Category: Generative AI

AB-620, Agentic AI, AI, Azure AI, Generative AI, Microsoft Certification

Monitor agents by using Application Insights (AB-620 Exam Prep)

 
This post is a part of the AB-620: Designing and Building Integrated AI Agent Solutions in Copilot Studio Exam Prep Hub.
This topic falls under these sections:
Integrate and extend agents in Copilot Studio (40–45%)
   --> Integrate agents with Azure
      --> Monitor agents by using Application Insights

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

As AI agents become more sophisticated and business-critical, monitoring their health, performance, reliability, and user interactions becomes essential. An AI agent that responds slowly, generates errors, experiences high failure rates, or consumes excessive resources can negatively impact business operations and user satisfaction.

Microsoft Copilot Studio integrates with Azure Application Insights, a feature of Azure Monitor, to provide comprehensive telemetry, diagnostics, and performance monitoring. Application Insights collects operational data from agents, allowing administrators and developers to observe agent behavior, troubleshoot issues, measure usage, and optimize performance over time.

For the AB-620 exam, you should understand how Application Insights integrates with Copilot Studio, what telemetry it collects, how to analyze monitoring data, and how monitoring supports production AI solutions.

What is Azure Application Insights?

Azure Application Insights is an application performance monitoring (APM) service within Azure Monitor.

It helps organizations:

  • Monitor application availability
  • Track performance
  • Diagnose failures
  • Analyze user behavior
  • Detect anomalies
  • Monitor dependencies
  • Measure response times
  • Identify bottlenecks
  • Generate alerts
  • Improve application reliability

Application Insights provides near real-time visibility into the operational health of applications, including AI-powered agents.

Why Monitor Copilot Studio Agents?

Production AI agents interact with users continuously. Monitoring helps answer questions such as:

  • Is the agent available?
  • Are conversations completing successfully?
  • Are responses taking too long?
  • Are external APIs failing?
  • Which topics are most frequently triggered?
  • Where are users abandoning conversations?
  • Are authentication failures occurring?
  • Are knowledge searches succeeding?
  • Is latency increasing?
  • Are recent deployments causing problems?

Without monitoring, identifying these issues can be difficult.

Monitoring Architecture

A typical monitoring architecture includes:

User
↓
Copilot Studio Agent
↓
Conversation Execution
↓
Telemetry Collection
↓
Application Insights
↓
Azure Monitor
↓
Dashboards
Alerts
Analytics
Logs

Every conversation can generate telemetry that is stored for analysis.

What is Telemetry?

Telemetry is operational data automatically collected from applications.

For Copilot Studio agents, telemetry may include:

  • Conversation start
  • Conversation end
  • User session
  • Topic activation
  • Tool execution
  • API calls
  • Response times
  • Exceptions
  • Authentication events
  • Dependency calls
  • Custom events
  • Prompt execution
  • Generative AI activity
  • User feedback
  • Errors

Telemetry provides the raw information used to monitor system health.

Types of Telemetry

Application Insights collects several categories of telemetry.

Requests

Measures requests processed by the agent.

Examples include:

  • User messages
  • Conversation requests
  • HTTP requests
  • API invocations

Useful metrics include:

  • Duration
  • Success rate
  • Failure rate

Dependencies

Tracks external services called by the agent.

Examples include:

  • REST APIs
  • Azure AI Search
  • Dataverse
  • SQL Database
  • SharePoint
  • Power Platform connectors
  • Azure OpenAI
  • Azure AI Foundry
  • External web services

Dependency tracking helps identify slow or failing external systems.

Exceptions

Captures unexpected errors.

Examples include:

  • Authentication failures
  • Timeout exceptions
  • API failures
  • Missing parameters
  • Invalid requests
  • Permission errors

Developers can use exception details to troubleshoot failures.

Traces

Trace telemetry records detailed execution information.

Examples include:

  • Topic execution
  • Diagnostic messages
  • Workflow progress
  • Variable values
  • Decision branches

Traces are especially useful during debugging.

Events

Custom events capture important business activities.

Examples:

  • Order submitted
  • Employee onboarded
  • Ticket created
  • Payment completed
  • Appointment scheduled

Organizations can define custom events for business-specific monitoring.

Availability

Availability monitoring tests whether an application is reachable.

It can detect:

  • Service outages
  • Connectivity failures
  • Regional problems
  • Downtime

Availability tests help ensure production agents remain accessible.

Metrics Commonly Monitored

Common operational metrics include:

  • Total conversations
  • Active users
  • Average response time
  • Request duration
  • API latency
  • Conversation completion rate
  • Conversation abandonment
  • Error count
  • Exception rate
  • Failed requests
  • CPU utilization (supporting resources)
  • Memory utilization (supporting resources)
  • Dependency performance
  • Token consumption (when available)
  • Cost trends

Integrating Copilot Studio with Application Insights

High-level integration typically includes:

  1. Create an Azure Application Insights resource.
  2. Enable monitoring.
  3. Connect the Copilot Studio environment.
  4. Configure telemetry collection.
  5. Deploy the agent.
  6. Review incoming telemetry.
  7. Build dashboards.
  8. Configure alerts.
  9. Monitor production activity.

Azure Monitor Integration

Application Insights is part of Azure Monitor.

Azure Monitor provides:

  • Centralized monitoring
  • Metrics
  • Log Analytics
  • Alerts
  • Dashboards
  • Workbooks
  • Automation
  • Diagnostic settings

Application Insights contributes telemetry to Azure Monitor, where it can be analyzed alongside other Azure resources.

Log Analytics

Telemetry is stored in Log Analytics, enabling powerful querying using Kusto Query Language (KQL).

Administrators can answer questions such as:

  • Which conversations failed today?
  • Which topics generate the most errors?
  • Which users experience timeouts?
  • What APIs are the slowest?
  • Which connector has the highest latency?
  • How many conversations exceeded five seconds?

Example Monitoring Scenarios

Scenario 1

Users report slow responses.

Application Insights reveals:

  • Average response time increased from 2 seconds to 12 seconds.
  • Azure AI Search dependency latency increased dramatically.

The administrator investigates the search service.

Scenario 2

A new deployment causes failures.

Monitoring identifies:

  • Spike in exceptions.
  • Failed API calls.
  • Authentication errors.

The deployment is rolled back.

Scenario 3

An external REST API becomes unavailable.

Application Insights shows:

  • Dependency failures
  • Timeout exceptions
  • Increased conversation failures

Administrators quickly identify the external dependency rather than blaming Copilot Studio.

Dashboards

Application Insights dashboards visualize operational health.

Typical dashboard components include:

  • Conversation volume
  • Requests per minute
  • Active users
  • Success rate
  • Failure rate
  • Exceptions
  • Response times
  • API latency
  • Dependency health
  • Geographic usage
  • Availability
  • Performance trends

Dashboards allow administrators to monitor systems without manually querying logs.

Alerts

Alerts automatically notify administrators when thresholds are exceeded.

Examples include:

  • Response time exceeds five seconds.
  • Error rate exceeds 3%.
  • Availability drops below 99%.
  • API failures increase suddenly.
  • Authentication failures spike.
  • Conversation completion rate decreases.

Alerts can trigger:

  • Email
  • SMS
  • Microsoft Teams notifications
  • Azure Automation
  • Logic Apps
  • Webhooks

Distributed Tracing

Many enterprise agents call multiple services during a single conversation.

Example:

User
↓
Copilot Studio
↓
Azure AI Search
↓
REST API
↓
Dataverse
↓
Azure AI Foundry
↓
Response

Application Insights correlates these operations into a single end-to-end transaction.

This allows administrators to identify exactly where delays occur.

Correlation IDs

Each conversation can be assigned a correlation ID.

This enables:

  • End-to-end tracing
  • Cross-service diagnostics
  • Root cause analysis
  • Log correlation
  • Easier troubleshooting

Correlation IDs are especially valuable in distributed AI systems.

Monitoring Generative AI Operations

Application Insights can help monitor:

  • Prompt execution
  • Model latency
  • API failures
  • Retrieval operations
  • Tool execution
  • Conversation completion
  • Dependency failures
  • User feedback events

While model-specific metrics may come from Azure AI Foundry or Azure OpenAI, Application Insights provides operational telemetry surrounding those interactions.

Security Considerations

Monitoring should avoid collecting sensitive information.

Best practices include:

  • Avoid storing secrets.
  • Minimize personal information.
  • Mask sensitive values.
  • Follow organizational compliance policies.
  • Apply RBAC to monitoring resources.
  • Encrypt telemetry in transit and at rest.
  • Retain logs according to governance requirements.

Cost Considerations

Application Insights pricing depends largely on:

  • Data ingestion volume
  • Log retention
  • Query frequency
  • Exported telemetry

Organizations should balance monitoring detail with storage costs.

Strategies include:

  • Sample telemetry.
  • Adjust retention periods.
  • Remove unnecessary events.
  • Archive historical logs.

Best Practices

  • Enable monitoring before production deployment.
  • Create dashboards for key performance indicators.
  • Configure proactive alerts.
  • Monitor dependency health.
  • Use distributed tracing.
  • Track conversation completion rates.
  • Review exceptions regularly.
  • Use KQL to investigate issues.
  • Protect sensitive telemetry.
  • Continuously optimize based on monitoring insights.

Common Exam Tips

For the AB-620 exam, remember the following:

  • Application Insights is part of Azure Monitor.
  • It provides application performance monitoring (APM).
  • It collects telemetry from running applications.
  • Telemetry includes requests, dependencies, exceptions, traces, events, and availability data.
  • Dependency monitoring helps diagnose failures in external systems.
  • Log Analytics uses Kusto Query Language (KQL) for querying telemetry.
  • Alerts can automatically notify administrators of operational issues.
  • Distributed tracing correlates activity across multiple services.
  • Correlation IDs enable end-to-end diagnostics.
  • Monitoring supports performance optimization, troubleshooting, and operational reliability.

Practice Exam Questions

Question 1

An administrator wants to determine why users are experiencing slow responses from a Copilot Studio agent. Which Azure service provides detailed performance telemetry for troubleshooting?

A. Azure Storage

B. Azure Application Insights

C. Microsoft Entra ID

D. Azure Key Vault

Answer: B

Explanation: Application Insights collects detailed telemetry such as response times, dependency performance, and exceptions, making it the primary service for diagnosing performance issues.

Question 2

Which type of Application Insights telemetry tracks calls from a Copilot Studio agent to Azure AI Search or external REST APIs?

A. Requests

B. Exceptions

C. Dependencies

D. Availability

Answer: C

Explanation: Dependency telemetry measures calls to external services, databases, connectors, APIs, and Azure resources, allowing administrators to identify slow or failing dependencies.

Question 3

A developer wants to investigate authentication failures generated during agent execution. Which telemetry type should they examine first?

A. Exceptions

B. Availability

C. Metrics

D. Workbooks

Answer: A

Explanation: Authentication failures typically generate exception telemetry, which records detailed information about errors encountered during execution.

Question 4

What is the primary purpose of distributed tracing in Application Insights?

A. Encrypt conversation history

B. Automatically translate telemetry

C. Compress monitoring data

D. Correlate activity across multiple services in a single transaction

Answer: D

Explanation: Distributed tracing connects telemetry from multiple services involved in processing a single request, enabling end-to-end diagnostics.

Question 5

Which language is used to query Application Insights data stored in Log Analytics?

A. T-SQL

B. Power Query M

C. DAX

D. Kusto Query Language (KQL)

Answer: D

Explanation: Log Analytics uses Kusto Query Language (KQL) to query, filter, summarize, and analyze telemetry data.

Question 6

An operations team wants to receive an email whenever an agent’s average response time exceeds five seconds. Which Azure Monitor capability should they configure?

A. Alerts

B. Availability tests

C. Workbooks

D. Sampling

Answer: A

Explanation: Azure Monitor alerts automatically notify administrators when configured thresholds or conditions are met.

Question 7

Which monitoring metric would BEST help determine whether users are abandoning conversations before completion?

A. CPU utilization

B. Conversation completion and abandonment rates

C. Azure subscription quota

D. Virtual machine availability

Answer: B

Explanation: Completion and abandonment metrics directly measure how successfully users finish conversations with the agent.

Question 8

Why are correlation IDs valuable when troubleshooting AI agents?

A. They reduce Azure costs.

B. They increase model accuracy.

C. They link telemetry across multiple services for a single conversation.

D. They automatically encrypt logs.

Answer: C

Explanation: Correlation IDs associate related telemetry from different services, making it easier to trace a request from start to finish.

Question 9

Which best practice helps protect sensitive information when using Application Insights?

A. Store authentication secrets in telemetry for debugging.

B. Collect every possible user input permanently.

C. Disable encryption to improve performance.

D. Mask sensitive data and apply role-based access control (RBAC).

Answer: D

Explanation: Sensitive information should be masked or excluded from telemetry, and access should be restricted using RBAC to support security and compliance.

Question 10

What is the primary benefit of monitoring external dependencies such as Azure AI Search, Dataverse, and REST APIs?

A. It automatically upgrades connectors.

B. It identifies latency and failures occurring outside the Copilot Studio agent itself.

C. It eliminates the need for application logging.

D. It reduces token consumption by language models.

Answer: B

Explanation: Dependency monitoring helps determine whether performance issues or failures originate in external services rather than within the agent, significantly speeding up root cause analysis.

