Built With MongoDB: Queenly
April 20, 2021 | Updated: October 17, 2022
Queenly founders Trisha Bantigue and Kathy Zhou grew up in low-income immigrant families, trying to balance their cultural upbringing with their desire to fit into their American lifestyle. To earn scholarships to pay for college, they both started participating in beauty pageants.
“Beauty pageants provide young women with the opportunity to kickstart their careers,” says Queenly Co-founder & CTO Kathy. “And one of the core parts of the pageant system is the evening gown — that was the spark of inspiration for us wanting to tackle the whole fashion industry.”
According to Kathy, women in America — especially outside the coastal cities — end up attending around 15 special occasions a year. “Whether it’s prom, beauty pageants, or other formal occasions, women need cost-effective formal outfits,” she says.
After working across leading Silicon Valley companies, Trisha and Kathy teamed up to build Queenly, a marketplace and search engine for the formalwear industry.
“No one has created a robust search engine for formal dresses,” says Kathy. “People are picky about formal attire — there’s so much consideration that goes into it, from neckline to hemline, silhouettes, colors, and fabrics. We’re trying to build a marketplace, do complex queries, and provide personalized recommendations.”
Queenly has 80,000 registered users and 50,000 dresses listed. The team of five (which is hiring!) is backed by Y-Combinator.
We recently sat down with Kathy to learn more about how her pageant experience has informed her career, her experience with MongoDB, and the challenges in building a formalwear business.
MongoDB: What was your first pageant experience like?
Kathy: It was really eye-opening: I have always been a shy person, and my number one fear is public speaking. What people don’t realize about pageants is that along with having to learn how to dress well, you also have to be able to speak well. You have to learn to speak from your heart and to communicate well. Gaining the confidence and soft skills to answer those pageant questions has also helped me in my career, helping me grow from an engineer to engineering leadership.
One of the most memorable questions from an early pageant was about what’s the most important thing you want to do in your pageant regime. I talked about how it’s okay for young women to both be nerdy and girly — you should be able to embrace all these different sides of yourself, and not fear falling into one box of being. I wish someone had told me that when I was younger. Now, I’m honored to be able to embrace both sides as a CTO, a Y-Combinator female founder, and a beauty pageant contestant.
MongoDB: Building a two-sided marketplace is a challenge. What did the minimum viable product look like?
Kathy: The MVP was very rough — I started by coding an iOS app part-time and during the weekends while I was still employed at Pinterest. The goal was to tackle the supply-side of the marketplace first to get people to upload dresses, so I optimized for creating a really easy dress-upload experience. You could only search for one size and one color at a time. Now, we’re using natural language processing query for search, and also a larger combination of different dress-type attributes. We’re also including reverse image search, and I’ve been working on tailored user recommendations.
MongoDB: How did you make decisions for your technical back end?
Kathy: Initially, we had very basic search and exploration using Google’s Firebase. It was very easy to set up and has a fairly good UI tooling, but its query capacity was something we were quickly outgrowing. At our stage of company, non-relational storages are a really great decision for the sake of speed and adaptability. As we’re working towards product-market fit, we need to move quickly in launching new user experiences and reworking old ones, so it’s important to have that flexibility in restructuring and reshaping our data.
As people say, we were building the airplane as we were flying. We needed to move fast so people could access and search for dresses quickly. Many of our users are women who live in the Midwest and the South where they may not have amazing internet access, so speed and performance are pretty important.
MongoDB: Are there specific features of MongoDB you're using, aside from Atlas?
Kathy: The most important aspects are the core functionality and the monitoring toolings and dashboards. Those are useful and come right out of the box. I’ve been meaning to take a look at search capabilities — I think it’s cool that there are indexes right out of the box. We’re trying to adapt our product as it goes, and figure out how to tag and enable different attributes on a dress.
MongoDB: What was the last good technical book or article you read?
Kathy: I really enjoy reading the Towards Data Science publication on Medium. They do a good job of covering different use cases as well as making different fields algorithms and data science/machine learning concepts more approachable.
Beyond that, I read several fashion magazines and pageant blogs because I think CTOs — and the technical side of the business — should really understand the users. I try to keep up with trends in fashion and retail to better understand the opportunity, and use that to influence how our product functions.
Looking to build something cool? Get started with the MongoDB for Startups program.