Legal & General - Stocks and shares ISA application
Making ISA investments easy
My role
I was contracted in for 6 months to lead the design of a complete end-to-end customer journey for online stocks and shares ISAs. The journey covered the pre-sales site, through to the product application, and my contract was extended for a further 3 months to deliver the Online-Self Service platform - where the users could view their investment in an online dashboard and carry-out transactions.
I was the Lead UX Designer the project and worked alongside a UI designer. I joined the newly assembled Agile delivery team to support with the MVP delivery and in addition to providing hands-on design and research I also provided guidance and advice on the approach, processes and ways of working within the team, e.g. helping the team to effectively integrate user-centred design practices into the Agile delivery framework.
Background
Financial services can be confusing, especially for people who are not that familiar with investments and how they work. Legal & General already had an online ISA application journey aimed at expert investors, or users with more advanced understanding of finance and what kinds of funds they were already looking for. However, they didn’t have an easy-to-use journey for finance novices. Legal & General wanted to launch and easy-to-use ISA application option for novice users, and allow them to keep track of and monitor the performance of their fund accordingly thereafter.
Project goals
Our goal:
Design and build an easy-to-use, completely digital ISA application journey for financial novices
Design and build an online self-service portal to allow users to manage and monitor their investment
Challenges with process
It was not possible to implement a full Double-Diamond approach on this project due to timeline restrictions. There was pressure to dive straight into the Develop and Deliver phase without fully defining the problem, as is often the case on fast-paced delivery projects. As a result, I knew I had to be pragmatic and to validate what I could as much as I could. One of the ways in which I was able to expedite design validation in the Develop and Deliver phase was to initiate a new practice within the team to conduct unmoderated user testing on UserZoom, which allowed us to get very rapid usability test data within each Sprint.
Assumptions
Prior to me joining the project, the team had just designed and gone live with a Pensions application journey. The assumption at the beginning was that there would potentially be various design patterns on the Pensions journey that could be reused and brought across to the ISA journey since they were effectively both stock and shares related and both required users to select their funds and risk level. My thoughts were that this was a fair assumption and it made sense not to reinvent the wheel, however it was also important to remember that they were still two different products with potentially quite different user groups so I still wanted to apply a fresh review of the whole Pensions journey to see what could work for the ISA journey and what could be improved improved for both.
Competitor analysis
I conducted a competitor analysis of some of the most relevant ISA application journeys online and went through the ISA application journey for as many companies as I could to see what we could learn from. Wealthify and Nutmeg appeared to have made their ISA calculators simple and engaging, with interactive sliders for engagement and terminology that utilised simple, non-jargon language. Nutmeg appeared to be quite good at guiding users along the journey by asking them certain questions about their risk appetite and then assigning them the relevant fund depending on their answers. Of all of them, I felt that Wealthify had the most compelling and easy to use ISA calculator. Since the ISA calculator was the pre-sales part of the journey I knew this would very important to improving conversion rate overall.
Expert review
Due to limited timelines, was not possible to test with users right at the start of the project, so to compensate for that I conducted an Expert Review of the Pensions journey. The Pensions product had only just gone live so we didn’t have enough user data to draw conclusions from how it might be working with users and also there was limited user testing done on it previously, so I kicked-off with an Expert Review to see what parts of the journey could be ported across to ISA and what parts could be improved. I used two methodologies for the Review:
1) Cognitive walkthrough - to start to understand step-by-step how easy/difficult it may be for users to progress on the path to conversion from pre-sales through to completing the application.
2) Heuristic evaluation - to give me a more general view on the overall website experience, including highlighting any inconsistencies in design pattern application.
Some outputs from the Expert Review of the Pension pre-sales Fund Selector can be found below:
Expert Review findings - Fund Selector
Redesign exploration
After I conducted the Expert Review and competitor analysis I followed-up with a design exploration of the calculator to figure out new ways to potentially improve the usability of the Fund Selector section. Since this was a key part of the pre-sales journey (and would significantly impact the amount of traffic going into the application flow) I wanted to make it interactive and engaging by using sliders and animations on the performance graph. I also wanted to get rid of any technical finance jargon that might put off newb investors. Unfortunately it wasn’t possible to implement the design below for MVP as it required significant development time, but it was put on the roadmap for a future release after it was shown to work well in usability testing.
User flow review
As part of the Expert Review I also conducted a review of the existing Pensions flow and potential ISA flow as it stood. I was able to identify two key areas for improvement in the flow that could potentially have a positive impact on conversion rate and help the user experience flow more naturally. The immediate design recommendations were to:
1) Move the ‘Input amount’ step next to ‘Select risk level’ since they were naturally associated.
2) Save a step in the journey by taking the ‘Email capture’ and combining it with ‘Personal details’ page.
Updated user flow
Below is the next iteration of the user flow, which was more comprehensive in terms of decision-points and unhappy paths. I used this document to liaise with the back-end developers early on in the project and to validate what was (and was not) technically feasible and also where all the back-end calls were happening. Together we were able to make various tweaks to the journey to optimise the flow, which I was quite happy with. When I proposed these updates to the rest of the team they immediately recognised the value and proposed updating the live Pensions journey in sync with our ISA build.
PI Planning
Not long after I joined the team we went into a Programme Increment (PI) planning session to map-out features for the upcoming sprints and position the features across the roadmap and to allow us to understand the dependencies across features. There was a challenge here as I didn’t have a massive amount of time to prep from the time of starting on the project to going in to the first PI session. As a result I had to make some assumptions and put together some hypotheses and a low fidelity user flow of the ISA journey. I also brought in insights from the competitor analysis, existing user insights from the business, generic company personas, and some assumptions and lessons learned from a parallel project that was recently delivered for the Pensions application product.
A-B testing on UserZoom
Due to the fast-paced nature of the project there was not always enough time (or budget) to usability test designs quick enough within each sprint. To work around this I conducted Guerilla user testing whenever I could, and towards the end of the project I was able to setup access to UserZoom, which made it possible to capture data from large numbers of users (working from home in their natural habitat) within a very short timeframe. This made it possible to share the findings with the scrum team in a quick turnaround and make several design iterations within each sprint when required. The design on the left was an example of an A-B test we did on UserZoom as part of an exploration into the best way to display charges to users.