Go to the AB-620 Exam Prep Hub main page

AB-620, Agentic AI, AI, Azure AI, Generative AI, Microsoft Certification

Configure custom prompts to use the Foundry model catalog (AB-620 Exam Prep)

 
This post is a part of the AB-620: Designing and Building Integrated AI Agent Solutions in Copilot Studio Exam Prep Hub.
This topic falls under these sections:
Integrate and extend agents in Copilot Studio (40–45%)
   --> Integrate agents with Azure
      --> Configure custom prompts to use the Foundry model catalog

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Overview

As organizations build increasingly sophisticated AI agents in Microsoft Copilot Studio, they often require more control over which large language models (LLMs) are used and how those models generate responses. While Copilot Studio includes powerful built-in generative AI capabilities, many enterprise scenarios benefit from connecting to models hosted in Azure AI Foundry (formerly Azure AI Studio).

Azure AI Foundry provides access to a large catalog of foundation models from Microsoft, OpenAI, Meta, Mistral AI, Cohere, Hugging Face, and many other providers. These models can be deployed within an Azure subscription and securely consumed by applications and AI agents.

One of the key capabilities covered in the AB-620 exam is configuring custom prompts in Copilot Studio that leverage models deployed through the Azure AI Foundry model catalog. This enables organizations to tailor agent behavior, use specialized models, implement reusable prompts, and satisfy governance requirements while maintaining enterprise security.

What is Azure AI Foundry?

Azure AI Foundry is Microsoft’s unified platform for building, evaluating, deploying, securing, and managing AI solutions.

It provides:

  • Model catalog
  • Prompt engineering tools
  • AI evaluation capabilities
  • Safety systems
  • Deployment management
  • Model monitoring
  • Responsible AI controls
  • Agent development tools
  • Integration with Copilot Studio

Rather than relying only on the default Copilot model, organizations can deploy one or more models within Azure AI Foundry and make them available to enterprise applications.

What is the Foundry Model Catalog?

The Model Catalog is a centralized repository containing hundreds of AI models.

Examples include:

  • GPT models
  • Phi models
  • Llama models
  • Mistral models
  • Cohere Command models
  • DeepSeek models (where available)
  • Open-source Hugging Face models
  • Vision models
  • Embedding models
  • Speech models
  • Multimodal models

Each model includes information such as:

  • Provider
  • Version
  • Licensing
  • Supported tasks
  • Context window
  • Token limits
  • Pricing
  • Deployment options
  • Performance benchmarks

Why Use the Model Catalog?

Organizations may choose custom models because they need:

  • Better reasoning
  • Lower latency
  • Lower cost
  • Longer context windows
  • Specialized coding abilities
  • Multilingual support
  • Vision processing
  • Image understanding
  • Document analysis
  • Industry-specific performance

Instead of using one model for every task, different prompts can target different deployed models.

What Are Custom Prompts?

A custom prompt is a reusable prompt template that defines how an LLM should perform a task.

Rather than asking the model a simple question, a custom prompt provides detailed instructions, context, formatting rules, and constraints.

Example:

Instead of:

Summarize this document.

A custom prompt might specify:

You are a financial analyst. Summarize this quarterly earnings report in less than 300 words. Highlight revenue changes, operating margin, risks, opportunities, and executive guidance. Produce the output as a Markdown table followed by three bullet points.

The additional instructions produce far more consistent outputs.

Benefits of Custom Prompts

Advantages include:

  • Consistent responses
  • Reusable instructions
  • Better formatting
  • Reduced hallucinations
  • Improved grounding
  • Easier maintenance
  • Centralized governance
  • Standardized business logic

How Copilot Studio Uses Foundry Models

The high-level workflow is:

  1. Deploy a model in Azure AI Foundry.
  2. Configure the deployment endpoint.
  3. Create or connect Azure AI resources.
  4. Connect Copilot Studio.
  5. Create a custom prompt.
  6. Select the deployed model.
  7. Pass user input into the prompt.
  8. Receive generated output.
  9. Continue the conversation.

The user typically does not know which model produced the response.

Typical Architecture

User
↓
Copilot Studio Agent
↓
Custom Prompt
↓
Azure AI Foundry
↓
Selected Model Deployment
↓
Generated Response
↓
Agent Response

Components Involved

A complete solution typically includes:

  • Copilot Studio
  • Azure AI Foundry
  • Azure AI Foundry Project
  • Model deployment
  • Azure AI Services resource
  • Authentication
  • Prompt template
  • Enterprise data
  • Optional Azure AI Search

Creating a Model Deployment

Before a prompt can use a model, the model must first be deployed.

Typical steps include:

  • Browse the Model Catalog.
  • Select a model.
  • Review licensing.
  • Choose deployment type.
  • Configure capacity.
  • Deploy the endpoint.
  • Test the deployment.
  • Secure the deployment.

The deployment creates an endpoint that applications can call.

Connecting Copilot Studio to Foundry

The connection typically involves:

  • Azure authentication
  • Managed identity or service principal
  • Endpoint configuration
  • Permissions
  • Environment configuration

After configuration, Copilot Studio can invoke deployed models as part of prompt execution.

Prompt Design Best Practices

Good prompts generally include:

Role

Tell the model who it is.

Example:

“You are an HR compliance specialist.”

Goal

Describe the objective.

Example:

“Review employee policies.”

Context

Provide supporting information.

Example:

“The organization operates in healthcare.”

Instructions

Explain exactly what should happen.

Example:

“Identify compliance risks.”

Constraints

Limit undesirable behavior.

Example:

  • Don’t speculate.
  • Use only supplied information.
  • Return JSON.

Output Format

Specify the expected structure.

Example:

Summary
Risks
Recommendations
Confidence Score

Prompt Variables

Custom prompts commonly accept variables.

Examples include:

  • User question
  • Customer name
  • Product
  • Ticket number
  • Region
  • Language
  • Conversation history
  • Retrieved documents

Variables make one prompt reusable for thousands of requests.

Example Prompt

Role:
You are an insurance claims specialist.
Task:
Review the submitted claim.
Context:
Use only supplied documents.
Output:
Return:
• Claim summary
• Fraud indicators
• Missing information
• Recommended next steps
Do not invent facts.

Choosing the Right Model

Different prompts benefit from different models.

Examples:

Customer support

  • Low latency
  • Low cost

Legal analysis

  • High reasoning ability
  • Large context window

Coding

  • Strong code generation

Document summarization

  • Long context support

Translation

  • Strong multilingual capabilities

Model Selection Considerations

Factors include:

  • Cost
  • Latency
  • Accuracy
  • Context length
  • Availability
  • Geographic region
  • Compliance requirements
  • Safety capabilities
  • Throughput
  • Scalability

Responsible AI Considerations

When configuring prompts, organizations should:

  • Avoid biased instructions.
  • Protect confidential information.
  • Minimize unnecessary personal data.
  • Ground responses in enterprise knowledge.
  • Validate generated output.
  • Apply content filtering.
  • Review prompts regularly.
  • Monitor model behavior.

Prompt Evaluation

Azure AI Foundry provides tools for evaluating prompts.

Organizations can measure:

  • Accuracy
  • Relevance
  • Faithfulness
  • Groundedness
  • Helpfulness
  • Safety
  • Toxicity
  • Hallucination rate
  • Latency
  • Cost

Evaluation helps determine whether prompt changes actually improve performance.

Prompt Versioning

As prompts evolve, organizations often maintain multiple versions.

Versioning enables:

  • Rollback
  • Testing
  • Controlled releases
  • A/B testing
  • Governance
  • Documentation
  • Change tracking

Common Enterprise Scenarios

Organizations frequently use Foundry-backed prompts for:

  • Customer support
  • IT help desks
  • HR assistants
  • Financial reporting
  • Contract analysis
  • Healthcare documentation
  • Manufacturing troubleshooting
  • Knowledge management
  • Compliance reviews
  • Executive reporting

Best Practices

  • Keep prompts focused on one objective.
  • Provide explicit instructions.
  • Specify output formats.
  • Use variables instead of hardcoding values.
  • Ground prompts with enterprise knowledge whenever possible.
  • Test prompts using multiple scenarios.
  • Monitor latency and token consumption.
  • Select the smallest model that satisfies business requirements.
  • Evaluate prompts continuously.
  • Version prompts before making production changes.

Common Exam Tips

For the AB-620 exam, remember:

  • Azure AI Foundry hosts deployed AI models.
  • The Model Catalog contains many foundation models from multiple providers.
  • Models must typically be deployed before they can be used.
  • Custom prompts provide reusable instructions for LLM interactions.
  • Prompt variables enable reuse across many conversations.
  • Azure AI Search can be combined with Foundry models for grounded responses.
  • Prompt evaluation measures quality and safety.
  • Responsible AI practices remain essential when using custom prompts.
  • Different prompts may use different deployed models.
  • Prompt engineering significantly affects response quality.

10 Practice Exam Questions

Question 1

An organization wants multiple Copilot Studio agents to use the same standardized instructions when summarizing financial reports. What is the best solution?

A. Create a custom prompt that all agents can reuse.

B. Rewrite the instructions in every topic.

C. Store the instructions inside Adaptive Cards.

D. Add the instructions to every user question.

Answer: A

Explanation: A reusable custom prompt centralizes instructions, promotes consistency, and simplifies maintenance across multiple agents.

Question 2

Which Azure AI Foundry component provides access to available foundation models?

A. AI Hub

B. Model Catalog

C. Prompt Flow

D. Azure Monitor

Answer: B

Explanation: The Model Catalog is the repository for browsing, evaluating, and selecting supported foundation models before deployment.

Question 3

A prompt instructs a model to answer only using retrieved enterprise documentation. What primary benefit does this provide?

A. Faster model deployment

B. Reduced token usage

C. Better grounding and fewer hallucinations

D. Automatic translation

Answer: C

Explanation: Restricting responses to trusted enterprise content improves factual accuracy and reduces unsupported or fabricated responses.

Question 4

Before a Copilot Studio agent can use a model from Azure AI Foundry, what must typically occur?

A. The model must be exported to Dataverse.

B. A Power Automate flow must be created.

C. A custom connector must be installed.

D. The selected model must be deployed.

Answer: D

Explanation: Models in the catalog are not directly consumable until they have been deployed to an endpoint.

Question 5

Which prompt component tells the model how it should behave?

A. Context

B. Output format

C. Role

D. Variables

Answer: C

Explanation: The role establishes the model’s persona or expertise, such as “You are a financial analyst.”

Question 6

Why should prompt variables be used instead of hard-coded values?

A. They improve model licensing.

B. They allow prompts to be reused for different inputs.

C. They reduce Azure subscription costs.

D. They eliminate authentication requirements.

Answer: B

Explanation: Variables enable a single prompt template to process many different user requests without modification.

Question 7

An organization compares several prompts for accuracy, groundedness, latency, and safety before production deployment. Which Azure AI Foundry capability are they using?

A. Deployment scaling

B. Resource monitoring

C. Model catalog browsing

D. Prompt evaluation

Answer: D

Explanation: Prompt evaluation measures prompt quality using metrics such as accuracy, groundedness, safety, and response quality.

Question 8

Which consideration is MOST important when selecting a model for a custom prompt?

A. The icon displayed in the model catalog

B. The browser used by administrators

C. The business requirements, including latency, cost, and reasoning capability

D. The number of Copilot Studio topics

Answer: C

Explanation: Model selection should align with workload requirements, balancing performance, cost, context length, and reasoning ability.

Question 9

A prompt specifies that output must always be returned as JSON with predefined fields. What prompt design principle is being applied?

A. Context injection

B. Output formatting

C. Authentication

D. Content indexing

Answer: B

Explanation: Explicitly defining the output structure increases consistency and simplifies downstream processing.

Question 10

Why should organizations maintain multiple versions of important production prompts?

A. To increase model context length

B. To reduce Azure subscription costs

C. To enable rollback, testing, governance, and controlled deployment of prompt changes

D. To eliminate authentication requirements

Answer: C

Explanation: Prompt versioning supports change management, testing, auditing, rollback, and safer deployment of updates without disrupting production agents.

Go to the AB-620 Exam Prep Hub main page

AB-620, AI, Azure AI, Generative AI, Microsoft Certification

Configure generative answers by using Azure AI Search with Foundry (AB-620 Exam Prep)

 
This post is a part of the AB-620: Designing and Building Integrated AI Agent Solutions in Copilot Studio Exam Prep Hub.
This topic falls under these sections:
Integrate and extend agents in Copilot Studio (40–45%)
   --> Integrate agents with Azure
      --> Configure generative answers by using Azure AI Search with Foundry

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Overview

One of the most powerful capabilities in Microsoft Copilot Studio is the ability to generate grounded, AI-powered responses using enterprise knowledge instead of relying solely on predefined topics. By integrating Azure AI Search with Azure AI Foundry, organizations can build intelligent agents that retrieve relevant information from enterprise content and use large language models (LLMs) to generate accurate, contextual responses.

For the AB-620: Designing and Building Integrated AI Agent Solutions in Copilot Studio exam, you should understand how Azure AI Search, Azure AI Foundry, and Copilot Studio work together to provide Retrieval-Augmented Generation (RAG) experiences.

Learning Objectives

After studying this topic, you should be able to:

  • Explain how Azure AI Search integrates with Copilot Studio.
  • Understand the role of Azure AI Foundry in generative AI solutions.
  • Configure generative answers using Azure AI Search indexes.
  • Understand Retrieval-Augmented Generation (RAG).
  • Configure enterprise knowledge grounding.
  • Understand indexing, chunking, embeddings, and vector search.
  • Apply security and governance best practices.
  • Troubleshoot common configuration issues.

What is Azure AI Foundry?

Azure AI Foundry is Microsoft’s unified platform for building, evaluating, deploying, and managing AI applications and agents.

