Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.focaldata.com/llms.txt

Use this file to discover all available pages before exploring further.

Once a study is launched, data collection begins. Interviews are conducted through AI-moderated conversations, using the approved discussion guide and stimulus as the source of truth. The aim is consistency without rigidity: every interview covers the required research questions, while participants can still express themselves naturally in their own words. Fieldwork runs continuously until the target number of completed interviews is reached.

How AI moderation works

Interviews are conducted by an AI moderator that follows the approved guide closely and manages the live interaction. In practice, the moderator:
  • Conducts the conversation turn by turn, asking one question at a time.
  • Follows the structure, sequencing, and intent of the approved guide.
  • Uses probes and follow-ups to encourage depth and explanation.
  • Introduces stimulus at the right moment, without altering meaning or adding context.
The moderator is not a free-form chatbot. It runs a method-led interview where coverage of objectives and consistency across participants matter as much as conversational flow. It does have licence to phrase questions conversationally, deviating from the exact wording in the guide where research best practice calls for it.
Moderation guardrails: one question per message; avoid over-relying on “why…?” as the opening word; ensure each research question has sufficient follow-ups.

Fieldwork duration

Fieldwork continues until the target number of completed interviews is reached.
AudienceTypical timeline
Broad (high incidence)24–48 hours
Moderate48–72 hours
Narrow (low incidence)72+ hours, potentially longer

How data quality is enforced

Quality is enforced at multiple points, not just a single check at the end.