From one sentence to a table you can export.
Four steps, and the fourth is the one most tools skip: keeping the people.
1. Brief
Describe who to study in natural language. Headcount, roles and culture are parsed from it. Name a country and names and languages follow it.
2. Staff
1 to 200 personas are generated with distinct jobs, personalities and attitudes. Inspect any of them on their person card.
3. Study
Survey, Message, Shock, Ask or Rumor. Each run is a persisted event, with answers written to memory.
4. Keep
Export CSV, JSON or PDF. Then continue the audience with memory, clear it, or clone it into Lab.
What a real week looks like.
An illustrative workflow, built from the instruments that ship today.
Staff
Brief: '80 first-job earners skeptical of another fintech app.' Artixol staffs the audience. Open a few person cards.
Baseline survey
Ask the same questions of everyone. Read the split and open two people on each side. Export the table.
Test the lines
Run two taglines in Lab on clones of the same people. Read control and treatment side by side.
Shock it
Inject a pricing change as a Shock. Then re-survey the same people to see who moved.
Keep the asset
Continue the sample for next month's brief. Same people, thicker memory, no re-recruiting.
Learning is memory, not retraining.
When a study runs, each persona writes episodic and semantic memories and updates topic attitudes (a 0 to 10 score with a note). The next study sees them. Personas become more themselves over time, which is the point of continuing an audience.
Why the fixed answer formats matter
Free text is hard to compare across 100 people. Each instrument records a fixed choice plus a reason, so you can tally, filter by role and export. The reasons stay in each persona's own voice.
Lab is the opposite of continuing
Continuing an audience keeps leftover memory on purpose. Lab clones the same people with clean memory so two treatments compare fairly. Choose the mode that matches your question.
Quick answers
What can I study with it?
You can run a Survey, a Message test, a Shock (a world event), an Ask (open question or 1:1 interview), a Rumor (word of mouth along relationships) and a Lab A/B experiment on cloned samples.
How long does a study take?
Most studies on an audience of a few dozen personas finish in minutes. Larger audiences take longer because each persona answers individually.
Bring a question. We will staff the audience.
Tell us who you need to understand. We will build the audience and run a study with you.