Today, Credible users seek to understand their mortgage buying power. There are a lot of different ways users wish to tackle their home-buying journey. Some start Credible's application flow with the intent to compare mortgage rates with their property in mind or others may just be browsing to understand buying power prior to even knowing where or when they could realistically purchase a home.
Credible offers no guidance in the mortgage buying process. Our current pre-approval form flow asks users all required questions needed to gauge users buying power through assets and income. Personal details are collected to create a dashboard they land on once the form is complete. We find users confused and abandoning the dashboard after going through the above lengthy pre-qualification form experience.
Credible has what is called a pre-approval form flow. It's a fancy way of saying when a user comes to our website curious about mortgage options, they fill out a 3 part form containing their current property and hosing details, financial assets, and personal information. This is used to generate a pre-approval number based on available data. It is Credible's best guess without having to run a hard credit check. We do this to help users understand their potential buying power and use that to guide them through their search process.
The dashboard you land on once your form is complete contains 2 key areas. One is the number you are pre-approved for, and the second is the next action you can take depending on what you're looking to gain. There were 3 primary CTAs a user could select from. The number of users actually completing the pre-qualified rate received flow was 0.1%. This flow describes what happens when a user sees a mortgage option they like within our dashboard, and selects that to apply for a loan through our lenders. This touch point is the official kickoff of borrowing money for a mortgage.
An initial heuristic evaluation allowed us to hypothesize 2 core issues the dashboard currently had that impacted conversion and overall engagement.
There are 3 main CTAs: Generate a pre-approval letter, compare mortgage rates, and my offer was accepted.Because there were no distinguishing factors between a primary or secondary CTA, users couldn't be certain what the next step to take was. From a business perspective, all 3 actions were key to a user's journey, and all equally increased engagement and odds of loan conversion. It was difficult for users to contextualize their process and where they stood in that current moment, and tie that back to how Credible can best assist them. There was not enough context provided to help users understand the value of each 3 actions and what they helped the user achieve. Even further than that, it did not contextualize how each action would meaningfully impact the overall journey from end to end. The question was raised: why would they do any if they don't even know what they do?Taking a mortgage out is a life changing decision that may be too difficult to make online.It was difficult to base our success on the number of loan conversions through the dashboard because taking out a mortgage is vastly different than measuring financial success through the number of users who apply for credit card loans. Taking a human-centered process of buying a home and making that completely digital proved difficult for certain personas. If they had less understanding of purchasing a home, i.e their first time, they did not want to do it in a digital environment where they couldn't ask questions and what ifs. Second, we could target personas that have purchased a home in the past and knew how to navigate through the process, but we would need to layer in an understanding of our competitive edge over other digital offerings.
Initial kickoff involved performing a task analysis to understand touch-points users have throughout the flow with assets, personal information, budgeting, and other considerations they provide through the form. This helped me understand what type of questions a user might be asking themselves throughout the flow. This identified 3 stages for where users might be in their journey when they come to Credible to understand their buying power.
I took the task analysis results and applied it to our current flow to find any areas of opportunity. Within that I looked at the different stages users experience when looking to purchase a home. I hypothesized what questions a user might have within each stage and used that to inform potential features we could test with users. This helped identify where users would gain the most value in addition to areas users may need more guidance and information on.
Applying the mortgage journey outside of Credible to our in-product experience showed us where users are mentally when they encounter the dashboard actions as they relate to each buying phase. Features were based on actionable tasks like generating pre-approval letters when in the "search" process, comparing mortgage rates within our loan dashboard, and post-offer acceptance flows which led users to the final mortgage application directly through lenders.
Combining all of the above we came up with potential features we could offer within the dashboard experience that would allow users to build on any existing gaps they had on the mortgage loan process and how their purchase power translated to a home (square footage, bedrooms, bathrooms, yard area, etc).
