Dan DeMeyere, ThredUp | Modern Retail Vanguard 2024
This is part of the Modern Retail Vanguard, a series highlighting the behind-the-scenes talent powering the world’s top retail brands. More from the series →
Dan DeMeyere, chief product and technology officer at ThredUp, was one of its first employees when he joined the company in 2010 as a junior engineer. He remembers the scrappy startup life of watching episodes of “Arrested Development” while working late nights – even spending the night at the office.
“Back then, it was like we were having fun,” DeMeyere says. “But it’s not something I would do again, not unless I really needed to.”
Flash forward 14 years later, and DeMeyere has risen through the ranks at ThredUp. He now oversees the technology that powers ThredUp’s resale platform, from the pricing algorithms to the user experience. Over the years, the company’s fortunes have changed dramatically, too. ThredUp went public in 2021, when it was valued at $3 billion. Since then, ThredUp’s stock has lost about 90% of its value. DeMeyere’s objective is to help course correct by scaling ThredUp’s artificial intelligence.
To do that, DeMeyere has gone back to his startup roots. A year ago, DeMeyere set up a dedicated team made up of engineering, product design and data science experts whose primary focus was to figure out how generative AI could be integrated at ThredUp. As DeMeyere puts it, “It was like a startup within the company.”
It became clear early on that AI-powered search was the application that made the most sense for ThredUp, which has more than 4 million SKUs on its platform.
ThredUp got to work building prototypes via hackathons. The company would pick a broad theme – say, chatbots – and then engineers would take a week to work on nothing else except developing a prodcut related to that theme.
One idea that resulted from the hackathons was ThredUp’s AI chatbot dubbed Style Chat, which the company released in August, along with a couple of other AI-focused search tools. Essentially, users can ask for an outfit idea using conversational language, and ThredUp’s bot will spit out a head-to-toe ensemble in a few seconds.
Hackathons, which once took place once or twice a year, now happen on a regular basis, as often as once a month.
It’s still early days, but ThredUp’s AI bet is starting to pay off. ThredUp’s customers are four times more likely to make a purchase if they have engaged with an AI feature on the website or app. New customers are also 55% more likely to revisit in the next 7 days if they use one of ThredUp’s AI tools.
“The adoption was a bit slow at the beginning, but then once customers use it a few times, it becomes their new norm,” DeMeyere says.