New: Customise your AI agent's behaviour
Great research isn't about one AI doing one thing well. It's about the AI doing your thing well — following your method, your tone, your standard of evidence.
Today we're shipping behavioural controls that let every team on Opini.ai shape how their AI agent conducts interviews. You choose how strictly the agent sticks to your guide, how deeply it probes and how it phrases its follow-ups. Every study gets its own configuration, saved as a profile you can reuse.
What you can control
- Script adherence: from strict — never deviate — to exploratory, where the agent follows any interesting thread.
- Probing depth: a slider that controls how many follow-ups the agent will chain per question.
- Tone: formal, conversational, playful — matched to your audience and brand voice.
- Language and locale: set primary and fallback languages so the agent respects the respondent's preference.
Why it matters
A B2B customer interview and a Gen-Z consumer study need very different interviewers. Configurable behaviour lets you keep one platform, one workflow, and still get an agent that behaves like it was trained for your specific study.
Available now on all plans. Head to your study settings to try it, or reach out and we'll walk you through a profile that fits your next launch.