Decreasing user drop-off rates at LendingTree

Mocks of conversational

Collaborators

Design mentors
Product manager
Engineering
UXR

Timeline

June 2021 - Aug 2021
(3 months)

Role

Interaction Design intern

LendingTree is an online marketplace for loans, credit cards, insurance, and more. We want to give you the best deal when shopping around for financial products.

Snapshots of the current Mortgage Purchase and Refinance flow

My main internship project was to improve LendingTree’s main upsell entry point - the Mortgage Purchase and Refinance flow - to decrease user drop-off. Many users will only partially complete this form before leaving the experience.

My goal was to research and design solutions to reduce this drop-off.

How might we…
increase the completion rate for our mortgage purchase form?

One of the biggest problems facing LendingTree is user drop-off. LendingTree's main source of revenue is product form completions, so having users fill out the entire form is essential to the business.

I focused on the issue of trust and confidence in the form, as these play crucial roles in influencing user decisions. The form asks for a high amount of PII, or personal identifiable information, which can make users feel hesitant or nervous.

Logos of LendingTree’s main competitors: Morty, Better & Rocket Mortgage. Young Alfred and Hello Fresh are both indirect competitors.

Understanding our competitors

LendingTree’s main competitors: Morty, Better & Rocket Mortgage. Young Alfred and Hello Fresh are both indirect competitors.

Across five competitors, I looked for elements of social proofing, copy tone, and visual progress indicators that served to build rapport between the business and the user. My findings showed:

  • Competitors grouped similar questions together on a single screen, and used a multi-page flow to not overwhelm users upon first sight.

  • Competitors provided a progress tracker and had a clear value upon completing the form.

  • Competitors are transparent as to why personal data is being collected.

  • Competitors used social proofing by displaying business partners’ logos to establish their credibility.

Ideating two solutions: step-by-step and grouped questions

Wireframes of the step-by-step solution

Using the insights from the competitive analysis, I designed wireframes for a step-by-step experience of the Mortgage Purchase and Refinance flow. The solutions in this design include:

  1. Including visual tools such as help icons and progress indicators to build a relationship with the user

  2. Welcoming, personalized copy based off of user input

  3. Reducing the number of questions from ~20 to 12 for a faster, more direct flow

I also used the insights from the competitive analysis to design another set of wireframes, this time with grouped questions. The solutions in this design include:

  1. A thematically grouped question experience based off of how sensitive the data is

  2. Smart conversational popups, illustrations, and motivational header copy to build a relationship with the user

  3. Also reducing the number of questions from ~20 to 12 for a faster, more direct flow

Wireframes of the grouped solution

With the help of rounds of feedback from cross-functional partners, I transformed my wireframes into high-fidelity mocks to present to users for testing.

Hi-fi mocks of the grouped solution

Hi-fi mocks of the step-by-step solution

53% of users view the new forms as trustworthy

76% of users said the popups and conversational copy were helpful

A randomized audience of ~300 users were presented with a set of 12 questions around expectations, attitudes, and level of trustworthiness for each design. User testing across both designs showed a generally positive attitude towards trustworthiness, UI elements like popups, and copy.


Talk to users early in the process

I was set on getting a set of high-fidelity mocks in front of users for testing, but this pushed me to rely on internal feedback rather than feedback from actual users while designing. In the future, I’d like to bring in user research earlier on so that users are involved in each stage of the design process.

Think outside of the box

Early on, I was challenged to explore a range design solutions by my design mentors. This resulted in numerous jam sessions with my design team that helped build team morale and spark creativity in my designs.

Reflections

Back to:
YMail ↩