It provides developers with tools to:

  • Build AI applications
  • Manage AI models
  • Connect enterprise knowledge
  • Evaluate AI responses
  • Deploy production AI solutions
  • Monitor model performance

When integrated with Copilot Studio, Azure AI Foundry supplies the AI models and orchestration capabilities that generate responses based on retrieved enterprise knowledge.

What is Azure AI Search?

Azure AI Search is Microsoft’s enterprise search platform.

Its responsibilities include:

  • Indexing enterprise content
  • Creating searchable knowledge repositories
  • Supporting keyword search
  • Supporting semantic search
  • Supporting vector search
  • Ranking relevant documents
  • Returning content used for grounding AI responses

Rather than generating answers from model training alone, Copilot retrieves relevant documents through Azure AI Search before asking the LLM to formulate an answer.

Understanding Retrieval-Augmented Generation (RAG)

This topic heavily emphasizes Retrieval-Augmented Generation (RAG).

Instead of relying entirely on the LLM’s pretrained knowledge:

  1. User asks a question.
  2. Azure AI Search searches indexed enterprise content.
  3. Relevant passages are retrieved.
  4. Retrieved content is passed to the LLM in Azure AI Foundry.
  5. The LLM generates a grounded response using that retrieved information.

Benefits include:

  • More accurate responses
  • Reduced hallucinations
  • Current enterprise information
  • Permission-aware answers
  • Citations and traceability (when configured)

High-Level Architecture

User
   │
   ▼
Copilot Studio
   │
   ▼
Azure AI Search
(Search Index)
   │
   ▼
Relevant Documents
   │
   ▼
Azure AI Foundry
(LLM)
   │
   ▼
Grounded Response
   │
   ▼
User

Components of the Solution

1. Enterprise Data Sources

Examples include:

  • SharePoint Online
  • OneDrive
  • Azure Blob Storage
  • SQL databases
  • Microsoft Fabric
  • PDF documents
  • Microsoft Teams files
  • Websites
  • Knowledge bases

2. Data Connectors

Connectors import content into Azure AI Search.

They support:

  • Scheduled indexing
  • Incremental updates
  • Metadata extraction
  • Content synchronization

3. Azure AI Search Index

The search index stores:

  • Text content
  • Metadata
  • Searchable fields
  • Filterable fields
  • Vector embeddings
  • Semantic configurations

Indexes are optimized for rapid retrieval.

4. Embeddings

Before semantic search can occur, documents are converted into numerical vectors called embeddings.

Embeddings allow the system to:

  • Compare meaning instead of exact wording
  • Find similar concepts
  • Improve retrieval accuracy
  • Support multilingual search

Example:

Question:

“How much vacation do employees receive?”

The document may say:

“Annual leave entitlement is 20 days.”

Keyword search may miss this.

Embedding search understands that both discuss vacation policies.

5. Chunking

Large documents are automatically divided into smaller sections.

Chunking improves:

  • Retrieval precision
  • Context quality
  • Token efficiency
  • Response accuracy

Poor chunk sizes often produce poor RAG performance.

6. Semantic Search

Semantic ranking considers:

  • Meaning
  • Intent
  • Context
  • Related concepts

Rather than matching words alone.

7. Vector Search

Vector search compares embedding similarity.

Advantages:

  • Better natural language understanding
  • Improved document matching
  • Better enterprise Q&A performance

Many enterprise deployments combine:

  • Keyword search
  • Semantic search
  • Vector search

Configuring Generative Answers

Typical configuration steps include:

Step 1

Create an Azure AI Search service.

Step 2

Create a search index.

Step 3

Import enterprise data.

Step 4

Configure indexing schedules.

Step 5

Enable semantic ranking.

Step 6

Configure vector search (if supported).

Step 7

Connect Azure AI Search to Azure AI Foundry.

Step 8

Connect the Foundry project to Copilot Studio.

Step 9

Enable Generative Answers.

Step 10

Test grounded responses.

Knowledge Grounding

Grounding ensures responses originate from approved enterprise information rather than model memory.

Grounding helps:

  • Improve accuracy
  • Reduce hallucinations
  • Maintain compliance
  • Support trustworthy AI

Security Considerations

Authentication typically uses:

  • Microsoft Entra ID
  • Managed identities
  • Role-based access control (RBAC)

Authorization should ensure:

  • Only authorized documents are searchable.
  • Sensitive data is protected.
  • User permissions are respected.

Monitoring

Administrators should monitor:

  • Search latency
  • Retrieval accuracy
  • Query success rates
  • Failed searches
  • Index freshness
  • Hallucination frequency
  • User feedback
  • Token consumption

Common Design Best Practices

Build high-quality indexes

Avoid indexing:

  • Duplicate content
  • Obsolete files
  • Incomplete documentation

Keep indexes current

Use incremental indexing.

Avoid stale enterprise knowledge.

Optimize chunk size

Too small:

  • Missing context

Too large:

  • Lower retrieval precision

Enable semantic ranking

Semantic ranking typically improves enterprise Q&A accuracy.

Use vector search

Vector search improves:

  • Similarity matching
  • Natural language understanding
  • Complex enterprise queries

Apply least-privilege security

Grant only the permissions required.

Validate responses

Test with:

  • Ambiguous questions
  • Synonyms
  • Long documents
  • Missing data
  • Permission-restricted users

Common Exam Scenarios

You should know when:

  • Azure AI Search should be used instead of static Topics.
  • Enterprise knowledge requires semantic search.
  • Vector search improves retrieval.
  • Azure AI Foundry generates responses after retrieval.
  • RAG is preferable to relying solely on an LLM.
  • Grounding reduces hallucinations.
  • Search indexes require re-indexing after significant data changes.
  • Semantic models and enterprise permissions affect response quality.

Exam Tips

  • Azure AI Search retrieves information—it does not generate responses.
  • Azure AI Foundry hosts and orchestrates AI models that generate responses.
  • Copilot Studio coordinates the conversation and calls Azure services.
  • RAG combines retrieval with generation to improve answer quality.
  • Embeddings power vector search.
  • Chunking directly affects retrieval accuracy.
  • Semantic search improves relevance beyond keyword matching.
  • Grounded responses are generally preferred over responses based solely on pretrained model knowledge.

Practice Exam Questions

Question 1

A company wants its Copilot Studio agent to answer employee policy questions using current HR documents instead of relying solely on the LLM’s pretrained knowledge. Which architecture should they implement?

A. Static Topics only

B. Retrieval-Augmented Generation using Azure AI Search and Azure AI Foundry

C. Power Automate flows only

D. Adaptive Cards with variables only

Correct Answer: B

Explanation: RAG retrieves relevant enterprise documents through Azure AI Search and passes them to Azure AI Foundry, allowing the LLM to generate grounded responses based on current organizational content.

Question 2

What is Azure AI Search primarily responsible for in a Copilot Studio generative answers solution?

A. Hosting large language models

B. Training AI models

C. Retrieving relevant enterprise content from indexed data

D. Managing Copilot Studio topics

Correct Answer: C

Explanation: Azure AI Search indexes and retrieves relevant enterprise content. It does not host or train language models.

Question 3

What is the primary purpose of document chunking during indexing?

A. Compress documents for storage

B. Improve retrieval accuracy by dividing large documents into manageable sections

C. Encrypt enterprise documents

D. Eliminate duplicate records

Correct Answer: B

Explanation: Chunking divides large documents into smaller, context-rich segments, enabling more precise retrieval during RAG.

Question 4

Which Azure service generates the natural language response after Azure AI Search retrieves relevant content?

A. Azure AI Foundry

B. Azure Blob Storage

C. Azure Monitor

D. Azure Key Vault

Correct Answer: A

Explanation: Azure AI Foundry provides access to large language models that synthesize retrieved content into conversational responses.

Question 5

Which technology enables Azure AI Search to retrieve documents based on semantic similarity rather than exact keyword matches?

A. Managed identities

B. RBAC

C. Vector embeddings

D. Power Automate

Correct Answer: C

Explanation: Vector embeddings represent document meaning numerically, enabling semantic similarity searches.

Question 6

Why is grounding considered an important capability in generative AI solutions?

A. It increases token limits.

B. It improves model training speed.

C. It ensures responses are based on trusted enterprise knowledge.

D. It replaces semantic search.

Correct Answer: C

Explanation: Grounding reduces hallucinations by anchoring AI responses to retrieved organizational content.

Question 7

An organization updates its policy documents every night. What is the best way to ensure the Copilot agent uses the latest information?

A. Retrain the language model nightly.

B. Configure scheduled or incremental indexing in Azure AI Search.

C. Restart Copilot Studio every morning.

D. Recreate the search index daily.

Correct Answer: B

Explanation: Scheduled or incremental indexing updates the search index efficiently without requiring complete re-creation or model retraining.

Question 8

Which component is responsible for coordinating the conversation and invoking Azure AI Search and Azure AI Foundry?

A. Azure Monitor

B. Azure AI Search

C. Azure AI Foundry

D. Copilot Studio

Correct Answer: D

Explanation: Copilot Studio orchestrates the conversational flow, calling Azure AI Search for retrieval and Azure AI Foundry for response generation.

Question 9

Which statement best describes vector search?

A. It searches only document titles.

B. It compares numerical representations of meaning rather than exact words.

C. It retrieves only structured database records.

D. It replaces semantic ranking entirely.

Correct Answer: B

Explanation: Vector search uses embeddings to compare semantic similarity, allowing retrieval of conceptually related content even when wording differs.

Question 10

A developer notices that the agent frequently provides incomplete answers because relevant information is split across large documents. Which improvement is most appropriate?

A. Disable semantic search.

B. Increase the model temperature.

C. Optimize document chunk sizes during indexing.

D. Replace Azure AI Search with keyword search only.

Correct Answer: C

Explanation: Appropriate chunk sizing improves retrieval quality by ensuring each indexed segment contains enough context while remaining focused, leading to more complete and accurate grounded responses.

Go to the AB-620 Exam Prep Hub main page

AB-620, AI, Generative AI, Microsoft Certification

Configure generative answers node (AB-620 Exam Prep)

 
This post is a part of the AB-620: Designing and Building Integrated AI Agent Solutions in Copilot Studio Exam Prep Hub.
This topic falls under these sections:
Plan and configure agent solutions (30–35%)
   --> Configure topics
      --> Configure generative answers node

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

The Generative Answers node is one of the most powerful features in Microsoft Copilot Studio. Unlike traditional chatbot responses that rely solely on pre-authored conversation paths, the Generative Answers node enables an agent to dynamically generate responses by retrieving information from approved knowledge sources and using a large language model (LLM) to compose natural, conversational answers.

For the AB-620 certification exam, you should understand how to configure the Generative Answers node, when to use it, how it retrieves information, how it differs from traditional topic responses, and how to optimize it for enterprise scenarios.

Learning Objectives

After studying this topic, you should be able to:

  • Explain the purpose of the Generative Answers node.
  • Understand how retrieval-augmented generation (RAG) works in Copilot Studio.
  • Configure the Generative Answers node within a topic.
  • Select appropriate enterprise knowledge sources.
  • Understand grounding and context.
  • Configure citations.
  • Control response generation behavior.
  • Recognize best practices for enterprise AI solutions.
  • Identify common exam scenarios.

What is the Generative Answers Node?

The Generative Answers node is a conversation node that enables Copilot Studio to generate AI-powered responses using one or more approved knowledge sources.

Unlike a standard Message node, which displays predefined text, the Generative Answers node creates responses dynamically based on retrieved information.

Example:

User asks:

“What are the company’s reimbursement policies for travel expenses?”

Instead of following a scripted topic, the Generative Answers node:

  1. Searches configured knowledge sources.
  2. Retrieves relevant documents.
  3. Grounds the AI model using the retrieved content.
  4. Generates a conversational answer.
  5. Optionally includes citations.

Why Use the Generative Answers Node?

Traditional topics work well for:

  • Frequently asked questions
  • Structured workflows
  • Decision trees
  • Business processes
  • Data collection

However, organizations often have thousands of documents that cannot realistically be converted into authored topics.

Examples include:

  • Employee handbooks
  • HR policies
  • Product documentation
  • Technical manuals
  • Knowledge base articles
  • Compliance documentation
  • Training materials
  • Internal procedures

The Generative Answers node allows the agent to answer questions directly from these sources without requiring authors to create individual conversation branches.

Traditional Topics vs. Generative Answers

Traditional TopicsGenerative Answers
Scripted responsesAI-generated responses
Predictable conversation flowDynamic conversational responses
Manual authoringKnowledge-driven generation
Best for business processesBest for knowledge retrieval
Requires maintenance of many topicsUses existing enterprise knowledge
Limited flexibilityHandles a wide variety of questions

Many enterprise agents combine both approaches.

How the Generative Answers Node Works

The process follows a Retrieval-Augmented Generation (RAG) pattern.

User Question
        ↓
Generative Answers Node
        ↓
Search Knowledge Sources
        ↓
Retrieve Relevant Content
        ↓
Ground the AI Model
        ↓
Generate Natural Language Response
        ↓
Display Answer with Citations

Rather than relying solely on the language model’s training data, the response is grounded in current enterprise knowledge.

What is Grounding?

Grounding is the process of providing relevant source material to the AI model before it generates a response.

Without grounding:

The model relies primarily on its pretrained knowledge.

With grounding:

The model bases its answer on approved enterprise content.

Grounding helps improve:

  • Accuracy
  • Relevance
  • Consistency
  • Trustworthiness
  • Compliance

Grounding is one of the most important concepts on the AB-620 exam.

