AI probing is a feature in Opini.ai that automatically generates a follow-up open-ended question based on a respondent’s initial input — regardless of the question type.
How It Works:
- The respondent answers any type of question (e.g., selects an option or writes a response).
- If AI probing is enabled, the system evaluates the response in real-time.
- Based on that answer, it generates an open-ended follow-up question to explore the topic further.
- The respondent answers the AI-generated question, giving you deeper, qualitative insight.
How to enable AI probing for your survey #
To enable AI probing on a question:
- Go to the Questionnaire tab of your survey project.
- Add a new question and select “AI Smart Probe” as the question type.
- In the probe setup area, write your instructions to the AI — this is where you tell the system what kind of follow-up you’d like to generate.
- Example: “If the respondent scores low on satisfaction, what would be a good follow-up question?”
- To reference the respondent’s actual answer, insert a value from a previous question:
- Click the “A” icon to open the value insert menu.
- Select the relevant question (e.g., Q3) so the AI can access the respondent’s answer.
- Use the dropdown next to your prompt to confirm which question response (e.g., Q3) the AI should use.

Example use cases for AI probes #
Use Case: Customer Feedback Surveys #
- Main question: Q1: How satisfied are you with our service?
- AI probe instruction: Based on the respondent’s answer in Q1, ask a follow-up question.
- Outcome: The AI generates a follow-up question based on the respondent’s answer:
- Unsatisfied: “What specifically was not satisfactory about your experience?”
- Neutral: “Can you elaborate on what specifically made you feel neutral about the topic?”
- Satisfied: “What specifically made you feel satisfied with your experience?”



Use Case: Employee Engagement #
- Main question: How do you feel about your workload?
- AI probe instruction: Based on the answer in Q3, generate a follow-up to explore what affects their workload perception.
- Outcome: The AI generates a follow-up depending on the response:
- Underloaded: “Which tasks or responsibilities make you feel underloaded, and have any recent changes affected your workload?”