I conducted research and usability testing with 3 groups of users. Our target persona for research was the loan officer since they deal with this exact experience daily with dozens of customers at a time, giving us insight into the biggest pain points and areas of hesitation. I also spoke with Credible site users and return home buyers.
The prototype we used in testing revealed three main feature asks that were grouped into high-level themes. Using the feedback provided, we broke down those key pieces of feedback into themes. This created a bigger picture for our customer asks, not just focusing on the specifics of feature desires mentioned in each call, but what the root of those asks is to holistically apply those concepts throughout the feature.
The main question through out research was what potential features could solve current user pain points experienced in our flow. We held brainstorming sessions and mapped out task analysis, user journeys, and our current flow to hypothesize what features could solve those pain points. Features were broken up into segments based on where a user might be within their home buying journey.
We learned that some users come to credible to understand their home buying power. They were curious to understand how much money they could be approved for and how that translated to a home. We referred to them as "dreamers" since they expressed their intent was low, and they wouldn't be ready to purchase anything for a few years, but liked to daydream about the potential.
The other group of users came in to compare rates to see which lender offered them the most competitive rates. These users were typically more informed and came in to the process either having been through our competitors flows or planned to go to competitors for comparison post Credible loan retrieval. Some feature ideas were focused on clarifying a home purchase journey or roadmap for users through visuals they could follow along with. This would show them where they are at currently and what steps are involved in order to fully "complete" the process before they can submit their final loan offer. Other feature ideas were based on levers that users could play around with to understand how much money they'd need to afford the house they want, and using data from aggregators to show average home size and details based on that mortgage loan amount.
I conducted 4 research sessions with loan officers to understand how customers ranked levels of importance in the home buying stages and which education pieces were crucial when making process decisions. When talking with return home buyers, we targeted cohorts that had already gone through a home purchase process to gain additional perspective on the difficulties, questions, and needs they had the first time around. We also wanted to understand what an ideal process for their second purchase would look like.
3 unmoderated tests were conducted via usertesting.com with 30 different customers. These unmoderated tests were utilized to hear their initial feedback on pain points within their journey and thoughts on the existing design. We used this method due to privacy laws regarding loan applications and private information within those.
Usability tests were performed to witness users' first-time interactions with different feature ideas to gauge, interest, usability, learnability, and perceived value. Research answered the open question of how to prioritize our features utilizing feedback on immediate needs. We then built out phases for MVP and the future of the product so we could ensure the foundation we were providing was in line with our long-term goals.
I implemented a survey on our dashboard page to understand what users wanted to gain from Credible or what their intent was for using Credible's service. The survey ran on the pre-approval dashboard. Users only saw this survey once they completed the form flow. It was live for 30 days reaching a total of 108 users.
The results were:
Usability tests were performed to witness users' first-time interactions with different feature ideas to gauge, interest, usability, learnability, and perceived value. Three main focus area in testing were:
After aggregating all the data and research insights, I utilized a brainstorm session with stakeholders to reach quicker decisions. This step of the process gave buy in from key stakeholders when moving towards implementation. We agreed that engagement and conversion is likely to increase if we offer experiences for users of all stages. By introducing the first step towards personalization, we wanted our MVP to be focused on 3 different dashboards depending on responses provided from the form flow regarding purchase time frame and current stage. Using these views we can recommend next steps based on those timelines and intentions.
The core learnings were based on budget constraints, exploration, and relevance.
Budget constraints
When inquiring on what is a make or break for users when considering a home purchase, monthly payments were most important. Users were seeking rates that fit into their comfort budget and our current product did not offer a monthly breakdown of cost.
Exploration
Users wanted to understand different outcomes and parameters that would impact their approach when buying a home. Users wanted to see individual scenario based questions answered through the tools. 44.4% of users were just looking to browse rates without intent to pursue further.
Relevance
Users wanted the experience to contain relevant data and pathways that applied to their current stage. Each user needed education around different stages of the process to guide them on the next best step.