Retrieval-Augmented Generation (RAG)

RAG combines two technologies:

  1. Information retrieval
  2. Large language model generation

Workflow:

User asks question
        ↓
Search enterprise knowledge
        ↓
Retrieve relevant documents
        ↓
Pass retrieved content to LLM
        ↓
Generate grounded response

Benefits include:

  • Reduced hallucinations
  • Current information
  • Organization-specific answers
  • Better transparency
  • Source citations

Supported Knowledge Sources

The Generative Answers node can retrieve information from multiple knowledge sources.

Common sources include:

  • Microsoft SharePoint
  • Microsoft OneDrive
  • Public websites
  • Internal websites
  • Azure AI Search indexes
  • Dataverse
  • Microsoft Fabric (through supported integrations)
  • Uploaded documents
  • Enterprise document repositories
  • Custom knowledge connectors

Organizations often combine several sources to create a unified knowledge experience.

Enterprise Knowledge Sources

Typical enterprise repositories include:

Human Resources

  • Employee handbook
  • Leave policies
  • Benefits guides

IT

  • Help desk documentation
  • Software manuals
  • Troubleshooting guides

Legal

  • Compliance policies
  • Governance documents
  • Regulatory guidance

Sales

  • Product documentation
  • Pricing guides
  • Competitive information

Customer Support

  • Knowledge articles
  • FAQ databases
  • Troubleshooting documentation

Adding a Generative Answers Node

Within a topic:

Trigger
     ↓
Ask Question
     ↓
Generative Answers Node
     ↓
Response

The node is inserted into the conversation where dynamic information retrieval is required.

Configuring Knowledge Sources

When configuring the node, developers specify where information should be retrieved.

Typical configuration options include:

  • One or more knowledge sources
  • Search scope
  • Search filters
  • Authentication
  • Citation behavior
  • Response generation options

Well-designed knowledge selection significantly improves answer quality.

Search Process

When a user asks a question:

  1. User query is analyzed.
  2. Relevant documents are identified.
  3. Best matches are selected.
  4. Relevant passages are extracted.
  5. Retrieved passages are provided to the AI model.
  6. AI generates the response.

The AI does not typically process every document in the repository—only the most relevant retrieved content.

Conversation Context

The Generative Answers node uses conversation context to improve relevance.

Example:

User:

Tell me about vacation policies.

Later:

What about contractors?

The second question is interpreted in the context of the first discussion, resulting in a more relevant response.

Maintaining conversational context creates a more natural interaction.

Using Variables

The node can incorporate variables collected earlier in the conversation.

Example:

Department = Finance

User asks:

What training is required?

The search can prioritize Finance-specific documentation, resulting in more targeted answers.

Citations

One of the major strengths of the Generative Answers node is the ability to include citations.

Example:

According to the Employee Handbook…

or

Source: HR Benefits Guide

Benefits include:

  • Increased transparency
  • Greater user confidence
  • Easier verification
  • Regulatory compliance
  • Reduced misinformation

Many enterprise deployments enable citations by default.

Benefits of Citations

Citations help users:

  • Verify information.
  • Locate original documents.
  • Confirm policy wording.
  • Build trust in AI-generated responses.
  • Distinguish grounded responses from general AI knowledge.

Organizations operating in regulated industries often consider citations essential.

When to Use the Generative Answers Node

Ideal scenarios include:

  • Employee self-service
  • Policy lookup
  • Technical documentation
  • Product information
  • Internal procedures
  • Knowledge management
  • Customer support
  • Training assistance
  • Compliance guidance

It is particularly effective when answers are based on existing documentation rather than transactional data.

When Not to Use the Generative Answers Node

Avoid using it when:

  • A deterministic business workflow is required.
  • Users must complete structured forms.
  • API calls are needed to update external systems.
  • Financial transactions must be executed.
  • Precise branching logic is required.
  • Data collection drives subsequent processing.

In these cases, traditional topics, actions, or agent flows are more appropriate.

Combining Topics and Generative Answers

Many enterprise agents use a hybrid design.

Example:

User asks question
        ↓
Topic starts
        ↓
Collect customer information
        ↓
Call API
        ↓
Generative Answers Node
        ↓
Display response
        ↓
Continue workflow

This combines structured processes with AI-powered knowledge retrieval.

Response Quality

High-quality responses depend on:

  • Accurate source documents
  • Well-organized knowledge repositories
  • Updated content
  • Appropriate search configuration
  • Effective grounding
  • Clear user questions

Even the best AI model cannot compensate for outdated or inaccurate source material.

Best Practices

When configuring the Generative Answers node:

  • Use trusted enterprise knowledge sources.
  • Remove outdated documents from repositories.
  • Organize content logically.
  • Enable citations whenever appropriate.
  • Test common user questions.
  • Use conversation context effectively.
  • Combine with traditional topics where needed.
  • Limit knowledge sources to those relevant for the intended audience.
  • Regularly review answer quality and user feedback.
  • Monitor changes to enterprise documentation to ensure responses remain accurate.

Exam Tips

For the AB-620 exam, remember:

  • The Generative Answers node retrieves information from configured knowledge sources rather than relying solely on the language model.
  • Retrieval-Augmented Generation (RAG) combines search with AI-generated responses.
  • Grounding improves response accuracy and reduces hallucinations.
  • Citations increase transparency and trust.
  • Traditional topics are best for deterministic workflows, while Generative Answers is best for knowledge retrieval.
  • Conversation context and variables can improve the relevance of generated responses.
  • Knowledge quality directly affects response quality.
  • Enterprise AI solutions commonly combine authored topics with Generative Answers to provide both structured workflows and dynamic knowledge retrieval.

Best Practices for Configuring Generative Answers

Microsoft recommends treating Generative Answers as a retrieval-augmented capability rather than allowing unrestricted AI generation. Well-designed agents retrieve authoritative information from trusted sources and then generate conversational responses grounded in that information.

1. Use Trusted Knowledge Sources

Always ground responses in enterprise-approved content.

Examples include:

  • SharePoint Online document libraries
  • Microsoft OneDrive
  • Microsoft Dataverse
  • Azure AI Search indexes
  • Company websites
  • Internal knowledge bases
  • FAQs
  • Product documentation
  • Policy manuals
  • Technical documentation

Benefits include:

  • More accurate responses
  • Reduced hallucinations
  • Easier governance
  • Better compliance

2. Keep Knowledge Current

The AI can only answer accurately if its knowledge is accurate.

Organizations should:

  • Remove obsolete documents
  • Archive outdated policies
  • Update procedures
  • Refresh FAQs
  • Review documentation regularly

Poor knowledge produces poor answers.

3. Write Good Source Content

Generative AI performs better when source documents are:

  • Clearly written
  • Well organized
  • Consistent
  • Free of contradictory information
  • Properly titled
  • Divided into logical sections

Instead of one 400-page manual, multiple focused documents often produce better retrieval results.

4. Limit Knowledge Scope

Avoid connecting every possible document source.

Instead:

  • Connect only relevant repositories.
  • Use Azure AI Search indexes.
  • Separate HR knowledge from IT knowledge.
  • Separate Finance knowledge from Customer Support knowledge.

Smaller knowledge domains generally improve retrieval accuracy.

5. Combine Topics with Generative Answers

Not every conversation should rely entirely on AI generation.

A common design pattern:

Customer asks question
        │
        ▼
Topic determines intent
        │
        ▼
If structured workflow needed
        │
        ▼
Run Topic
        │
        ▼
If informational question
        │
        ▼
Run Generative Answers
        │
        ▼
Return grounded response

This hybrid approach provides predictable business logic while leveraging AI for knowledge retrieval.

6. Provide Conversation Context

Generative Answers work best when they receive context.

Instead of asking:

“Vacation”

Ask:

“Explain the employee vacation policy for full-time employees.”

The additional context helps retrieve more relevant information.

7. Protect Sensitive Information

Knowledge sources should respect organizational security.

Examples:

  • HR documents
  • Payroll records
  • Legal contracts
  • Medical information
  • Financial reports

Ensure users only receive information they are authorized to access.

8. Test with Real User Questions

Instead of testing only ideal scenarios:

Try questions such as:

  • “How do I reset my laptop?”
  • “What’s our refund policy?”
  • “Can I carry unused vacation days?”
  • “How do I submit an expense report?”

Testing natural language improves overall solution quality.

Common Design Patterns

Pattern 1: IT Help Desk

User:
My laptop won't connect to Wi-Fi.
↓
Generative Answers searches:
• IT documentation
• Network troubleshooting guides
• FAQ articles
↓
Returns troubleshooting steps.

Pattern 2: HR Assistant

User:
How many sick days do I receive?
↓
Search HR policy documents
↓
Generate policy explanation.

Pattern 3: Customer Support

Customer:
Can I return an opened product?
↓
Search return policy
↓
Generate customer-friendly response.

Pattern 4: Product Assistant

Customer:
Does Model X support Wi-Fi 6?
↓
Search product specifications
↓
Generate answer from documentation.

Common Mistakes

Mistake 1

Connecting outdated documentation.

Result:

Incorrect answers.

Mistake 2

Connecting documents containing conflicting information.

Result:

Inconsistent responses.

Mistake 3

Expecting the AI to know company policies without connected knowledge.

Result:

Hallucinations.

Mistake 4

Using Generative Answers for transactional workflows.

Instead use:

  • Topics
  • Agent flows
  • Actions
  • Power Automate
  • Connectors

Mistake 5

Providing vague prompts.

Example:

Tell me about benefits.

Better:

Explain the health insurance benefits available to full-time employees.

Exam Tips

For the AB-620 exam, remember the following:

  • The Generative Answers node is designed for grounded, AI-generated responses based on connected knowledge.
  • It is not intended to replace structured business workflows.
  • Knowledge quality directly impacts response quality.
  • Azure AI Search enhances enterprise-scale retrieval.
  • Security permissions should govern access to enterprise knowledge.
  • Topics and Generative Answers are commonly used together.
  • Custom prompts can influence the tone, format, and style of responses.
  • Multiple knowledge sources can be combined within a single agent.
  • Testing with realistic user questions is essential before deployment.
  • Monitoring response quality helps identify gaps in documentation and knowledge sources.

Practice Exam Questions

Question 1

A company wants its AI agent to answer employee questions using official HR documentation while minimizing hallucinations.

Which feature should be configured?

A. Variables only

B. Generative Answers connected to HR knowledge sources

C. Conversation transcripts

D. Adaptive Dialogs

Answer: B

Explanation: Connecting the Generative Answers node to authoritative HR documentation grounds responses in trusted enterprise content and significantly reduces hallucinations.

Question 2

Which scenario is the BEST use case for the Generative Answers node?

A. Creating new Dataverse tables

B. Processing payroll transactions

C. Answering questions from company documentation

D. Deploying solutions between environments

Answer: C

Explanation: The Generative Answers node excels at retrieving information from connected knowledge sources and generating natural-language responses based on that information.

Question 3

An organization notices inconsistent answers because two policy documents contain conflicting information.

What should the administrator do FIRST?

A. Increase AI temperature.

B. Disable generative responses.

C. Add more connectors.

D. Remove or reconcile conflicting documentation.

Answer: D

Explanation: Conflicting source content leads to inconsistent retrieval and responses. The underlying documentation should be reviewed and updated before modifying AI settings.

Question 4

Why should organizations regularly update connected knowledge sources?

A. To improve Power Automate performance

B. To reduce licensing costs

C. To increase connector limits

D. To ensure AI responses reflect current information

Answer: D

Explanation: Generative Answers relies on the connected knowledge. Outdated documents can result in inaccurate or obsolete responses.

Question 5

A developer wants an agent to execute an approval process after answering a policy question.

Which design is MOST appropriate?

A. Use only the Generative Answers node.

B. Replace topics with variables.

C. Combine Topics or Agent Flows with Generative Answers.

D. Disable AI responses.

Answer: C

Explanation: Generative Answers handles informational responses, while Topics and Agent Flows manage structured business processes such as approvals.

Question 6

Which practice generally improves retrieval accuracy?

A. Connecting every available document repository

B. Allowing unrestricted internet searches

C. Increasing conversation length

D. Limiting knowledge sources to relevant content

Answer: D

Explanation: Restricting knowledge sources to relevant, high-quality content reduces noise and improves the relevance of retrieved information.

Question 7

Which characteristic makes enterprise documentation easier for Generative Answers to use?

A. Random organization

B. Duplicate information

C. Clear structure with logical sections

D. Multiple conflicting versions

Answer: C

Explanation: Well-structured, clearly organized documents improve indexing, retrieval, and answer generation.

Question 8

An HR chatbot should ensure employees only access information they are authorized to view.

Which consideration is MOST important?

A. Conversation length

B. Prompt creativity

C. Variable naming

D. Knowledge source security and permissions

Answer: D

Explanation: Access controls and security permissions should be enforced so that users only receive information they are authorized to access.

Question 9

A user asks, “How do I submit an expense report?”

What should be included in testing before production deployment?

A. Only technical validation

B. Only connector authentication

C. Realistic user questions that reflect actual usage

D. Only performance testing

Answer: C

Explanation: Testing with realistic, natural-language questions helps ensure the agent performs well under real-world conditions.

Question 10

Which statement BEST describes the role of the Generative Answers node?

A. It replaces all Topics and Agent Flows.

B. It performs database schema migrations.

C. It automatically builds Power Automate flows.

D. It generates grounded responses using connected knowledge sources.

Answer: D

Explanation: The Generative Answers node retrieves information from configured knowledge sources and uses AI to generate conversational, context-aware responses based on that content.

