came to the task analysis of an affordability calculator.
Recruitment. It was important that respondents had some need for a calculator. I decided to recruit home seekers, who have used a property calculator in the last 6 months.
Methodology. The study had to accomplish two things - 1) understand user needs and pain points with property calculators and 2) provide inputs to improve our existing calculator. As such, I designed the research as an in-depth interview combined with a usability test on our own calculator. If there was time, we also performed a usability test on a competitor’s calculator. Due to budget and logistical constraints, we conducted these as remote interviews.
Competitive research. We examined other calculator designs (e.g. MoneySmart, pictured above) and consolidated them into Figma.
Summary of methodology and sampling. As shown in the final report
Remote Fieldwork
BlueJeans installation. To facilitate screensharing and recording, we made it very clear during recruitment that the BlueJeans app needed to be installed on their primary device (either smartphone or desktop). We also had fallback options in the form of WhatsApp and normal phone calls, which we used to walk respondents’ through the process of installation. Additional time was budgeted in our planning for technical difficulties.
Screenshare issues. While we did provide support during the call, we conducted extensive pre-testing and troubleshooting before the call to cover any possible scenarios with screensharing. One thing to note: when screen sharing on BlueJeans, it was impractical to monitor a video feed of the respondent’s face at the same time.
Ensuring a dynamic interview. Remote research can often lead to respondents feeling less engaged. This was overcome by spending more time and effort into establishing rapport, and by increasing the frequency and directness of think out loud prompts (e.g. “what are you thinking about?”, “why so quiet?”, “why did you stop?”, “why are you scrolling up and down so much?”).
Screenshare + Usability. Screenshot from the video recording of the usability test on a smartphone.
Interview notes. We kept our notes as close as possible to the respondent’s talking points - using exact quotes as much as possible.
Analysis and Reporting
Organising our thoughts. After each interview, the designer and I would always discuss our thoughts and note them down. Referencing our notes, we consolidated our thoughts on post-its, organised as an affinity map later. This allowed us to progressively structure our thoughts.
Template creation/ report sketching. To get the report structure, we sketched out a flow of the report, slide by slide, on paper. This acted as a way for us to write the report while ensuring all points were covered. It also allowed for a strong collaborative spirit, with each of us building on each other’s ideas.
Additional data sources + video usage. As a result of this collaborative exercise, the designer and I also managed to pull in other pieces of data for us in the report - previous quantitative research, Google Analytics data. We also included a short video clip of the usability test to illustrate our points.
Affinity mapping. Photo of our affinity map, with extensive notes to help our thought process.
Report sketching. Some sketches of the report, before shifting to Google Slides.
Outcomes
Design. The affordability calculator was greatly simplified, in a way that answered to the users’ basic needs. Some improvements: (1) Input + output fields were reduced, while those that remained were explained in greater detail, and/ or were clearly labelled as optional; (2) The design also facilitated “playing” with the input numbers; (3) The layout was made more readable through larger fonts and a more intuitive layout.
A/B Testing and launch. The new calculator was a clear improvement over the previous iteration - with completion rates improving by 90%.
Fintech business. The new calculator was launched in Dec 2019, just before the soft launch of PG Finance as a mortgage marketplace. The mortgage calculator results subsequently act as a channel to funnel customers into PG Finance’s service.
Redesigned affordability calculator. Completion rates improved significantly after the deployment of the redesign.
Reflection
Designer-Researcher collaboration. This was the second time I’d collaborated with this designer, and it was already a greatly positive experience. The combination of each of our strengths and knowledge helped to elevate the research. In future, I can also see potential for me to play a more consultative role, allowing the designer to lead the research.
Data collaboration. Combining data from Google Analytics and past quantitative research with the qualitative research allowed for a much more convincing report, with greater applicability of the insights. In future, I’d like to go deeper into website behavioural data to complement my research, to create more depth to the insights.