Vectorial raises $3 million to simulate consumers for research

Vectorial's pitch rests on a gap its CEO describes: companies now generate code with AI faster than ever, but consumer research still runs on surveys that take weeks and reach only 50 to 100 participants. Taranveer Singh says many smaller startups skip research altogether and simply assume they know what consumers want.

The company argues prior attempts fall short for different reasons. AI-moderated interviews still depend on real humans and small samples, while synthetic agents are not grounded in real consumer behavior. Vectorial instead embeds simulated audiences in the development workflow, drawing on SAPIENS, a behavior model trained on latent socio-demographic, psychographic and behavioral traits across more than 2.5 million real consumers.

SenseAI Ventures' Rahul Agarwalla frames the bet as a new class of AI aimed at predicting human behavior rather than generating the next word. Singh expects the method to become the standard for new consumer products within roughly four years, a claim the startup will have to prove as it expands its dataset and builds SAPIENS 2.0.