UserTold.ai
Open appGet Started

Study Runtime

UserTold interviews capture real product usage, preserve what participants say and do, ask planned follow-up questions, extract reviewable evidence, route work, and watch future interviews for recurrence after Linear completion.

During observation, the runtime stays silent, then hands that context to a talk debrief.

Three Segment Modes

Every segment uses one participant-facing mode.

Observe

The participant uses your product while UserTold captures product usage, speech, clicks, navigation, and available snapshots. The assistant stays silent. Pauses, loops, confusion phrases, and help requests are preserved as evidence for interpretation and debrief.

Best for: usability testing, task completion studies, real workflow observation.

Speak

The assistant delivers a scripted one-way message. Use it for intros, task instructions, transitions, and thanks. It is not a live help mode.

Best for: task handoffs, consent reminders, short setup or wrap-up messages.

Talk

GPT-Live-1 handles the full-duplex voice conversation and interruptions. UserTold's delegated backend agent reasons over the Study and grounded context, chooses approved tools, and calls complete_segment when the segment is done.

Best for: discovery interviews, context conversations, and debriefing after an observation segment.

Observe To Talk Handoff

  1. speak gives the task instructions.
  2. observe captures behavior without interruption.
  3. When realtime_analysis is enabled, the runtime periodically turns recent speech, site behavior, and page changes into a running private evidence note in Markdown.
  4. talk starts immediately from the latest valid evidence checkpoint, or from a short fallback when there is no checkpoint. It does not wait for a model call or receive a raw transcript or action log.
  5. Late analysis updates reach the debrief only if they change the picture, so the interviewer does not restart mid-question.
  6. The delegated backend agent combines evidence with the goals and calls complete_segment when more questions would not improve the record; GPT-Live-1 remains the spoken interface.
  7. Final speak thanks the participant or closes the interview when the script calls for it.

The participant's struggle stays visible for debrief and extraction.

Deterministic Advancement

Production scripts advance only by deterministic criteria:

  • max_duration_s
  • user Done / step_done
  • url:<substring>
  • action:<selector-or-pattern>
  • complete_segment for talk segments
  • scripted speak completion

Use URL and action rules when the product exposes a clear completion signal. Use max_duration_s as a safety valve for observation. In talk, the runtime sends a private wrap-up instruction one minute before max_duration_s; the assistant should finish naturally and call complete_segment before the hard limit force-advances. Goals guide analysis and planned debriefs, not observe-mode advancement.

Designing Studies

Usability

Use Speak → Observe → Talk → Speak. Enable Pre-debrief analysis on the planned Talk step when its questions should draw on the observed task. The complete onboarding usability recipe includes the goal, task, time limit, debrief instructions, and realtime_analysis flag. See follow-up questions for the dashboard steps.

Discovery

Use planned talk segments when the research question is conversational. Split long interviews into focused segments with clear goals so complete_segment has a natural boundary.

How It Connects To Evidence Extraction

After the interview, the captured record feeds evidence extraction:

  • transcript and participant quotes
  • navigation and interaction history
  • observation context and page snapshots
  • study goals and segment timing
  • planned debrief answers grounded in observed behavior

This separation keeps the live runtime predictable while still giving extraction enough evidence to explain friction, desired outcomes, workarounds, and recovery.

See also

Try this with your own users

Copy the prompt and paste it into your AI assistant.

View prompt
Help me apply this research approach to my product and set up a suitable interview in UserTold.

Read this guide first: https://usertold.ai/docs/study-runtime

Use my existing UserTold connection. If it is not connected, help me connect through https://mcp.usertold.ai/mcp and complete browser authorization. Ask for missing project details, use the available UserTold tools, and walk me through any steps that need the dashboard.