Go to the AB-620 Exam Prep Hub main page

AB-900, Agentic AI, AI Governance, Generative AI, Microsoft Certification

Manage prompts, in Microsoft Copilot, including saving, sharing, scheduling, and deleting (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Perform basic administrative tasks for Copilot
      --> Manage prompts, in Microsoft Copilot, including saving, sharing, scheduling, and deleting

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

Microsoft 365 Copilot allows users to create and reuse prompts to streamline repetitive work such as drafting emails, summarizing documents, generating reports, or analyzing data. From an administrative perspective, understanding how prompts are managed is important for governance, productivity, and consistency across an organization.

Prompts can be treated as reusable productivity assets that users can store, distribute, and manage over time—especially when Copilot is used at scale across Microsoft 365 apps.

1. What are Copilot prompts?

A Copilot prompt is a natural language instruction given to Copilot to generate output. For example:

  • “Summarize this meeting in five bullet points.”
  • “Draft a project update email for stakeholders.”
  • “Analyze this Excel dataset and highlight trends.”

Prompts can be:

  • One-time (ad hoc usage)
  • Saved for reuse
  • Shared across users or teams
  • Scheduled for recurring execution (in supported scenarios)

2. Saving prompts

Saving prompts allows users to reuse effective instructions without rewriting them.

Key characteristics:

  • Stored in a user-accessible prompt library or prompt experience
  • Can be reused across Microsoft 365 apps (Word, Teams, Outlook, etc.)
  • Helps standardize repetitive business tasks

Benefits:

  • Increases productivity
  • Encourages consistent output formatting
  • Reduces time spent recreating complex prompts

Example:

A finance analyst saves a prompt:

“Summarize quarterly revenue performance and highlight anomalies.”

3. Sharing prompts

Prompts can be shared with other users or teams to promote consistency.

Sharing capabilities include:

  • Sharing with individuals or groups
  • Embedding prompts into team workflows
  • Distributing best-practice prompts across departments

Use cases:

  • Standard HR onboarding email drafts
  • Sales proposal templates
  • IT troubleshooting responses

Governance consideration:

Shared prompts should align with organizational policies to avoid:

  • Exposure of sensitive instructions
  • Use of non-compliant content templates

4. Scheduling prompts

Scheduling allows prompts to be executed at defined intervals or triggered conditions (depending on Copilot capabilities and integration context).

Examples of scheduled prompt usage:

  • Daily summary of emails in Outlook
  • Weekly project status report generation
  • Regular data analysis summaries in Excel

Benefits:

  • Automates repetitive reporting tasks
  • Ensures timely information delivery
  • Reduces manual effort

Important note:

Scheduling capabilities may depend on:

  • Copilot-enabled workflows
  • Microsoft 365 integrations (Power Automate or agent-based automation)

5. Deleting prompts

Prompts can be deleted when they are no longer needed or are outdated.

Reasons for deletion:

  • Prompt is obsolete or inaccurate
  • Organizational standards have changed
  • Security or compliance concerns
  • User no longer needs the prompt

Administrative considerations:

  • Deleted prompts may not be recoverable depending on retention policies
  • Enterprises may enforce governance policies around prompt lifecycle management

6. Administrative and governance considerations

When managing prompts at scale, administrators should consider:

Security

  • Prevent sharing of sensitive prompts containing confidential logic
  • Ensure prompts do not encourage data leakage

Compliance

  • Align prompt usage with Microsoft Purview policies
  • Ensure prompts do not bypass organizational controls

Lifecycle management

  • Define rules for retention, reuse, and deletion
  • Standardize prompt libraries for departments

User enablement

  • Provide curated prompt libraries
  • Encourage adoption of approved prompt templates

7. Key exam takeaway

For AB-900, focus on the fact that Copilot prompt management includes:

  • Saving prompts for reuse
  • Sharing prompts across users or teams
  • Scheduling prompts for recurring tasks (where supported)
  • Deleting prompts for governance and lifecycle control

These capabilities support productivity while requiring governance oversight in enterprise environments.

Practice Exam Questions (10)

1.

What is the primary benefit of saving Copilot prompts?

A. It increases network bandwidth usage
B. It allows reuse of effective instructions
C. It disables prompt security controls
D. It deletes old conversations automatically

Answer: B
Explanation: Saving prompts enables reuse of effective instructions, improving productivity and consistency.

2.

An organization wants to standardize email drafts across departments. Which feature supports this goal?

A. Prompt deletion
B. Prompt sharing
C. Device enrollment
D. Data loss prevention

Answer: B
Explanation: Sharing prompts allows standardized templates and instructions to be distributed across teams.

3.

Which scenario best represents a scheduled Copilot prompt?

A. A one-time email draft request
B. A manually typed search query
C. A daily summary report generated automatically
D. A deleted conversation thread

Answer: C
Explanation: Scheduled prompts run at defined intervals, such as daily report generation.

4.

Why might an administrator enforce governance rules on shared prompts?

A. To increase storage capacity
B. To reduce CPU usage
C. To prevent exposure of sensitive or non-compliant content
D. To disable Copilot licensing

Answer: C
Explanation: Shared prompts may contain sensitive logic, so governance ensures compliance and security.

5.

What typically happens when a prompt is deleted?

A. It is permanently removed from the prompt library
B. It becomes read-only
C. It is converted into a system alert
D. It is automatically shared with all users

Answer: A
Explanation: Deleting a prompt removes it from the library, although retention policies may affect recoverability.

6.

Which of the following is a valid use case for saved prompts?

A. Running antivirus scans
B. Reusing a formatted project status report request
C. Managing device drivers
D. Configuring network routing

Answer: B
Explanation: Saved prompts are used for repeatable tasks like structured reports or summaries.

7.

What is a key risk of unmanaged prompt sharing?

A. Increased CPU performance
B. Exposure of sensitive instructions or business logic
C. Faster email delivery
D. Reduced storage costs

Answer: B
Explanation: Unmanaged sharing can expose sensitive organizational logic or data-handling instructions.

8.

Which Microsoft 365 principle is most relevant to managing Copilot prompts?

A. Hardware lifecycle management
B. Identity federation
C. Information governance
D. Network segmentation

Answer: C
Explanation: Prompt management relates to information governance, including control over content and usage.

9.

What is a benefit of scheduling prompts in Copilot-enabled workflows?

A. It eliminates user authentication
B. It automates repetitive reporting tasks
C. It disables Microsoft 365 apps
D. It increases manual effort

Answer: B
Explanation: Scheduled prompts automate recurring tasks like reports and summaries.

10.

Which action supports prompt lifecycle management in an enterprise environment?

A. Random prompt duplication
B. Unrestricted external sharing
C. Deleting outdated prompts based on policy
D. Disabling all Copilot features

Answer: C
Explanation: Removing outdated prompts helps maintain compliance and ensures only relevant prompts are retained.

Go to the AB-900 Exam Prep Hub main page

AB-900, Agentic AI, AI, Generative AI, Microsoft Certification

Identify use cases for custom agents (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Understand features and capabilities of Copilot and agents
      --> Identify use cases for custom agents

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

Custom agents in Microsoft 365 Copilot extend Copilot’s built-in capabilities by allowing organizations to create tailored AI assistants focused on specific business processes, data sources, and workflows. Unlike general Copilot experiences, custom agents are designed to operate within defined boundaries, grounded in organizational knowledge and governed data.

What are custom agents?

A custom agent is a specialized AI assistant built on top of Microsoft 365 Copilot that can:

  • Use organization-specific knowledge sources (SharePoint sites, files, Dataverse, web connectors, etc.)
  • Follow predefined instructions and behaviors
  • Perform scoped tasks such as answering domain questions, generating structured outputs, or assisting workflows
  • Operate with Microsoft 365 identity and security controls

They are typically built using tools like Copilot Studio and integrated into Microsoft 365 experiences such as Teams, SharePoint, or Copilot chat.

Key characteristics of custom agents

Custom agents differ from general Copilot usage in several important ways:

They are purpose-built, meaning they are designed for a specific function such as HR support or IT helpdesk assistance. They are also data-grounded, relying on selected enterprise knowledge sources rather than broad internet knowledge.

They are governed, meaning they respect Microsoft 365 permissions, Microsoft Purview policies, and organizational compliance boundaries.

Finally, they are interactive and task-oriented, often guiding users through structured processes rather than only responding to ad-hoc questions.

Common use cases for custom agents

1. HR and employee support agents

Custom HR agents are commonly used to:

  • Answer questions about leave policies, benefits, and onboarding
  • Guide employees through HR workflows
  • Retrieve policy documents from SharePoint or HR systems

This reduces HR ticket volume and improves employee self-service.

2. IT helpdesk and support agents

IT-focused agents can:

  • Troubleshoot common issues (password resets, device setup, VPN access)
  • Provide step-by-step remediation guidance
  • Surface knowledge base articles from internal documentation

These agents help reduce repetitive IT support requests.

3. Sales and customer support agents

Sales agents are used to:

  • Summarize customer accounts and opportunities
  • Retrieve CRM data and product information
  • Generate sales emails or proposals

Customer support agents can also respond to common inquiries using approved knowledge bases.

4. Knowledge management agents

Organizations use agents to:

  • Provide structured access to company policies and documentation
  • Answer questions across multiple SharePoint sites
  • Improve search and discovery of internal content

These agents are especially valuable in large enterprises with distributed knowledge.

5. Finance and operations agents

Custom agents in finance or operations can:

  • Assist with budget tracking queries
  • Explain financial reporting definitions
  • Summarize operational KPIs or dashboards

They typically connect to controlled datasets and reporting systems.

6. Project and workflow assistants

These agents help teams by:

  • Tracking project status updates
  • Summarizing meeting notes
  • Guiding users through standardized workflows (e.g., project intake, approvals)

When to use custom agents vs standard Copilot

Custom agents are most appropriate when:

  • A repeatable business process exists
  • The organization has curated knowledge sources
  • Responses must follow strict formatting or rules
  • Domain-specific accuracy is required (HR, finance, IT, legal)

Standard Copilot is better for:

  • General productivity tasks (writing, summarizing, brainstorming)
  • Ad hoc questions that do not require structured workflows or specialized data

Governance considerations

Custom agents inherit Microsoft 365 security and compliance controls, including:

  • Microsoft Entra ID authentication
  • Microsoft Purview sensitivity labels and DLP policies
  • Role-based access control (RBAC)
  • Data access restricted by user permissions

This ensures agents do not expose information beyond what a user is authorized to see.

Summary

Custom agents in Microsoft 365 Copilot are specialized AI assistants designed for targeted business scenarios. They extend Copilot by adding organizational knowledge, structured workflows, and governance controls. Their primary value lies in automating repetitive tasks, improving knowledge access, and supporting domain-specific processes across departments such as HR, IT, finance, and operations.

Practice Exam Questions (10)

1.

A company wants an assistant that can answer employee questions about vacation policies using only internal HR documents stored in SharePoint. What is the best solution?

A. Use a custom Copilot agent grounded in HR SharePoint content
B. Use Microsoft Excel Copilot only
C. Use a Power BI dashboard
D. Use a generic web Copilot chat

Answer: A
Explanation: A custom agent can be grounded in specific SharePoint HR content and provide controlled, policy-based responses.

2.

Which scenario best represents a use case for a custom agent?

A. Writing a marketing email from scratch
B. Generating creative ideas for a product name
C. Answering general trivia questions
D. Guiding users through an IT password reset workflow

Answer: D
Explanation: IT helpdesk workflows are structured, repeatable, and ideal for custom agents.

3.

What is a key benefit of using custom agents in Microsoft 365 Copilot?

A. They bypass Microsoft security controls for faster responses
B. They only use public internet data
C. They enforce organizational policies and use approved data sources
D. They eliminate the need for user authentication

Answer: C
Explanation: Custom agents respect Microsoft 365 governance and use controlled enterprise data.

4.

A finance team wants an AI tool that summarizes monthly budget reports stored in controlled datasets. Which capability is most appropriate?

A. Custom finance agent grounded in approved financial data
B. Personal Microsoft Word Copilot
C. Bing search integration
D. Email auto-responder rules

Answer: A
Explanation: Finance use cases require structured, governed access to internal datasets.

5.

Which tool is commonly used to create custom agents for Microsoft 365 Copilot?

A. Power Automate only
B. Copilot Studio
C. Azure DevOps
D. Microsoft Access

Answer: B
Explanation: Copilot Studio is used to build and configure custom agents.

6.

What distinguishes a custom agent from standard Microsoft 365 Copilot?

A. It can only work offline
B. It uses only unstructured internet data
C. It is built for specific business scenarios and uses curated data sources
D. It replaces all Microsoft 365 applications

Answer: C
Explanation: Custom agents are scoped to specific business needs and data sources.

7.

Which is a valid HR-related use case for a custom agent?

A. Generating random social media posts
B. Answering employee benefit questions from policy documents
C. Editing video content
D. Running system diagnostics on servers

Answer: B
Explanation: HR agents provide policy-based answers from controlled documentation.

8.

What ensures a custom agent does NOT expose unauthorized data?

A. Internet firewall rules
B. Microsoft Defender antivirus only
C. User identity and Microsoft 365 permissions
D. Manual approval of every prompt

Answer: C
Explanation: Access is controlled through Microsoft Entra ID and existing permissions.

9.

When should a custom agent be preferred over standard Copilot?

A. When tasks are ad hoc and creative
B. When structured workflows and specific business rules are required
C. When browsing public websites
D. When no data sources are needed

Answer: B
Explanation: Custom agents are ideal for structured, repeatable workflows.

10.

Which department would most likely benefit from a knowledge management agent?

A. HR requesting policy document access
B. Users playing games
C. Graphic design teams creating artwork
D. Hardware repair technicians fixing printers

Answer: A
Explanation: Knowledge management agents help retrieve and summarize internal policies and documentation.

