Background:
Members are facing issues finding the right kind of loan. Once they do, the application process is tedious.
Pain Points:
Too much information is packed in small amount of space. It’s confusing & overwhelming for users.
Solution:
I designed an entire consumer lending application with neatly defined loan categories and an easy application flow.
Duration:
3 Months
My Role:
Solo UX Researcher+UX Designer. Collaborated with analytics, business, developers and PMs.
Success:
-
Quantity of online Lending applications completed: Increased by 23%
-
Drop-off rates for Auto loans: Reduced by 2%
-
Click-through-Rate (CTR) changes for Exclusive offers: Increased by 7%
Defining Problem
and Scope
The brief given to me was to completely redesign the current "Consumer Lending" product line on RBFCU's website (Desktop, mobile browser) as well as for the upcoming native app.
My first task was to understand:
-
What are users currently doing/saying?
-
Why are they facing certain issues?
-
How to fix their problems and up to what extent?
​
​
WHAT
To understand what are users currently doing- I decided to use a platform my employer uses called "FullStory" which lets the UX researcher see click stream analysis, drop-offs, etc.
WHY
To find out why they are facing certain issues RBFCU member email surveys as well as in-person interviews were conducted.
HOW
To understand to how to fix these issues I decided to use mixture of qualitative and quantitative methods such as online usability testing and task analysis.
What are the
Current Pain Points?
Using FullStory clickstream analysis and multiple internal stakeholder reviews, I realized there were four primary issues that my design strategy needed to address:
​
-
Members facing difficulties finding the right kind of loan.
-
Members not filling out the application properly or dropping off in the middle of the process.
-
The most revenue generating lending product- Auto Loan is deemed to be the most confusing one.
-
“Exclusive Offers” something the bank provides to members with good credit score, is not getting expected response.
​
One of the later stage card sorting-brainstorming sessions
Pain Point: Loan Selection
Using quantitative research method such as FullStory gave me an idea about the broad problems regarding the main landing page.
-
Large quantity of vehicle loan categories were confusing for potential auto loan customers.
-
Members who were not visiting the page for auto loan were facing issues locating the right kind of loan.
-
Some of the loan categories were redundant or were leading to empty states. It was ending up in drop-offs or rage-clicks.
Current Lending Products Page
Brainstorming Sessions
Pain Point:
Mid-application Drop-offs
Based on requirements captured from user surveys and in-person tests+interviews with three members, some of the following issues were noticed:
Most participants were annoyed that they had to calculate on their own -how much amount they will be paying monthly.
Currently all helpful reference links are grouped in this sidebar. In-person test participants couldn't complete related tasks and expressed it was confusing for them. Insights from online platforms mirrored these results.
​
-
Participants were frustrated by long forms that needed to be filled, with multiple fields and drop-downs on a few long web-pages, especially on mobile devices.
-
Also, majority of participants noted that they were not happy with the visual look of the whole application.
Pain Point: Auto Loan is Especially Confusing
It was made clear during internal stakeholder reviews that Auto loan was the most revenue generating and most widely used lending product. And due to it's larger monetary value it has the longest flow that includes adding co-applicants, details of vehicle required, details of trade-in vehicle, etc.
If the member had a desired or trade-in vehicle, the form was becoming quite lengthy.
Even if a member has a pre-approval code, at no point they were required to manually enter it.
Loan Purpose contained about 10 options in the drop-down, some of which were confusing users due to verbiage. Some of the options were unnecessary.
Current process doesn't take enough information about co-applicant. The bank has previously faced some identity theft related issues due to this.
Pain Point:
Exclusive Offers
-
In the current flow once the member clicks on "Apply for a new loan" in main navigation, this pop-up appears. It shows any special offers a user is offered based on their credit.
-
Participants ignored these offers in majority. When asked why some blamed it on the content's pop-up ad-like nature.
-
Some participants also noted that they ignored it as there was no cause and effect behind it's appearance.
Coming up with Actionable Design Goals
Discoveries
-
Members understand that loan application process can be long, but they are not thrilled about it's length and complexity.
-
Whenever the member has to do any kind of mental calculation, probability of drop-off is higher.
-
The current design has less quantity of clicks. But that does not necessarily equal to great user experience.
Deeper Insights
-
The long application forms need to be broken up into short, friendly, conversational flows. Needs to start with broader, easier questions, gradually going into more complex content.
-
Avoid making users calculate anything on their own that will take them out of the flow. Also, make sure they have to manually type as little as possible.
-
In the process of simplifying the flow, number of clicks may increase. As long as it's intuitive and takes the flow forward with no redundancies, it should work.
Design Solutions
Based on the insights drawn, I came up with design solutions for the issues discovered.
Solution for Loan Selection
-
By teaming up with stakeholders from consumer lending and business analyst, first I unpacked all the loans we provide and how they can be distributed with no redundancy.
-
Once an initial decision was made, I implemented that in early wireframes and checked if the flow was logical and intuitive to users via UsabilityHub.
