
Built by researchers. Made for everyone.
ReChat is a patented platform for designing, running, and analyzing structured discussions at any scale.
ReChat was invented at Princeton University.
Surveys capture opinions. Messaging platforms move conversations. Focus group tools gather reactions. ReChat brings all of this together in one place — structured discussions you design, moderate, and analyze in real time, with any mix of human and AI participants. All in your browser.
In 2019, a political science researcher at Princeton was trying to study something deceptively simple: what actually happens when people talk to each other about politics. Not what they say in a survey. Not their individual opinions. The real thing. Back-and-forth. Opinions shifting mid-conversation. Arguments that build, collapse, and rebuild.
The options were frustrating. Building custom software from scratch meant months of work and serious money. Bringing everyone into the same physical room was slow and expensive, and then COVID made it impossible. Existing chat platforms were designed for communication, not research. No structure, no control, no clean data. And online focus-group tools mostly just routed responses to a moderator. Participants never actually talked to each other.
None of it was good enough. There had to be something flexible enough for different research designs, but simple enough that you didn't need a programmer to run a study.
That frustration became ReChat. You describe what you want: how many people per group, what they see, how the discussion should flow. The platform handles the rest.
What started as a workaround for one researcher quickly became something others needed too. Polling firms, market research teams, educators, and organizations all kept running into the same wall: they needed real group discussions, without the infrastructure headache. It turns out a lot of people want to run structured discussions. Very few want to build the tools to do it.
ReChat was introduced in a published study in the Journal of Experimental Political Science.
Read the published study
Six principles shaped how we build ReChat
Every feature exists because researchers asked for it. We don't add complexity for its own sake.
Understanding human behavior is hard enough. Your software shouldn't make it harder. ReChat should feel obvious to use.
We price ReChat so that anyone — individuals, teams, and organizations — can afford to run structured discussions.
We built the platform around structure because we believe the quality of your insights depends on the quality of your conversation design.
The line between human and AI interaction is shifting faster than most platforms can keep up. ReChat was designed from the beginning to support that evolution.
Every transcript, response, and data point captured on ReChat belongs to you. We believe everyone using our platform should have complete ownership of what their conversations reveal.
ReChat draws inspiration from academic research in communication, deliberation, behavioral studies, and digital interaction systems. The platform is designed to support evolving needs across universities, research centers, analytical organizations, and collaborative institutions seeking more advanced approaches to conversation-based insight generation.
To make structured discussion accessible to anyone who needs to understand how people think, decide, and interact.
To make structured human interaction — between people, between people and AI, and between AI agents — something anyone can design, study, and learn from.
We focus on making one thing work really well at one time. Our background is in research, so we understand the problems you're trying to solve.
Best features come from users telling us what's broken or missing.
Clean transcripts with accurate timestamps.
When you email us, you get a person who understands your work.
We ship improvements regularly. If something isn't working, we don't wait to fix it.
Every session, transcript, and participant response is handled with strict data integrity. What happens in your discussion stays yours.
Every decision we make about the platform is tested against real use cases. We don't build on assumption.
Whether you're running one session or hundreds, ReChat performs consistently. We build for the researcher who needs reliability as much as features.
We don't add AI because it's trending. Every AI capability on ReChat is designed with a clear research purpose, to extend what's possible, not replace human judgment.