Our research staff continues to collect, interpret, and classify sentiment-based data, leveraging their unique relationships with respondents. Each interview undergoes a quality control process, which also serves as the preprocessing step for our GenAI analysis. This ensures that the data is coherent and clear for our AI model to process. By placing our team in the centre of the analysis process, we ensure that AI outputs add value and align with our goals.

We use GenAI to probe our data per the classifications we define, rather than relying on the tool to handle the definition of our data.

We use our large language model (LLM) to produce outputs across the various subsets of our data. We can look at, for example, results per different segments of respondents, respondents in different regions, or the sentiments held by respondents at different times.  

We understand that the qualitative data we collect during our interviews is essential to your strategic thinking and problem solving. Therefore we include our outputs in your daily prelim reports.