Go to the AB-900 Exam Prep Hub main page

AB-900, Agentic AI, AI, Generative AI, Microsoft Certification

Identify use cases for Analyst (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Understand features and capabilities of Copilot and agents
      --> Identify use cases for Analyst

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

Analyst is an advanced reasoning capability in the Microsoft 365 Copilot ecosystem that is designed to help users analyze structured data, identify trends, perform complex calculations, generate forecasts, and produce data-driven insights. Rather than simply summarizing information, Analyst works through data methodically, much like an experienced business analyst or data analyst.

While standard Microsoft 365 Copilot can answer questions, summarize documents, or create presentations, Analyst specializes in data analysis and quantitative reasoning. It can evaluate datasets, perform multi-step calculations, identify patterns, compare scenarios, and generate visualizations and recommendations based on the available data.

For the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals exam, you should understand:

  • What Analyst is
  • How Analyst differs from standard Microsoft 365 Copilot
  • Typical business scenarios where Analyst is valuable
  • The types of data Analyst can analyze
  • How Microsoft 365 permissions and governance apply
  • Best practices and limitations when using Analyst

What Is Analyst?

Analyst is an AI capability that focuses on turning data into actionable insights.

Typical tasks include:

  • Analyzing spreadsheets
  • Identifying trends
  • Comparing datasets
  • Creating summaries
  • Performing calculations
  • Producing forecasts
  • Detecting anomalies
  • Supporting business decisions

Instead of answering a single question, Analyst performs a structured analytical process to produce meaningful conclusions.

Example request:

“Analyze our quarterly sales data, identify the fastest-growing regions, explain why revenue declined in Q2, and forecast next quarter’s revenue.”

Rather than providing a brief response, Analyst may:

  • Examine multiple worksheets
  • Perform statistical calculations
  • Identify trends
  • Compare time periods
  • Detect unusual changes
  • Create charts
  • Generate a written business summary

How Analyst Differs from Standard Microsoft 365 Copilot

Standard Microsoft 365 Copilot focuses on everyday productivity tasks such as:

  • Drafting emails
  • Summarizing meetings
  • Creating presentations
  • Writing documents
  • Answering general questions

Analyst focuses on:

  • Data analysis
  • Statistical reasoning
  • Forecasting
  • Trend analysis
  • Scenario comparisons
  • Financial analysis
  • Business intelligence
  • Quantitative decision support

Think of standard Copilot as an AI assistant for productivity, while Analyst functions more like a business intelligence analyst or data analyst.

Data Sources Used by Analyst

Analyst can work with information stored across Microsoft 365, including:

  • Microsoft Excel workbooks
  • SharePoint document libraries
  • OneDrive files
  • Microsoft Teams shared files
  • CSV files
  • Tables
  • Financial reports
  • Business reports
  • Operational data
  • Microsoft Graph organizational context

Depending on the scenario, Analyst may combine structured business data with organizational context to provide richer insights.

Microsoft Graph and Analyst

Like other Microsoft 365 Copilot capabilities, Analyst uses Microsoft Graph to locate organizational data.

Microsoft Graph provides:

  • File locations
  • User permissions
  • Organizational relationships
  • Shared documents
  • Teams collaboration
  • SharePoint content
  • OneDrive files

Analyst only analyzes information the requesting user is authorized to access.

Common Business Use Cases

1. Sales Analysis

Sales managers can ask Analyst to:

  • Compare regional sales
  • Identify top-performing products
  • Analyze customer trends
  • Calculate revenue growth
  • Highlight underperforming markets

Example:

“Compare this year’s sales with last year and explain the largest changes.”

2. Financial Analysis

Finance teams can use Analyst to:

  • Analyze budgets
  • Compare expenses
  • Forecast revenue
  • Calculate profit margins
  • Identify cost drivers
  • Review spending patterns

3. Forecasting

Analyst can identify trends and generate forecasts such as:

  • Future sales
  • Inventory demand
  • Budget projections
  • Staffing needs
  • Customer growth

Forecasts help organizations plan future operations.

4. Trend Identification

Businesses often need to understand how metrics change over time.

Analyst can identify:

  • Seasonal trends
  • Growth patterns
  • Declining performance
  • Customer behavior
  • Revenue changes
  • Operational improvements

5. Executive Dashboards

Executives often receive large datasets.

Analyst can summarize:

  • Key performance indicators (KPIs)
  • Revenue
  • Profitability
  • Customer growth
  • Operational metrics
  • Performance against targets

Instead of reviewing hundreds of rows of data, leaders receive concise business insights.

6. Operational Analysis

Operations teams can analyze:

  • Manufacturing output
  • Service performance
  • Delivery times
  • Inventory levels
  • Workforce productivity
  • Supply chain efficiency

7. Human Resources Analytics

HR departments can analyze:

  • Employee turnover
  • Hiring trends
  • Training completion
  • Workforce demographics
  • Performance metrics
  • Recruiting pipelines

8. Marketing Performance

Marketing teams can evaluate:

  • Campaign effectiveness
  • Customer engagement
  • Website traffic
  • Conversion rates
  • Lead generation
  • Advertising performance

9. Risk Analysis

Organizations can identify:

  • Financial risks
  • Operational risks
  • Performance anomalies
  • Unexpected changes
  • Outliers in business data

Analyst helps prioritize issues requiring investigation.

10. Scenario Analysis

Analyst can compare multiple business scenarios.

Example:

“What happens if sales increase by 10% while operating expenses increase by 5%?”

Analyst performs the calculations and explains the business impact.

Types of Questions Analyst Can Answer

Examples include:

  • Which products are growing fastest?
  • Which departments exceeded budget?
  • What caused this month’s revenue decline?
  • Which customers generate the highest revenue?
  • Which regions are underperforming?
  • What trends appear over the past three years?
  • Which business unit has the highest operating costs?
  • Which employees completed required training?

Charts and Visualizations

Analyst may generate or recommend visualizations such as:

  • Bar charts
  • Line charts
  • Pie charts
  • Trend graphs
  • Comparison tables
  • KPI summaries

These help users understand complex datasets more quickly.

Security and Governance

Analyst follows the same Microsoft 365 security model as Microsoft 365 Copilot.

It respects:

  • Microsoft Graph permissions
  • SharePoint permissions
  • OneDrive permissions
  • Teams permissions
  • Microsoft Purview sensitivity labels
  • Data Loss Prevention (DLP) policies
  • Retention policies
  • Microsoft Defender protections

Analyst cannot analyze data that a user is not authorized to access.

Benefits of Analyst

Organizations benefit because Analyst can:

  • Reduce manual spreadsheet analysis
  • Improve decision-making
  • Accelerate reporting
  • Identify hidden trends
  • Detect anomalies quickly
  • Improve forecasting accuracy
  • Increase productivity
  • Support data-driven business decisions
  • Reduce repetitive analytical work

Limitations of Analyst

Although powerful, Analyst has limitations.

It:

  • Depends on the quality of the underlying data.
  • Cannot correct inaccurate source data automatically.
  • Cannot access restricted information.
  • Should not replace expert financial, legal, or scientific judgment.
  • May require users to validate calculations.
  • Cannot override Microsoft 365 permissions.
  • Cannot bypass Microsoft Purview protections.

Human review remains essential before making important business decisions.

Best Practices

Microsoft recommends that organizations:

  • Maintain clean, accurate datasets.
  • Store data in well-organized Microsoft 365 locations.
  • Apply Microsoft Purview sensitivity labels appropriately.
  • Use meaningful table names and column headers.
  • Ask clear, specific analytical questions.
  • Validate AI-generated calculations.
  • Review forecasts before implementing business decisions.
  • Monitor data quality regularly.
  • Ensure users have appropriate permissions.
  • Combine Analyst insights with human expertise.

Researcher vs. Analyst

Although both capabilities use advanced reasoning, they serve different purposes.

FeatureResearcherAnalyst
Primary purposeResearch and knowledge synthesisData analysis and quantitative reasoning
Main data typesDocuments, emails, meetings, knowledgeStructured datasets, spreadsheets, reports
Typical outputResearch reportsAnalytical reports and business insights
FocusInformation gathering and synthesisCalculations, trends, forecasting, analysis
Common usersExecutives, project managers, researchersFinance, sales, operations, business analysts

Exam Tips

For the AB-900 exam, remember these key points:

  • Analyst specializes in structured data analysis, calculations, forecasting, and trend identification.
  • It differs from Researcher, which focuses on knowledge discovery and multi-source research.
  • Analyst commonly works with Excel workbooks, reports, and business data.
  • It uses Microsoft Graph to locate organizational information.
  • It respects Microsoft 365 permissions and Microsoft Purview governance policies.
  • Analyst supports decision-making but does not replace human expertise.
  • Organizations should validate AI-generated analyses before acting on them.
  • Security, compliance, and permissions remain enforced throughout the analysis process.

10 Practice Exam Questions

Question 1

Which type of task is Analyst primarily designed to perform?

A. Configuring Microsoft Entra ID

B. Performing structured data analysis and generating business insights

C. Managing Microsoft Teams meetings

D. Assigning Microsoft 365 licenses

Correct Answer: B

Explanation: Analyst specializes in analyzing structured data, identifying trends, performing calculations, and producing data-driven insights.

Question 2

A finance manager wants to forecast next quarter’s revenue using historical sales data. Which Microsoft 365 Copilot capability is most appropriate?

A. Analyst

B. Microsoft Defender

C. SharePoint Advanced Management

D. Communication Compliance

Correct Answer: A

Explanation: Analyst is designed for forecasting, trend analysis, and financial modeling based on structured datasets.

Question 3

Which Microsoft technology helps Analyst locate organizational files and data while respecting permissions?

A. Microsoft Intune

B. Microsoft Defender

C. Microsoft Graph

D. Microsoft Sentinel

Correct Answer: C

Explanation: Microsoft Graph provides access to organizational relationships, files, and collaboration data while enforcing existing permissions.

Question 4

Which business scenario is the best use case for Analyst?

A. Writing a company newsletter

B. Comparing quarterly sales performance across regions

C. Creating a SharePoint site

D. Managing Exchange Online mailboxes

Correct Answer: B

Explanation: Analyst excels at comparing structured business data and identifying meaningful trends.

Question 5

What happens if a user does not have permission to access a financial spreadsheet?

A. Analyst requests administrator approval automatically.

B. Analyst bypasses permissions during analysis.

C. Analyst uses the spreadsheet but hides sensitive values.

D. Analyst cannot analyze the spreadsheet for that user.

Correct Answer: D

Explanation: Analyst respects Microsoft 365 permissions and cannot access data that the requesting user is not authorized to view.

Question 6

Which Microsoft Purview capability continues protecting sensitive data analyzed by Analyst?

A. Windows Hello

B. Device encryption

C. Sensitivity labels and Data Loss Prevention (DLP) policies

D. Microsoft Edge profiles

Correct Answer: C

Explanation: Microsoft Purview protections remain in effect whenever Analyst accesses organizational content.

Question 7

Which output would Analyst most likely generate?

A. A statistical summary showing sales trends and revenue forecasts

B. A new Microsoft 365 tenant

C. A user password reset

D. A SharePoint site template

Correct Answer: A

Explanation: Analyst produces analytical reports, calculations, forecasts, and business insights based on structured data.

Question 8

What is one limitation of Analyst?

A. It automatically corrects inaccurate source data.

B. It replaces professional financial judgment.

C. It ignores incomplete datasets.

D. It depends on the quality and completeness of the available data.

Correct Answer: D

Explanation: Analyst’s output is only as reliable as the underlying data and should be reviewed before making important decisions.

Question 9

Which statement best distinguishes Analyst from Researcher?

A. Analyst focuses on structured data analysis, while Researcher focuses on gathering and synthesizing information from multiple knowledge sources.

B. Analyst manages Microsoft 365 licenses, while Researcher manages permissions.

C. Analyst is used only in Microsoft Excel, while Researcher is used only in Microsoft Word.

D. Analyst replaces Microsoft Graph, while Researcher replaces Microsoft Purview.

Correct Answer: A

Explanation: Analyst specializes in quantitative analysis, whereas Researcher specializes in deep research and knowledge synthesis.

Question 10

Why should organizations validate Analyst-generated insights before making business decisions?

A. Analyst always produces incorrect calculations.

B. Analyst can modify the organization’s data automatically.

C. Human review is still necessary because AI-generated analyses should be verified before acting on them.

D. Analyst cannot generate forecasts.

Correct Answer: C

Explanation: Analyst is a powerful decision-support tool, but organizations should validate AI-generated analyses and recommendations using human expertise before making significant business decisions.

Go to the AB-900 Exam Prep Hub main page

AB-900, Agentic AI, AI, Generative AI, Microsoft Certification

Identify use cases for Researcher (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Understand features and capabilities of Copilot and agents
      --> Identify use cases for Researcher

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

Researcher is an advanced AI-powered reasoning capability available within the Microsoft 365 Copilot ecosystem. It is designed to perform multi-step, in-depth research tasks that require gathering information from multiple sources, analyzing large amounts of data, synthesizing findings, and presenting comprehensive, well-organized results.

Unlike standard Copilot experiences, which typically generate responses from a single prompt, Researcher performs more sophisticated reasoning by combining enterprise knowledge stored in Microsoft 365 with, when appropriate and permitted, external information sources. It is intended to help users complete tasks that would normally require hours of manual research.

