PLATFORMS
Augmented opinion research
AI support across each step of a study, from drafting the questionnaire to categorising the data and interpreting what it says.
The problem
Challenge
The value of a research study lies in the thorough analysis its researchers conduct and the concrete paths of action they suggest. Yet several parts of the cycle absorb the time that analysis would need the most: questionnaire wording, thousands of open-ended answers sorted largely by hand, the first pass at spotting patterns. And when the data and findings arrive, decision makers are handed conclusions without the means to interpret them. Research is commissioned to be used, and almost all of the effort goes into producing it.
What it changes
Impact
We put technology at the service of researchers: the platform supports the entire research process, from questionnaire development to data management and categorisation, and works with the tools you already use: the questionnaire is deployed to your own research platform, and the data comes back by upload or API connection.
Our teams of practitioners, analysts and developers optimised the automation before bringing it to scale for our partners. It follows the same methodology and rigor as any research study, relieving analysts and researchers of long, repetitive tasks, to allow more time for analysis.
Under the hood
How it works
From a question about what people believe to findings your team can act on.
01
Select the research steps
As a researcher or practicioner, you know which research step(s) is worth optimizing. Select it and the tool will optimize it.
02
Questionnaire and convertion
Draft and refine the instrument with AI support, testing the wording, the logic and the coverage against what the study needs to answer. Export the questionnaire to your research platform.
03
Bring the data back and categorise it
Upload the results or connect your research tool, combine them, then categorise the responses and any text data at scale against the frame you define, with analysts reviewing the output.
04
Interpret your data
An assistant configured for research data answers questions about the findings in plain language, so analysts can go further faster and colleagues can read the data without one.