PlaysbyMoat8
Back to the library
The workflow guideOperations

Group audience questions for a live Q&A

Collect incoming questions and use an assistant to identify common themes for the moderator.

6 implementation stepsSimple

How this works

Collect incoming questions and use an assistant to identify common themes for the moderator.

Cluster questions by the issue the audience wants answered while preserving distinct viewpoints. A frequent theme deserves visibility, but a rare safety, accessibility or clarification question should not disappear merely because it has few votes.

Before you start

Defined operating scope

Live events receive overlapping audience questions that exceed moderator attention.

Required source material

Obtain Audience questions; Session topics. If connectors are unavailable, use authorized files with the same fields.

Decision and exception rules

Two questions concern the same topic but opposite concerns. Keep separate subgroups or explicitly preserve both perspectives.

Prepare the inputs

question_stream
Required

IDs, text and arrival times

Retain original wording and duplicate relationships.

session_scope
Required

Topics and speaker expertise

Identify off-topic or later-session questions.

grouping_rules
Required

Equivalence and theme criteria

Do not merge opposite premises.

moderator_preferences
Required

Time limit and selection criteria

Keep the human selection decision explicit.

Implementation, step by step

  1. Establish the working record

    • Confirm the selected records, effective versions and responsible owner.
    • Check question_stream, session_scope, grouping_rules, moderator_preferences before beginning. Record the source version and the selected scope; an unavailable value stays unknown rather than becoming an inferred fact.

    Checkpoint: The input record identifies the selected question_stream and any missing required fields.

  2. Collect questions in a shared table

    • Collect questions in one table with stable IDs and avoid counting repeated submissions as independent interest without checking.
    • Check each cluster’s representative question against all member questions.

    Checkpoint: A saved intermediate result for: Collect questions in a shared table.

  3. Group equivalent questions

    • Group genuinely equivalent questions while retaining each original and any important differences.
    • Reconcile counts to distinct input IDs and inspect minority or contradictory questions.

    Checkpoint: A saved intermediate result for: Group equivalent questions.

  4. Draft concise representative questions

    • Draft neutral representative wording that captures the common request without adding a premise the audience did not state.
    • Check each cluster’s representative question against all member questions.

    Checkpoint: A saved intermediate result for: Draft concise representative questions.

  5. Let the moderator choose what to ask

    • Present theme counts, minority questions and a time-aware shortlist for the moderator to select and revise.
    • Reconcile counts to distinct input IDs and inspect minority or contradictory questions.

    Checkpoint: A saved intermediate result for: Let the moderator choose what to ask.

  6. Package the result and unresolved work

    • Preserve the meaning of minority questions and avoid inventing audience concerns.
    • The moderator chooses what is asked; the assistant does not invent audience demand.

    Checkpoint: A prioritized Q&A list with recurring themes. Unresolved items remain visible in the review record.

Decisions along the way

Two questions concern the same topic but opposite concerns.

Keep separate subgroups or explicitly preserve both perspectives.

A minority question raises a material unresolved point.

Surface it independently rather than discarding it for low frequency.

A question contains identifying or abusive content.

Prepare a neutral version or flag it for moderation while retaining restricted original evidence.

What you will produce

cs-video-event-question-clustering-result.json

JSON

A prioritized Q&A list with recurring themes. Label proposed and completed states separately.

Include: cluster_id · theme · question_ids · representative_question · count · minority_view · moderator_selection · status · evidence_reference · review_note

cs-video-event-question-clustering-review.md

Markdown

Keep the evidence and unresolved decisions needed to apply this result, including exceptions involving moderator_preferences.

Include: source_reference · question · owner · next_action

Check the result

  • Check each cluster’s representative question against all member questions.
  • Reconcile counts to distinct input IDs and inspect minority or contradictory questions.
  • Inspect question_stream and session_scope in both the original and output; confirm that record identity and interpretation survived the transformation.
  • Preserve the meaning of minority questions and avoid inventing audience concerns.

When something goes wrong

Two questions concern the same topic but opposite concerns.

Keep separate subgroups or explicitly preserve both perspectives. Retain the affected evidence and recovery status.

A minority question raises a material unresolved point.

Surface it independently rather than discarding it for low frequency. Retain the affected evidence and recovery status.

A question contains identifying or abusive content.

Prepare a neutral version or flag it for moderation while retaining restricted original evidence. Retain the affected evidence and recovery status.

A worked example

SYNTHETIC EXAMPLE · NOT A REPORTED BUSINESS RESULT

Request

Group audience questions for a live Q&A for the following bounded example; prepare the result for review.

Inputs

Synthetic example: Q-1 and Q-2 ask how to start a pilot. Q-3 asks whether pilots should be avoided when data is incomplete.

Expected result

Synthetic expected result: Group Q-1 and Q-2 as starting steps; keep Q-3 as a separate readiness question. Representative wording does not imply everyone wants immediate deployment.

Run a small pilot

Start small

Trial a representative batch containing the ordinary case and the exception demonstrated above.

Record the baseline

Record current preparation time, review effort and correction types for the same records.

Agree success criteria

Agree acceptable review and error limits; measure distinct questions represented; moderator preparation time and missed important topics.

Stop or revise

A question contains identifying or abusive content. Pause affected records and apply this recovery: Prepare a neutral version or flag it for moderation while retaining restricted original evidence.

Keep the scope clear

  • The moderator chooses what is asked; the assistant does not invent audience demand.
  • Limit public display of personal details from submitted questions.

Check before using

Preserve the meaning of minority questions and avoid inventing audience concerns.

Business activities

  • Coordinate meetings and follow-ups

Editorially adapted workflow; not independently tested.

FROM A RECIPE TO YOUR WORKFLOW

Bring the company context together.

This workflow needs reliable inputs before it needs a better prompt. Moat8 Vault brings working information into a readable archive for the AI tools you use.

For this recipe: Audience questions · Session topics

Start with one recurring task, agree which sources to use, and review the first result with your team.

Discuss this workflow

What does your team do?

Choose the recurring work you would like to improve.

A field guide, made for work.

Plays is a curated library of anonymized business workflows. Each recipe explains the inputs, steps, tools, and checks you can adapt to your team. These are editorial guides, not independently tested implementations.

When you share a company website, the guide may read a limited number of public pages on that website. Messages and retrieved page text are sent to our AI provider, NVIDIA, to recommend relevant workflows. Share a general description rather than confidential information.

Saved workflows and selected activities stay in this browser. Your chat history and selected company website are kept in this tab and cleared when you reload or start a new chat. The server does not store conversations. Live source monitoring is not enabled.

If you allow analytics, PostHog receives page views and actions such as workflow saves, skill downloads and link clicks. Random browser and session identifiers connect these actions without identifying you by name. Your choice and browser identifier expire after 90 days; sessions restart after 30 minutes of inactivity or 24 hours. We use local storage, not analytics cookies. Messages, company URLs, search text, full referrers and IP addresses are not sent. There are no person profiles or session recordings. You can decline or withdraw in Analytics settings; Do Not Track and Global Privacy Control keep analytics off.

Read the Plays Privacy Notice ↗