Synthetic user feedback
Active Personas turn a static persona document into an LLM-powered stand-in for a real user. Ask it what it thinks of a screen, run a heuristic evaluation, or send it to walk through your live website. Get feedback in the persona's own voice, on demand, to complement what you learn from real users.
A prototype built on the research of Simaremare & Edison (ICSOB 2024 and 2025). Read the papers.
A product development team can ask an Active Persona to explore a web app and say how it feels to use, to gain early insight into its usability.
Describe a target user: goals, tech proficiency, accessibility needs, tone, background. Plain text, no model involved.
Bind the Persona to an LLM configuration and an impersonation approach. The same Persona can run against several models to compare tone and criticality.
A Panel is a shared chat space for your team and one or more Active Personas. Attach screenshots, @mention an AP, ask what it thinks.
Run the built-in 20-item Nielsen heuristic questionnaire, or send an AP to drive your live site step by step and report what it did, saw, and felt.
One primitive for every conversation: one user and one AP, a whole team with several APs, or a broadcast to all of them at once.
Instantiate one Persona against Gemini, Llama, GPT, and Claude side by side. In the study, the choice of model shifted how critical the feedback was.
The study's 20-statement questionnaire on Nielsen's 10 heuristics, answered with a Likert rating and a rationale per statement, so you can compare personas and models.
An AP drives a real browser through your site, screenshot by screenshot, using a registered credential bound to its allowed origins, then reports back in persona.
Download a Panel as a PDF, or fetch it as structured JSON through the API: the screenshots shown, the questions asked, every reply, and the Likert ratings.
Actions are logged, each agent turn keeps a process trace, and credentials are redacted before anything is written down.
Hear your users' perspective while there is still time to act on it. Ask an Active Persona what confuses it on a screen, send it through your web app, or run a structured usability review, as often as you like, at any stage of development.
The approach comes from peer-reviewed research. When we compared Active Persona feedback on a transit app with what real users said in interviews and app-store reviews, the two lined up well. The study also showed that the persona you write and the model behind it shape how critical the feedback is, which is why you can compare several side by side. Use it next to your user research: it helps you find the questions worth asking sooner and walk into interviews better prepared.
The research behind it
Questions, feedback, or interested in trying Active Personas with your team? We would like to hear from you.
mario.simaremare@bth.seThis application is part of a research project supported by
ActivePersonas