Lesson 15.4: Topics, Knowledge, and Generative Answers

Module 15: Copilot Studio Fundamentals

Lesson 15.4: Topics, Knowledge, and Generative Answers

Intermediate Last verified: 2026-06-02
Availability and governance note: Copilot Studio features, channels, authentication, generative AI settings, and Power Platform policies vary by tenant, licensing, environments, and admin governance.

Lesson Promise

Use the right mix of authored conversation and source-grounded answers.

Real-World Scenario

A user may ask a predictable question like 'reset my password' or a broad knowledge question like 'what is included in onboarding week one?'

Core Concept

Topics are useful for predictable flows, explicit steps, decision trees, and controlled handoffs.

Knowledge and generative answers help the agent answer questions from configured sources without hand-authoring every possible phrase.

Strong agents combine both: controlled paths where precision matters and grounded answers where source-based flexibility matters.

Step-By-Step Workflow

  1. List predictable intents that need controlled handling.
  2. Create topics for those intents and define trigger phrases.
  3. Map broad FAQ questions to knowledge sources.
  4. Define answer format and citation/source review expectations.
  5. Test the same question in multiple phrasings.
  6. Revise topics or knowledge when the agent chooses the wrong path.

Prompt Lab

Bad Prompt

Use AI for all answers.

Better Prompt

Use topics for password reset and billing handoff; use knowledge for approved onboarding FAQ answers.

Expert Prompt

Create a topic and knowledge strategy for this agent. Classify each user need as controlled topic, knowledge/generative answer, action, escalation, or out-of-scope. Include trigger phrases, source requirements, expected answer style, and test cases.

Hands-On Exercise

Build a topic-vs-knowledge map for an FAQ agent.

Deliverable

A topic and knowledge strategy table.

Copilot Studio Fundamentals Checklist

Common Mistakes

  • Treating Copilot Studio as a chatbot page builder instead of an agent lifecycle platform.
  • Creating too many hand-authored branches before clarifying the agent's purpose.
  • Using knowledge without checking freshness, permissions, and source ownership.
  • Publishing before testing with realistic user language.
  • Ignoring environment, DLP, authentication, and admin policy implications.
Pro tip: Use topics for predictable conversations and controlled paths; use knowledge and generative answers for source-grounded questions that should not require a hand-authored branch for every wording.

Quiz / Checkpoint

Why use topics and knowledge together?

Topics control predictable flows; knowledge provides flexible source-grounded answers for broader questions.

Official Sources To Verify

For AI founders and marketers

Want your AI product explained to a large AI-native audience?

Kingy AI helps AI companies turn complex products into clear, useful YouTube videos that drive awareness, product understanding, demos, clicks, and search visibility.