UX Design • 2017


  • Understand why users do not report hires (how BuildZoom makes money).

  • Re-design the core revenue-driving experience, taking into account insights gained and the existing UI library.


  • I primarily led this project during my 3-month internship, with mentorship from my boss (Head of Product).

  • BuildZoom is an online marketplace that connects contractors to homeowners.

  • Tools Used: email analytics software, MixPanel, FullStory, User Interviews, Sketch, Invision, KeyNote.


  • Insights from ~5 user interviews.

  • High fidelity prototype for mobile and desktop platforms, boasting a brand new user flow and UI. Justified via A/B testing.


“This modal is ugly, redesign it.”

At first glance, I thought this would be a simple UI touch-up. However, after probing deeper into the problem, I realized that the project was an actual UX overhaul. 

The original modal I was asked to redesign.


I used email analytics software, MixPanel, and FullStory to create hypotheses I would later test.

This modal was clicked into from an email that was sent to homeowners, asking them to submit the final amount that they and the contractor had agreed on for a specific home project.

The email was opened 60% of the time, but no homeowners actually clicked the “Accept This Bid & Hire” call to action button. This directly impacted BuildZoom’s revenue.

Screenshot from data-analytics tool, manifesting number of clicks into the email where users saw the modal.

I watched video recordings of actual homeowners interacting with the modal via FullStory. Most homeowners, would hover around the modal for a bit and then x out of the screen.

From this, I developed a few hypotheses:

  1. Most homeowners needed time between receiving a bid from a contractor and hiring him.

  2. It is unclear to homeowners what clicking “Accept This Bid & Hire” entails.

  3. The modal, from a purely aesthetic perspective, was not engaging.

I then interviewed 3 real homeowners to find out more:

  1. Homeowners wanted more context when deciding whether to hire, like final price, contractor location, BuildZoom score (based on permit data), and previous reviews of the contractor.

  2. Homeowners expected bidding to be a competitive process and not a single decision.

  3. Homeowners needed more time when hiring.

“[I] took a few days to hire, I didn’t hire the first moment I opened the modal.”


I translated my findings into three possible solutions.

I presented all 3 options to the team. Two were simple UI fixes where I changed visual styling. The last solution was more of a UX overhaul that provided more context for the user. I was given the go-ahead to pursue the UX overhaul for both mobile and desktop platforms.


I put together a final flow for a mobile and desktop experience.

Final mobile flow.

Key improvements in the new design were all directly related to user insights:

  • More context regarding contractor i.e. the location, final price, project description, additional documents (i.e. blueprints, hardware/furniture images, etc.).

  • Clear that inputted information is confidential. Addressed need for user trust by explicitly indicating that the information inputted (particularly the hiring price) was confidential.

  • Better navigation would allow homeowners to go back and forth and return to the modal, giving them more time to decide whether to officially hire a contractor.

A quick GIF on the final desktop flow.

View Prototype >


  1. I would have had better documentation of all options I presented to the team to demonstrate various thought processes & iterations.

  2. I would user test my designs with homeowners to validate my design decisions.

  3. I would A/B test certain copy to see what resonated more with users.


Check out a fun video I made when redesigning the BuildZoom’s careers page (another of my projects this summer)!


OTHER WORK: Cisco / StarButter / Dayli