For the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals exam, you should understand:

  • What Researcher is
  • How it differs from standard Microsoft 365 Copilot experiences
  • Typical business scenarios where Researcher provides value
  • The types of data Researcher uses
  • How Microsoft 365 security, permissions, and governance continue to apply
  • The limitations and best practices for using Researcher

What Is Researcher?

Researcher is an advanced AI capability that helps users perform complex research tasks by:

  • Collecting information from multiple sources
  • Comparing information
  • Identifying patterns
  • Summarizing findings
  • Producing structured reports
  • Citing supporting information where applicable
  • Performing iterative reasoning before generating a final response

Rather than simply answering a question, Researcher can develop a complete research workflow.

Example request:

“Prepare a report comparing our organization’s cloud migration strategy with current industry best practices and identify potential risks.”

Instead of providing a brief summary, Researcher may:

  • Review internal project documentation
  • Examine meeting notes
  • Analyze SharePoint documents
  • Review emails
  • Compare current practices with publicly available information (when configured)
  • Produce a detailed report with recommendations

How Researcher Differs from Standard Copilot

Standard Microsoft 365 Copilot focuses primarily on helping users complete everyday productivity tasks such as:

  • Drafting emails
  • Summarizing meetings
  • Creating presentations
  • Rewriting documents
  • Generating tables
  • Answering questions

Researcher extends these capabilities by emphasizing:

  • Multi-step reasoning
  • Long-form research
  • Deep analysis
  • Information synthesis
  • Strategic recommendations
  • Comprehensive reporting

Think of standard Copilot as an AI assistant, while Researcher functions more like an AI research analyst.

Data Sources Used by Researcher

Researcher can analyze information from multiple Microsoft 365 sources, including:

  • SharePoint sites
  • OneDrive files
  • Microsoft Teams conversations
  • Outlook emails
  • Microsoft Word documents
  • Excel workbooks
  • PowerPoint presentations
  • OneNote notebooks
  • Microsoft Graph organizational relationships

Depending on organizational configuration and licensing, Researcher may also incorporate approved external information sources.

Microsoft Graph and Researcher

Researcher relies heavily on Microsoft Graph.

Microsoft Graph provides:

  • Organizational relationships
  • User permissions
  • File locations
  • Emails
  • Meetings
  • Calendar events
  • Conversations
  • Shared documents
  • Collaboration history

Researcher uses Microsoft Graph to locate relevant information efficiently.

Importantly, Researcher never bypasses Microsoft Graph permissions.

If a user cannot access a document, Researcher cannot use it.

Common Business Use Cases

1. Market Research

Researcher can help organizations:

  • Compare competitors
  • Analyze market trends
  • Summarize industry reports
  • Identify emerging technologies
  • Evaluate customer behavior

Example:

“Research the latest AI adoption trends in financial services.”

2. Executive Briefings

Executives often require concise summaries from large volumes of information.

Researcher can:

  • Summarize multiple meetings
  • Combine reports
  • Review emails
  • Produce executive-ready briefing documents

3. Project Research

Large projects often generate hundreds of documents.

Researcher can help summarize:

  • Requirements
  • Risks
  • Decisions
  • Milestones
  • Meeting notes
  • Design documents

Instead of reading dozens of files manually, Researcher consolidates the information.

4. Policy Analysis

Organizations frequently maintain hundreds of internal policies.

Researcher can:

  • Compare policies
  • Identify inconsistencies
  • Summarize requirements
  • Highlight missing documentation

5. Compliance Research

Researcher can assist with:

  • Reviewing compliance documentation
  • Summarizing regulatory guidance
  • Comparing policies against standards
  • Organizing compliance evidence

It does not replace formal compliance or legal reviews.

6. Sales Preparation

Sales teams can use Researcher to prepare for customer meetings by combining:

  • Previous emails
  • Meeting notes
  • Proposal documents
  • Customer presentations
  • Product documentation

The result is a comprehensive customer briefing.

7. Product Research

Product managers may ask Researcher to:

  • Compare product requirements
  • Analyze customer feedback
  • Summarize bug reports
  • Review feature requests
  • Recommend priorities

8. Knowledge Discovery

Employees often spend significant time searching for information.

Researcher can locate and combine information from:

  • Multiple SharePoint sites
  • Teams chats
  • Emails
  • Documents
  • Internal knowledge bases

This significantly reduces research time.

9. Strategic Planning

Leadership teams may ask Researcher to:

  • Compare business strategies
  • Analyze organizational performance
  • Review previous planning documents
  • Summarize lessons learned
  • Generate strategic recommendations

10. Report Generation

Researcher can generate:

  • Research reports
  • Project summaries
  • Risk analyses
  • Business cases
  • Recommendation documents
  • Decision-support reports

How Researcher Protects Organizational Data

Researcher follows the same Microsoft 365 security model as Microsoft 365 Copilot.

It respects:

  • Microsoft Graph permissions
  • SharePoint permissions
  • OneDrive permissions
  • Teams permissions
  • Microsoft Purview sensitivity labels
  • Data Loss Prevention (DLP) policies
  • Retention policies
  • Microsoft Defender protections

Researcher cannot retrieve information users are not authorized to access.

Benefits of Researcher

Organizations benefit because Researcher can:

  • Reduce manual research time
  • Improve decision-making
  • Consolidate information from multiple sources
  • Produce consistent reports
  • Improve knowledge discovery
  • Increase employee productivity
  • Reduce duplicate work
  • Accelerate project planning

Limitations of Researcher

Although powerful, Researcher has limitations.

It:

  • Only accesses authorized data.
  • Depends on data quality.
  • Cannot invent missing information.
  • May produce incomplete answers if source data is incomplete.
  • Does not replace human judgment.
  • Does not override organizational permissions.
  • Cannot bypass compliance policies.
  • Should not be considered a legal or regulatory authority.

Users should always review AI-generated conclusions before making important business decisions.

Best Practices

Microsoft recommends that organizations:

  • Ensure SharePoint permissions are accurate before deployment.
  • Apply Microsoft Purview sensitivity labels consistently.
  • Implement DLP policies.
  • Organize content with meaningful names and metadata.
  • Maintain high-quality documentation.
  • Encourage users to write specific research prompts.
  • Review AI-generated reports before distribution.
  • Train employees on responsible AI usage.
  • Monitor adoption and usage.
  • Continuously improve information governance.

Exam Tips

For the AB-900 exam, remember these key points:

  • Researcher is designed for complex, multi-step research tasks, not simple productivity tasks.
  • It uses Microsoft Graph to locate organizational information.
  • It respects all existing Microsoft 365 permissions.
  • Microsoft Purview policies continue to protect data.
  • Researcher can combine information from multiple Microsoft 365 services.
  • It supports report creation, analysis, and decision-making.
  • Human review remains important for critical decisions.
  • Researcher improves productivity but does not replace subject matter expertise.

10 Practice Exam Questions

Question 1

Which type of task is Researcher primarily designed to perform?

A. Multi-step research and analysis across multiple data sources

B. Installing Microsoft 365 applications

C. Managing user licenses

D. Configuring SharePoint permissions

Correct Answer: A

Explanation: Researcher is intended for advanced research, reasoning, and analysis that combines information from multiple sources into comprehensive results.

Question 2

How does Researcher locate relevant organizational content?

A. By ignoring file permissions

B. By using Microsoft Graph to identify accessible organizational data

C. By copying data into a separate database

D. By downloading every SharePoint site locally

Correct Answer: B

Explanation: Researcher relies on Microsoft Graph to discover relationships, files, emails, meetings, and other Microsoft 365 content while respecting user permissions.

Question 3

Which scenario is the best use case for Researcher?

A. Changing a user’s password

B. Assigning Microsoft 365 licenses

C. Comparing multiple project documents and generating a strategic summary

D. Creating a new SharePoint site

Correct Answer: C

Explanation: Researcher excels at analyzing multiple documents and producing synthesized reports or recommendations.

Question 4

Which Microsoft 365 security principle applies to Researcher?

A. Researcher automatically grants access to restricted documents.

B. Researcher temporarily elevates user permissions.

C. Researcher ignores sensitivity labels during analysis.

D. Researcher only accesses content the user is already authorized to view.

Correct Answer: D

Explanation: Researcher follows the same permission model as Microsoft 365 Copilot and cannot access unauthorized content.

Question 5

Which Microsoft technology provides the organizational relationships that Researcher uses?

A. Microsoft Defender

B. Microsoft Entra ID

C. Microsoft Graph

D. Microsoft Intune

Correct Answer: C

Explanation: Microsoft Graph connects users, files, meetings, emails, calendars, and collaboration data that Researcher uses during analysis.

Question 6

Which business activity is a common use case for Researcher?

A. Replacing Microsoft Purview

B. Producing executive briefing reports by combining information from multiple Microsoft 365 sources

C. Managing Azure subscriptions

D. Configuring firewall rules

Correct Answer: B

Explanation: Researcher can consolidate organizational information into executive-ready reports and summaries.

Question 7

What limits the information that Researcher can include in its responses?

A. Internet bandwidth only

B. Microsoft licensing costs only

C. The amount of SharePoint storage available

D. The user’s existing Microsoft 365 permissions and governance policies

Correct Answer: D

Explanation: Researcher can only use data that the requesting user is authorized to access, and it remains subject to governance controls.

Question 8

Which Microsoft Purview capability continues protecting organizational information when Researcher accesses documents?

A. Sensitivity labels and Data Loss Prevention (DLP) policies

B. Printer management

C. Windows Update

D. Device drivers

Correct Answer: A

Explanation: Microsoft Purview policies, including sensitivity labels and DLP, remain fully enforced when Researcher accesses organizational content.

Question 9

Why should users review Researcher-generated reports before acting on them?

A. Researcher cannot create reports.

B. AI-generated findings should be validated because human judgment is still required for important decisions.

C. Researcher always produces incorrect results.

D. Researcher automatically changes organizational data.

Correct Answer: B

Explanation: While Researcher can significantly accelerate analysis, users remain responsible for verifying conclusions and making informed decisions.

Question 10

Which statement best describes Researcher compared to standard Microsoft 365 Copilot?

A. Researcher replaces Microsoft Graph.

B. Researcher is only available in Microsoft Teams.

C. Researcher focuses on deep research, multi-step reasoning, and comprehensive analysis rather than routine productivity tasks.

D. Researcher only summarizes email messages.

Correct Answer: C

Explanation: Standard Copilot primarily assists with day-to-day productivity, whereas Researcher specializes in complex analysis, reasoning, and report generation.

Go to the AB-900 Exam Prep Hub main page

AB-900, Agentic AI, AI, Generative AI, Microsoft Certification

Identify which Copilot features can be enabled or disabled (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Understand features and capabilities of Copilot and agents
      --> Identify which Copilot features can be enabled or disabled

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

One of the primary responsibilities of a Microsoft 365 Copilot administrator is understanding which Copilot features can be controlled through administrative settings. Organizations often have different security, compliance, and business requirements, so Microsoft provides administrators with the ability to enable or disable various Copilot capabilities at the tenant, service, and user levels.

For the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals exam, you should understand:

  • Which Copilot capabilities administrators can control
  • Where these controls are configured
  • Why organizations may enable or disable specific features
  • Which capabilities are always governed by Microsoft 365 permissions rather than simple on/off settings
  • How licensing affects feature availability

Why Organizations Control Copilot Features

Organizations don’t always want every AI capability immediately available to every employee.

Common reasons include:

  • Meeting regulatory requirements
  • Protecting sensitive information
  • Conducting pilot deployments
  • Managing licensing costs
  • Limiting access to experimental features
  • Preventing users from accessing external AI services
  • Reducing organizational risk

Microsoft allows administrators to gradually introduce Copilot while maintaining governance.

Administrative Control Layers

Copilot features can be managed through several layers.

Control LayerPurpose
LicensingDetermines who is entitled to use Copilot
Microsoft 365 Admin CenterEnables or disables Copilot services and manages user assignments
Microsoft Entra IDControls user and group access
Microsoft PurviewApplies compliance, DLP, retention, sensitivity labels, and governance
SharePoint Advanced ManagementControls content access and oversharing protection
Microsoft DefenderProtects against threats affecting Copilot-accessible content
Individual Microsoft 365 AppsMay provide application-specific Copilot settings

These controls work together rather than independently.

Features That Can Be Enabled or Disabled

Administrators can control several Copilot capabilities.

1. Microsoft 365 Copilot Licenses

The most fundamental control is license assignment.

Without a license:

  • Users cannot access Microsoft 365 Copilot.
  • Copilot chat within Microsoft 365 apps is unavailable.
  • AI-powered productivity experiences remain disabled.

Administrators assign or remove licenses through the Microsoft 365 Admin Center.

2. Copilot Chat Availability

Organizations can choose whether users have access to:

  • Microsoft 365 Copilot Chat
  • Enterprise data grounding
  • AI conversations within Microsoft 365

This allows phased deployments.

Example:

  • IT department enabled
  • Executive team enabled
  • Finance enabled later
  • Entire organization enabled after testing

3. Copilot in Individual Microsoft 365 Apps

Copilot experiences exist across multiple applications, including:

  • Word
  • Excel
  • PowerPoint
  • Outlook
  • Teams
  • OneNote

Organizations may decide when to introduce Copilot features within these workloads depending on readiness and licensing.

4. Intelligent Meeting Features

Some Teams AI features can be managed by administrators, including:

  • Intelligent meeting recap
  • AI-generated meeting summaries
  • Suggested action items
  • Meeting notes
  • Transcript availability

Organizations handling confidential meetings may choose to limit some AI-generated meeting experiences.

5. Plugins and Connectors

Administrators can manage:

  • Microsoft Graph connectors
  • Third-party plugins
  • Custom connectors
  • Agent access to external systems

Disabling unnecessary plugins reduces security risk.

