Caylee Betts, Product Designer


When I joined Porch, the company had quickly raised $100M and grown to nearly 500 employees in two years. On the product design team, I helped develop the logged-in experience, where homeowners can start, manage and share home projects; hire professionals; and get inspiration. I also lead the design for Porch's installation services program, working with retail partners like ATG and Wayfair.

Role: Senior Product Designer

Retail Installation Services

A new line of business Porch offered was installation services to customers of retailers like Wayfair and ATG. When shopping for something like a ceiling fan on a partner site, the user could opt-in to add Porch installation. I lead the design of this, which was an interesting pursuit as we learned that each partner cared about different components of the flow, and each partner had different restrictions that affected the flow. After several screenshot/Photoshop-hacked renderings, I was given the go-ahead to create generic (non-branded) flows that I believed solved the problems that our experience with the first few retailers had surfaced. These flows were used for future pitches, reducing the prep time for those meetings. Our team would customize the flow for the partner after definite interest was there, rather than before.

The Flow
The user arrives on product detail page on a site like Wayfair. The user can opt-in to installation services before clicking “Add to Cart.” A Porch modal is fired, giving an overview of how the service works, asking the user to confirm their ZIP code, and to select service type (for a ceiling fan, there would be “first-time installation” and “replacement,” for instance). Once the user confirms/adds installation, they are taken to the cart. The item is there with a Porch Installation sub-item. The user checks out for the item and the retailer sends Porch the user’s information. Porch emails the user to finalize installation details and price.

Include contextual details and micro-interactions in the user's experience helps to direct the focus, the movement, and to educate the user. In this case, we had a lot of information to convey, but not all of it was relevant to each installation type we were offering. So, only after the installation type is selected does the user see this information. This also calls more attention to information as it appears once the user makes a selection. I also dimmed the primary button while inactive. Once it becomes active, it alerts the user they've completed all of the steps by illuminating.

An example of an error state.

UX Problems and Solutions
Porch only services certain areas, and doesn’t have the customer’s definitive location. We chose to collect location data and only display the installation option if within service zone for the pilot/v1.

Porch is able to gather certain information from the retailer after checkout - name, contact info, shipment date - but no payment information. To prevent checkout abandonment, we couldn’t easily ask the user in the middle of the flow to enter their payment info and then have them enter it again at checkout. Our work around was to finalize installation details, including payment, after the retailer’s flow.

Porch couldn’t calculate the exact price of the installation service until gathering more information from the user and matching them with a pro. This required us to be really vague during the retail flow, and overall was a really tough problem for us to solve. We felt that the best choice was to flat-rate these services to reduce confusion, and collect payment through the retailer. The retailers did not allow us to collect payment on their sites, and Porch’s process with pros wasn’t ready to streamline this.

Porch couldn’t schedule the exact time and date for the service because the user’s shipping option was selected after the Porch modal. If the ceiling fan isn’t there, Porch can’t send someone to install it. I really pushed (successfully) to remove the scheduling feature because of the repetition it would cause. In the end, we opted to handle all particulars after the retail flow. We considered the first iteration of this program to come to Porch as a “hot lead” rather than a definite sale.

Best Places to Own a Home Survey Site

The intention of this project was to improve Porch’s SEO by creating a micro-site worthy of sharing and then linking to all the service pages in the area. We collected data from 10,000 US homeowners. The results were super interesting but from a small sampling, so it was tough to design for.

I lead the design on this project. The first step was to digest the data. Designs were requested before the data was collected, but I knew it was important to base the designs on the data that was found, so I worked closely with the project’s data scientist to massage interesting stories from the raw data. What we came up with was a template for each city that highlighted strengths and weaknesses and allowed users to compare with other cities, as well as a national level view that displayed the most interesting nation-wide facts and featured a comparative map.

Homeowner Experience

My first project at Porch was on the team imagining the next phase of the homeowner experience. Prior to our work, homeowners could search for professionals, start projects, and claim their home on the site. The forward vision for this product was more of a homeowner's social network, where you could also browse content (sales, projects, recommendations within your neighborhood), get inspiration and advice on improving your home, comprehensively manage your home projects, and find and coordinate pros for your jobs. Because "premium" professionals create revenue for Porch, we wanted to create social trust amongst neighbors, highlighting pros doing great work in the homeowner's neighborhood, and make finding and booking a pro as seamless as possible.

We worked on two interesting initiatives within the Homeowner Experience: Home Assistant and Home Health Score. The Home Assistant was a subscription-based service (piloted in Seattle and Chicago) that aimed to generate revenue on the Homeowner side of the business by allowing users to turn over home management or a project like a remodel to their Porch Home Assistant. It was our job to educate unpaid users on this premium option. The Home Health score was partially created to draw attention to the Home Assistant while also being a play for more partnerships. Here’s how it worked: all homes had a rating before being claimed on Porch based on things like location, age, and any public information about upkeep or improvement. Once claimed, the onboarding flow helps to improve your score, and eventually the site recommends a variety of improvements to raise your score even more. One of those being the Home Assistant program. The Home Health Score and Home Assistant both garnered interest from many big name insurance companies, but in the end, neither survived the major pivot of Porch discontinuing the entire Homeowner program.

Booking App

As user data started to come in for Porch’s relatively new booking app, I was brought in to roll out small updates and test new features. I love mobile work and would have enjoyed spending more time on this.


Porch was the top choice in my job search. I was looking for a company that was fast-growing and creating a consumer product (not super common in Seattle). As an avid startup industry follower, I was really excited to look under the hood of one of Seattle’s most promising startups. My time at Porch has definitely prepared me for working quickly, staying on my toes, and executing all sorts of requirements. I’ve taken many learnings away to be applied to future roles!