Personas.
Simulated message testing
Improve the message before a customer sees it.
One of fifteen tools the studio built for its own work. We build the same kind of app for clients, shaped around the job they need doing.
- Made for
- Testing messages before real customers see them
- Shape
- Native Mac app
- Data
- Stored locally; AI requests send relevant content to the configured provider
What Personas does.
Personas turns interviews, surveys and support conversations into customer profiles. Every trait links back to the evidence you supplied, so you can see what each profile is based on.
Test several versions of a message across the panel at once. Each version comes back scored, with likely objections and a suggested rewrite. Research is stored locally; AI requests send relevant research, profiles and messages to the configured provider. Simulated reactions are prompts for further research, not customer validation.
Explore possible reactions to a message, then check important assumptions with real customers.
Why we built Personas.
Every business writes messages it can't easily test: the price rise email, the new campaign line, the delicate note to a best client. A real focus group costs thousands of pounds and takes weeks to organise, so in practice, most messages go out untested.
We had the raw material all along: real interviews, survey responses and support conversations sitting in project folders. Personas was built to put that research to work, turning it into customer personas we could actually ask.
How Personas earns its place in a working day.
Add your research
Paste interview notes or survey responses. AI drafts the profiles, with each trait tied to its supporting evidence.
Check the evidence
Every trait shows its source and a confidence score. Anything that doesn't ring true, you edit; the AI never blocks you.
Test the message
Put two to five versions of a message in front of the panel. Each receives simulated scores, possible objections and a suggested rewrite.
Compare simulated reactions
For bigger decisions, the profiles discuss the idea together. You receive a structured report showing agreements, objections and differences.
Message choices get rehearsed
Simulated scores help us spot questions to investigate with real customers.
Old research earns its keep
Interviews and surveys that used to be filed and forgotten now sit behind every test we run.
The awkward messages get rehearsed
Price changes and difficult announcements go through the panel first. Possible objections surface for us to investigate before sending.
Made to fit the work it needs to do.
Personas stores profiles and assets locally. Optional AI features use your API key and send the request content to the configured provider, so local storage does not mean offline AI processing.
One rule shaped the whole build: the AI drafts, a person decides. Nothing is ever locked behind an AI result; every field can be edited by hand, and every claim shows the evidence it rests on.
AI drafts each customer profile from your research, answers questions in character and scores message options. Every profile remains editable, and the supporting evidence stays visible.
Could your business use something like this?
Tell us the job in plain English. We'll say what would solve it, what it would cost, and whether a desktop app is the right shape for it.