Lendio approval calculator
Increased approval acceptance by 42% by building a calculator that moves data, and the customization process, out of email and into the hands of users.
Overview
Lendio is a small business loan marketplace, making it easier to discover the right financing to support your business. When businesses are approved for a loan, they have a range of possibilities. In fact, a typical $50,000 approval has over 1.2 million unique customizable options to choose from. This should provide business owners the flexibility to dial in the details and find the best option for their situation. However, the current process only gives businesses insight into one option at a time. This prolongs approval times and obscures important details.
I led the research and design to establish a new connection with lenders and build a customizable calculator to give users the flexibility to see the whole approval, resulting in an increase of approval acceptance by 42%.
My role
Product designer leading research, discovery and design.
The team
1 Designer, 1 Product Manager and 3 Engineers
Timeline
6 months · 2019 - 2020
Company
User interviews revealed a broken connection
After talking with users and observing the current process, we learned that we had a broken connection; all of our data was being passed through email and manually pasted into our system. Not only is this extremely inefficient and unsustainable, but it became clear it was often inaccurate. And when dealing with financing, the details need to be right.
Problems
700-1000 approval emails are sent to a single inbox daily.
It is the job of 2-3 approval administrators to scour the email and manually copy and paste data into our own system.
Our staff doesn’t trust the data. Because data is manually entered, mistakes are inevitable. Most of the time data is good, but one bad encounter hurts the whole system.
The data is static and unchangeable. Approvals are customized 90% of the time, but the only way to do so is through a lengthy back and forth process that reinforces our reliance on emails.
“How can we expect to show business owners this information when we don’t trust it ourselves?”
Taking a small step towards the vision
In the future, we want to build a fully featured experience for small business owners, where they can apply for, customize and compare a range of approvals. After our initial research we knew we were a long way from this, but we wanted to start!
Our goal was simple, get out of email.
This was the first step in building a new data foundation that is consistent, fast, and reliable. For this project, we decided to focus on our internal users interacting with approvals to test our ideas and slowly build up towards our vision.
Building an experience for user needs
Because data has been gated through emails, our internal Funding Managers act as the layer between lenders and businesses. I spent multiple sessions interviewing and watching them interact with end users to see how they work. And I learned that their workflow is expansive and varied; they are using custom spreadsheets, proprietary calculators, sticky notes, and everything but the data in our system.
Although the methods differed, the process each user went through was the same. As new loan approvals come through, users first glance to compare, then review the details to determine fit, and finally, interact with the data to customize to the perfect option.
With this insight, and after a few rounds of collaborative sketching and wireframe testing, I had a design the team felt confident about using as a foundation. Approvals would be represented as individual cards that displayed all of the basic information to compare at a glance. Each card could be expanded to reveal the "calculator" allowing the user to explore and customize.
Sample of all my sketches
Final wireframe exploration used for testing
Testing the "real thing"
I had been testing the approval calculator at every stage of the process. But to be honest, I don't think I always knew what I was testing for. Showing sketches and early wireframes of the layout to users helped me discover a direction. Testing simple prototypes ironed out usability wrinkles. That is all necessary, but our direct feedback still consisted mostly of head nods, thumbs-up, and exclamations of "looks great!"
Stepping back, I noticed a pattern. Those users who said “looks great” often followed up by saying, "...if it works."
"I could use this... but only if it is going to be right.
"It's a cool toy, but will it work?"
"I've been burned in the past... I would double check the email."
The design had gotten to the point that I was confident that users could use the feature. Yet, there was still risk in whether a user would use the feature. We had to show that it worked.
So we built a quick live prototype! To test, we manually input real approval information into our prototype to power the calculations. I would then reach out to the assigned staff and send them a link to the calculator. This gave us a chance to test the real thing. Real data with real borrowers.
We built a working prototype that users could customize real approvals with.
Test results
88% of users were able to successfully customize the deal. "Those sliders were awesome!"
75% of users felt confident in calculations.
100% of users confirmed that all the calculations were correct after double-checking.
We learned that we needed to show users that they could trust this new product. It would likely be a slow process of change, but we were encouraged by the results of our testing. We saw excitement grow and trust regained.
The New Approvals
Compare the options
Some approval offers are better than others. I designed the cards to highlight the most important information so users can make quick decisions across multiple approvals to determine which ones are worth looking into.
Review the details
When an approval looks promising, users can expand it to see all of the details. I designed this to be compact and stay in view of the other approvals. This provides users the option to focus on the specific approval or continue to compare details across at a more granular level.
Customize to find the right fit
With familiar interactions, users can customize all the available options of the approval to find the right fit. At launch, we were trying to win back user trust and were competing against years of learned behavior. Knowing this, I made every effort I could to design an experience that felt comfortable, familiar and fast. It wasn’t enough to just be good, it needed to be better than what they currently do!
Outcomes
— At launch —
Integrated with 1 lender and had 2 more in conversation.
78% adoption of the new calculator among our staff. It was fun to see people interacting with the new calculator around the office.
No negative impact to approval acceptance rate.
— 8 months later —
Integrated with 3 lenders with many more onboarding.
~250% increase in daily usage of the calculator.
42% increase in approval acceptance rate for integrated lenders.
We were really excited about the results we were seeing from this project. The calculator was faster, more accurate, and led to better outcomes for our customers. Lendio works with over 75+ lenders, so we had a ways to go, but we were well along the path of getting this experience out of email and into the hands of users.
