From recording to a ready-made recommendation — without manual interview analysis
Respondents provided spoken feedback, but the analysis was still prepared manually by listening to recordings, extracting quotations and compiling results. The decision-maker received material to interpret independently rather than a ready-made recommendation.
Before
The agency combined quantitative and qualitative research. Surveys were fast, with figures available immediately. Voice interviews provided richer material, but the recording was followed by laborious work: listening, extracting answers and combining statements from more than a dozen people into one coherent picture. Several days passed before the analysis reached the decision-maker.
Discovery
The key question was not about recording, but about what happened afterwards. We tested whether the system could aggregate statements from multiple respondents into a single coherent synthesis and how to build an interface that would not discourage respondents by requiring them to enable their microphone.
What we tested
- Transcription and group synthesis — can the system process recordings from more than a dozen interviews and aggregate the responses into one coherent synthesis instead of producing separate summaries for each respondent?
- Simplicity for respondents — can a mobile voice survey be as intuitive as a standard form, allow users to return to an interrupted session and accommodate those who do not wish to record their answers?
What changed
Respondents recorded their answers using a simple mobile application. Once collected, the system processed the recordings automatically, with no manual work by an analyst. Audio files were deleted after processing. The researcher could view aggregated results, charts and a ready-made synthesis of responses instead of spending hours listening to recordings.
Client decision
The application is used to conduct research.
Would you like to test a similar area in your company?
Book a meetingFree 30-minute online consultation