Extract specific insights from conversational surveys

Conversational surveys generate rich, natural dialogue but analysing hundreds of transcripts manually takes days of coding work. Deep probe automates this process, letting you simply describe what insights you want to extract and get structured, quantifiable results directly in your report.

The example below shows a practical application of how to extract insights from your conversational survey responses using Deep probe.

Working example: Analysing Australians’ future sentiment and personal aspirations

In a conversational survey exploring 100 Australians’ dreams for the future, the research team wanted to understand individual respondents’ future sentiment and main aspirations in minutes.

Instead of spending hours manually reviewing transcripts, they used Deep probe to automatically transforms conversational data into structured insights in minutes.

Categorising respondents’ future sentiment

Deep probe setup

1. Navigate to the Deep probe tab in the experiment report and click Add a deep probe .
2. Select Add LLM analysis .
3. Describe what you want to extract from your conversational responses. In this example, the request was What is this person's sentiment about their future?.
4. The system automatically suggests categories as the output type with options including “Positive,” “Negative,” “Mixed,” and “Uncertain/unclear.”
5. After reviewing the suggested categories, researchers can accept the recommendations or modify them based on their analysis objectives. In this example, the researcher accepted all suggested categories as they aligned with the analysis goals.

Generated prompt for analysis

The system generates an analysis prompt based on the settings. Final output quality depends on this prompt and LLM processing. Refine the prompt and preview results to ensure it captures what you want to discover before processing begins.

The following shows an example prompt generated by the LLM based on the request in this example.

Request: What is this person's sentiment about their future?

Response:

Outputs overview

In this case, Deep probe assigns a sentiment category to each respondent. Please note that classification accuracy varies depending on prompt and LLM processing and cannot guarantee 100% accuracy. Conjointly recommends reviewing outputs and editing individual results where needed.

Based on these individual results, the system summarises the category outputs under the Summary for category outputs tab. The distribution chart shows percentage breakdowns across the categories. In this case, 43% of respondents expressed positive sentiment, 21% expressed negative sentiment, 31% had mixed sentiment, and 5% were uncertain.

Summarising individual aspirations

Deep probe setup

1. Navigate to the Deep probe tab in the experiment report and click Add a deep probe .
2. Select Add LLM analysis .
3. Describe what you want to extract from your conversational responses. In this example, the request was Summarise this person's key aspirations and priorities.
4. The research team chose string as an output type to generate text summaries.

Generated prompt for analysis

The system generates an analysis prompt based on the settings. Final output quality depends on this prompt and LLM processing. Refine the prompt and preview results to ensure it captures what you want to discover before processing begins.

The following shows an example prompt generated by the LLM based on the request in this example.

Request: Summarise this person's key aspirations and priorities

Response:

Outputs overview

In this case, the Deep probe generates individual summaries that captured the respondent’s core priorities in a concise, scannable format. Please note that summarisation accuracy varies depending on prompt and LLM processing and cannot guarantee 100% accuracy. Conjointly recommends reviewing outputs and editing individual results where needed.

Based on these individual results, the Summary for string outputs tab presents the overall AI-generated summary identifying overall key themes and patterns across all responses.

Best practices for deep probing in conversational survey analysis

• Write clear, specific prompts that focus on extracting one type of information per analysis. Broad prompts produce inconsistent results.
• Define mutually exclusive categories to ensure clean classification. Overlapping categories reduce the clarity of your percentage summaries.
• Consider context when setting up categories, accounting for how respondents might express concepts differently in natural dialogue.
• Test with preview samples before processing all data to save time and processing costs.
• Consider running multiple deep probe analyses on the same conversational data to extract different insights. Each analysis can focus on a specific aspect of the conversation.
• Combine with existing insights from your conversational survey’s AI summaries and sentiment analysis for comprehensive understanding.

Other example analysis requests for conversational surveys

Beyond extracting particular mentions of factors, you can identify broader patterns, sentiments, and themes across your conversational data. These requests demonstrate versatile applications for different research objectives.

Other Deep probe use cases

You can also read about:

• Coding open-ended responses.
• Synthesising individual respondent data.
• Creating calculated values from existing data.