Sequen gets $16 million to bring TikTok-style personalization technology to a consumer company

At Etsy, Zoë Weil helped achieve a $1 billion increase in gross merchandise volume in one year by improving the online marketplace’s AI classification systems. With her new startup Sequen, she aims to transfer her years of work and those of her founders involved in AI research and product development to other companies in the consumer space.

The company, which just closed $16 million in Series A funding, offers real-time personalization and segmentation infrastructure — technology used by the world’s largest tech companies but inaccessible to other large consumer companies due to the massive data sets required.

While people outside the tech industry may not understand what this technology entails, anyone who uses consumer apps like TikTok, Instagram, or YouTube has been a target of these systems.

“Modern technology doesn’t just recommend content anymore,” explains Weil, CEO of Sequen. “It bends your will in subtle ways over time to make you actually want things. In fact, the technology has gotten so good that many people suspect platforms are eavesdropping on their conversations.”

Weil attributes this phenomenon to the so-called grand event model. While large language models (LLMs) used by chatbots like ChatGPT generalize text, large event models generalize event flows and human behavior in particular. This technology has use cases beyond simply building a better algorithm.

Weil believes Sequen could eventually replace the cookie — a tracking technology that personalizes end-users’ web experiences, but in a way that has raised privacy concerns and spurred regulation.

“Our big event models learn from direct user actions, not just clicks and swipes, but also swipes, conversations, and things within a given session — not static profiles or third-party cookies,” Weil says. “This is how you do real-time personalization, even with sparse data,” she adds. “So, yes, we’re opening up TikTok’s algorithms to Fortune 500 companies that don’t have the infrastructure to do that… but I would say we’re taking it a step further.”

Companies that work with Sequen integrate with the startup’s RankTune platform, allowing them to access Sequen’s parametric and real-time ranking models through APIs. (Sequen customers already use some sort of internal API to run their trapping suite, so they just replace their own API with Sequen’s.)

Furthermore, Sequen’s technology does not interfere with privacy like cookies because it is based on real-time data – no user identity is needed to personalize results. It is fast, with the decision-making process taking less than 20 milliseconds.

“Our large event models are able to generalize to the event streams they get in real time,” Weil says. “It doesn’t matter who performs these events – they are able to perceive and understand the events without relying on the user’s identity. So, in fact, the user’s identity is not relevant at all.”

Despite this more privacy-conscious aspect, Sequen says its technology can still show “an insane increase in revenue,” Weil claims.

In one example, a large furniture company saw revenue increase by 7% after switching to Sequen, whereas a 0.4% rise was previously considered a win. Another client, Fetch Rewards, saw a 20% increase in net revenue in less than 11 days. She also works with a streaming media company and an online travel agency.

The system is priced based on requests per second (RPS), with tiers offering up to 500 RPS, 1000 RPS, etc., with price reductions as levels increase. Among its first five clients, contracts were in the seven figures.

“What we’ve seen consistently across the board is people choosing the top tier, because once they see us in one use case, they want to adopt us for their entire platform,” Weil points out.

Will started her career in the industry through the research side of things, but quickly realized she preferred building products. She has spent most of her time so far helping companies develop these types of classification products to generate business value from them, which is what led her to create Sequen.

Now, in less than 18 months, the company has processed about 10 billion orders per month and won business at a handful of Fortune 500 companies. Its offering includes proprietary technology, including large event models, classification models, algorithms, and more.

At the startup, Will is joined by Ethan Benjamin, who worked with her at Etsy, and co-founders Mo Afshar and Alexander Tom. Rafael Luca recently joined from Meta to become Sequen’s Chief Production Officer. Headquartered in New York, the company’s 14-person team includes employees from DeepMind, Meta, Anthropic, and elsewhere.

Sequen’s Series A was led by White Star Capital and Threshold Ventures, with participation from its previous investors, including Greycroft, which led its seed round. To date, Sequen has raised $22 million.