-
I implemented certain changes in a paper-based prototype to save time and be more flexible with the modifications.
One of the design review sessions with the team
Main landing page
Personal loans page
-
After a few design and testing iterations, this design was made for the final product.
-
Only the main six categories were displayed on the landing page. Once the user chooses the kind of loan they want, they don't need to see all the other children categories at all.
-
In early requirement capturing phase, it was found that there were some loans not classified properly. All these loans were confirmed to be some form of personal loans by our business analyst. So, they were grouped together accordingly.
Solution for Mid-application Drop-offs
As discussed in the "Pain Points" section earlier, some of the major reasons discovered for drop-offs were- users' need to break the flow due to own calculations, all reference material being in one section, long length and formal nature of application forms, etc. Let's see how these issues were tackled.
Calculator Design
-
During user interviews, members made it clear that they would like to see an estimation of how much amount they need to shell out on a monthly basis.
-
A paper mock-up was created and reviewed by the whole team.
-
Based on stakeholder feedback a high fidelity version was made.
-
It was subjected to quantitative testing where certain issues came up.
-
Such as some members not wanting to pay down payment, some not having much idea of their credit score or not interested in seeing a monthly payment (especially for loans of smaller amounts).
-
Based on test insights users were given option to provide certain information.
-
In the A/B test performed in later stages, the version that provided the option to see loan term in both month and year units along with an estimated end date was majorly favored and hence was implemented in the final prototype.
To avoid calculation related drop-offs, I designed our own calculator within the application.
Reference Material
After detailed discussions with the business analyst, I divided all reference material, assigned it to corresponding loans and provided access to it from within each loan's flow itself.
Reference material for Personal Loans
"Form" Format of the Application
-
As discussed earlier, members understood that a loan application will be long but did not appreciate the tedious forms with stern questions. A quick Competitive Analysis revealed that some of the players in consumer lending market space are opting for ways to make loan application process as friendly as possible.
-
I conducted Heuristic Analysis of our whole application. Based on insights, divided the content according to it's nature which included major categories such as -Contact information, Employment/Income, Accommodation information, Loan purpose/sub-type information, etc.
-
I transformed the typical label-entry field format into more conversational Q&A flow. Made sure to design content in a way that users can choose from given options and have to manually enter information only when absolutely necessary. Also tested three order of occurrences and chose the best received one.
Original form
Solution for Auto Loan
Changing conceptual navigation of entire loan application to conversational style positively affected Auto loan flow as well. But this longest and most widely used product flow needed some extra attention.
Selecting Auto Loan Purpose
-
With the help of BA and consumer lending team, I changed verbiage for some of the confusing loan purpose options. And removed a few options that were not used anymore.
-
During brainstorming, the team realized that some of the individual options can actually be nested based on their nature. This helped me clean things up in the design as well.
Auto Loan Purpose Redesign
Desired & Trade-in Vehicles
-
For Trade-ins, members can either directly work with RBFCU or with one of our vendors Carvana.
-
Based on discussion with my colleague in analytics, we get a lot of calls from members as current page doesn't make it crystal clear that they can opt for Carvana and even if they do get it, they have to get out of our flow and search for the said vendor's website.
-
I made sure to provide vendor related information on a pop-up modal, so the members are well informed and are not taken out of our flow.
Trade-In Redesign
Co-applicant Page Redesign
Adding Co-applicants
-
Just taking co-applicant's SSN has left the bank with some legal troubles in the past and the design needs to solve that problem.
-
Data analytics suggested that majority of times the co-applicant is already an RBFCU member and joint account owner. But the possibility of them being a non-member was not completely refuted.
-
So I decided to auto-populate with the list of the member's existing joint account holders along with an option of adding a non-RBFCU member in the application.
-
Also based on internal stakeholder input, I added income information inquiry for member co-applicant and more information requirement for non-members.
Solution for Exclusive Offers
-
Both exclusive offer related issues that came up in users survey were solved by making some changes in the location of the content.
-
Instead of them popping up when user clicks on "Apply for new loan", they were added as an independent category on the loans landing page.
-
Apart from that, since Business Unit really wanted more visibility for these offers, an "Exclusive Offers" section was also added on the "Account Summary" (main landing) page.
Exclusive offers new flow
'Consumer Lending' Full Process Flow
Measuring Success
& Next Steps
Some of the metrics the analytics team was tracking quarter after the new design went live:
-
Quantity of online Lending applications completed: Increased by 23%
-
Drop-off rates for Auto loans: Reduced by 2%
-
Click-through-Rate (CTR) changes for Exclusive offers: Increased by 7%
-
Incoming call volume for content that's provided on the website: No improvement noted in first quarter.
-
Overall click-stream analysis (It's a multi-factor metrics- was still being analyzed).
-
Overall changes in Consumer Lending turnover in the long term (It's a multi-factor metrics- was still being analyzed).
​