6. Copilot Agents

Administrators can control:

  • Which agents are available
  • Who can create agents
  • Who can publish agents
  • Which departments can access specific agents

For example:

Human Resources might publish an HR Benefits Agent while Finance publishes an Expense Policy Agent.

7. Web Grounding

Some Copilot experiences include information from:

  • Microsoft Graph
  • Public web content
  • Organizational content

Organizations may configure which experiences are available depending on licensing and organizational policies.

Features That Cannot Simply Be “Turned Off”

Some Copilot behaviors are governed by Microsoft 365 security rather than feature switches.

Examples include:

Microsoft Graph Permissions

Copilot never ignores permissions.

If a user lacks permission to a file:

  • Copilot cannot retrieve it.
  • There is no setting that overrides SharePoint permissions.

SharePoint Permissions

Copilot always honors:

  • Site permissions
  • Folder permissions
  • File permissions
  • Restricted SharePoint sites

Administrators manage access by changing SharePoint permissions—not Copilot settings.

Microsoft Purview Policies

If Microsoft Purview blocks data through:

  • Sensitivity labels
  • DLP policies
  • Retention policies

Copilot follows those controls automatically.

Microsoft Defender Policies

Security policies continue protecting data regardless of Copilot.

Examples include:

  • Safe Links
  • Safe Attachments
  • Threat protection
  • Malware detection

Copilot cannot bypass Defender protections.

Enabling Copilot Through Licensing

Most Copilot functionality depends on licensing.

Typical process:

  1. Purchase licenses.
  2. Assign licenses.
  3. Configure organizational settings.
  4. Enable users or groups.
  5. Monitor adoption.
  6. Expand deployment gradually.

Removing the license immediately removes access.

Feature Rollout Strategies

Many organizations deploy Copilot in phases.

Example rollout:

PhaseUsers
PilotIT department
Early adoptersBusiness champions
Department rolloutHR, Finance, Sales
Enterprise rolloutEntire organization

This minimizes disruption and allows administrators to gather feedback.

Feature Controls for Copilot Agents

Agent administrators can typically control:

  • Agent publishing
  • Agent availability
  • Knowledge sources
  • Connector permissions
  • Agent sharing
  • Agent lifecycle
  • Agent retirement

These settings help prevent unauthorized AI experiences.

Managing Experimental Features

Microsoft periodically releases:

  • Preview capabilities
  • Experimental AI experiences
  • Early-access functionality

Organizations can often choose whether these features are available.

Many enterprises disable preview features until internal testing is complete.

Monitoring Enabled Features

Administrators should monitor:

  • License assignments
  • Usage reports
  • Adoption metrics
  • Agent activity
  • Security alerts
  • Compliance reports
  • AI interactions (where supported)

Monitoring helps determine whether enabled features are providing value while remaining compliant.

Best Practices

Microsoft recommends:

  • Start with a pilot group.
  • Assign licenses only to intended users.
  • Review SharePoint permissions before deployment.
  • Apply Microsoft Purview protection policies first.
  • Enable only required plugins.
  • Monitor adoption regularly.
  • Review security settings before enabling new AI capabilities.
  • Use least-privilege access.
  • Periodically review agent permissions.
  • Train users before broad rollout.

Exam Tips

For the AB-900 exam, remember these key points:

  • Licensing is the primary method of enabling Microsoft 365 Copilot.
  • Administrators can enable or disable access for users and groups.
  • Copilot always respects Microsoft Graph permissions.
  • Microsoft Purview protections continue to apply to Copilot.
  • SharePoint permissions cannot be bypassed by Copilot.
  • Administrators can manage plugins, connectors, and agents.
  • Many organizations use phased deployments.
  • Security and governance controls remain in effect regardless of Copilot features.

10 Practice Exam Questions

Question 1

What is the primary requirement for a user to access Microsoft 365 Copilot?

A. Membership in the Global Readers group

B. Assignment of an appropriate Microsoft 365 Copilot license

C. Creation of a Copilot agent

D. A Microsoft Teams Premium license

Correct Answer: B

Explanation: A Microsoft 365 Copilot license is required before users can access Copilot experiences.

Question 2

An administrator wants to introduce Copilot to only the IT department before rolling it out company-wide. What is the recommended approach?

A. Disable Microsoft Graph

B. Remove SharePoint permissions

C. Assign Copilot licenses only to the IT department

D. Create separate Microsoft 365 tenants

Correct Answer: C

Explanation: Administrators commonly pilot Copilot by assigning licenses only to selected users or groups.

Question 3

Which security principle does Microsoft 365 Copilot always follow?

A. It ignores file permissions for administrators.

B. It grants temporary access to files during conversations.

C. It respects existing Microsoft Graph and Microsoft 365 permissions.

D. It automatically shares documents across departments.

Correct Answer: C

Explanation: Copilot only accesses content the user already has permission to view.

Question 4

Which capability can administrators commonly control?

A. Whether users can access Copilot agents

B. Whether Copilot can ignore sensitivity labels

C. Whether Microsoft Graph indexes SharePoint

D. Whether SharePoint stores documents

Correct Answer: A

Explanation: Administrators can manage agent availability, publication, and access permissions.

Question 5

What happens if a user’s Microsoft 365 Copilot license is removed?

A. Existing AI conversations become public.

B. SharePoint permissions are deleted.

C. Copilot access is removed from that user.

D. Microsoft Graph stops indexing organizational content.

Correct Answer: C

Explanation: Removing the Copilot license removes the user’s entitlement to Copilot services.

Question 6

Which Microsoft technology automatically continues enforcing sensitivity labels when users work with Copilot?

A. Microsoft Defender for Endpoint

B. Microsoft Purview

C. Microsoft Intune

D. Microsoft Planner

Correct Answer: B

Explanation: Microsoft Purview applies data protection controls, including sensitivity labels, regardless of whether Copilot is used.

Question 7

Why might an organization disable certain Copilot plugins?

A. To reduce security risks from unnecessary external integrations

B. To increase Microsoft Graph indexing speed

C. To improve Outlook mailbox quotas

D. To eliminate SharePoint storage limits

Correct Answer: A

Explanation: Limiting plugins reduces the organization’s attack surface and helps maintain governance.

Question 8

Which feature continues protecting documents even after Copilot is enabled?

A. Microsoft Graph indexing

B. Microsoft Purview DLP policies

C. Copilot prompts

D. AI-generated summaries

Correct Answer: B

Explanation: Data Loss Prevention policies remain fully enforced when Copilot accesses organizational data.

Question 9

What is a common best practice when deploying Microsoft 365 Copilot?

A. Enable every Copilot feature for all employees immediately.

B. Remove SharePoint permissions before deployment.

C. Begin with a pilot deployment and expand gradually.

D. Disable Microsoft Purview during rollout.

Correct Answer: C

Explanation: A phased rollout allows administrators to validate security, governance, and user adoption before organization-wide deployment.

Question 10

Which statement about SharePoint permissions and Copilot is correct?

A. Copilot can temporarily bypass SharePoint permissions.

B. Copilot automatically grants access to related files.

C. Administrators can disable SharePoint permissions while keeping Copilot enabled.

D. Copilot only accesses SharePoint content the user is already authorized to view.

Correct Answer: D

Explanation: Copilot always honors existing SharePoint permissions and cannot access content beyond the user’s authorized access.

Go to the AB-900 Exam Prep Hub main page

AB-900, Agentic AI, AI, Generative AI, Microsoft Certification

Compare Copilot monthly license model to Pay-as-You-Go, including SharePoint (AB-900 Exam Prep)

 
This post is a part of the AB-900: Microsoft 365 Copilot and Agent Administration Fundamentals Exam Prep Hub.
This topic falls under these sections:
Perform basic administrative tasks for Copilot and agents (25–30%)
   --> Understand features and capabilities of Copilot and agents
      --> Compare Copilot monthly license model to Pay-as-You-Go, including SharePoint

Note that there are 10 practice questions (with answers) at the end of each section to help you solidify your knowledge of the material. Also, there are 4 practice tests with 30 questions each available from the hub's main page below the exam topics section.

Introduction

Microsoft offers multiple licensing models for AI experiences across Microsoft 365. Understanding these licensing options is important for administrators who plan deployments, manage costs, and determine which AI capabilities are available to users.

For the AB-900 exam, you should understand the differences between:

  • Microsoft 365 Copilot monthly user licensing
  • Pay-as-you-go (consumption-based) licensing
  • SharePoint Copilot licensing
  • When each licensing model is appropriate

The exam focuses on understanding the concepts rather than memorizing pricing.

Why Multiple Licensing Models Exist

Organizations vary greatly in how employees use AI.

Some organizations:

  • Have employees who use AI all day.
  • Need AI integrated into Microsoft 365 apps.
  • Require predictable monthly costs.

Other organizations:

  • Use AI occasionally.
  • Need specialized agents.
  • Want to pay only when AI is used.

Microsoft therefore offers both subscription-based and consumption-based licensing.

Microsoft 365 Copilot Monthly License Model

The traditional Microsoft 365 Copilot license is assigned to individual users.

Each licensed user receives access to Copilot experiences across supported Microsoft 365 applications.

Examples include:

  • Word
  • Excel
  • PowerPoint
  • Outlook
  • Teams
  • OneNote
  • Microsoft 365 Chat

The license is:

  • Assigned per user
  • Monthly subscription
  • Predictable recurring cost

Characteristics of the Monthly License

The monthly model provides:

  • Full Microsoft 365 Copilot experience
  • Unlimited daily usage (subject to service limits)
  • Personalized AI assistance
  • Microsoft Graph integration
  • Cross-app experiences
  • Enterprise security and compliance

This model is best for employees who regularly use Copilot throughout their workday.

Typical Monthly License Scenario

A financial analyst uses Copilot every day to:

  • Analyze Excel workbooks
  • Draft reports
  • Summarize meetings
  • Create PowerPoint presentations
  • Search organizational knowledge

Because AI is used continuously, a monthly license provides predictable costs.

Benefits of Monthly Licensing

Advantages include:

  • Predictable budgeting
  • No need to monitor consumption
  • Continuous access
  • Simplified administration
  • Consistent user experience
  • Ideal for heavy users

Limitations of Monthly Licensing

Considerations include:

  • Fixed monthly cost regardless of usage
  • Not ideal for occasional users
  • Every user requires their own license
  • Organizations may over-license infrequent users

Pay-as-You-Go Licensing

Pay-as-you-go (PAYG) is a consumption-based licensing model.

Instead of paying for every user every month, organizations pay based on actual AI usage.

Think of it similarly to cloud computing services:

  • More usage = higher cost
  • Less usage = lower cost

Characteristics of Pay-as-You-Go

Pay-as-you-go provides:

  • Usage-based billing
  • Flexible scaling
  • No requirement for every user to have a monthly Copilot license
  • Cost based on AI requests or service consumption (depending on the service)

This model is especially useful for agents and certain AI scenarios.

Benefits of Pay-as-You-Go

Advantages include:

  • Lower upfront costs
  • Pay only for actual usage
  • Flexible deployment
  • Easy experimentation
  • Ideal for seasonal workloads
  • Good for occasional users

Limitations of Pay-as-You-Go

Potential drawbacks include:

  • Variable monthly costs
  • Budget forecasting is more difficult
  • Requires monitoring usage
  • Heavy usage may become more expensive than subscription licensing

Comparing Monthly Licensing and Pay-as-You-Go

Monthly LicensePay-as-You-Go
Fixed monthly costUsage-based cost
Licensed per userConsumption-based
Predictable budgetingVariable spending
Best for daily usersBest for occasional use
Continuous Copilot accessPay only when AI is used
Simpler cost managementRequires usage monitoring

Microsoft 365 Copilot Chat

Organizations should understand that Microsoft offers AI experiences beyond the traditional monthly Copilot license.

For example:

  • Microsoft 365 Copilot Chat is available to Microsoft 365 users.
  • Organizations can extend Copilot Chat with agents.
  • Some agent usage can be billed using pay-as-you-go licensing rather than requiring every user to have a full Copilot subscription.

This provides flexibility for organizations with mixed AI usage patterns.

SharePoint and Copilot

SharePoint includes AI capabilities that help users work with documents, sites, and organizational knowledge.

Examples include:

  • Summarizing documents
  • Answering questions about files
  • Generating page content
  • Assisting with document creation
  • Improving knowledge discovery

SharePoint Agents

One important capability is SharePoint agents.

A SharePoint agent can:

  • Be created from a SharePoint site or document library
  • Answer questions using approved SharePoint content
  • Help users locate organizational knowledge
  • Reduce the need to manually search documents

For example:

A Human Resources SharePoint site may contain:

  • Employee handbook
  • Benefits guide
  • Leave policies
  • Training documents

An HR SharePoint agent can answer employee questions using those documents.

SharePoint Pay-as-You-Go

Organizations can use SharePoint agents without assigning every user a full Microsoft 365 Copilot license.

Instead, administrators can configure consumption-based billing.

Benefits include:

  • Lower cost for occasional users
  • Easy pilot deployments
  • Department-specific AI
  • Flexible scaling

This makes SharePoint agents attractive for organizations wanting targeted AI experiences without licensing every employee.

Choosing the Right Licensing Model

Choose Monthly Licensing When

  • Employees use Copilot every day.
  • AI is integrated into daily workflows.
  • Predictable monthly budgeting is important.
  • Users need full Copilot functionality across Microsoft 365.

Examples:

  • Executives
  • Project managers
  • Analysts
  • Consultants
  • Sales professionals
  • Knowledge workers

Choose Pay-as-You-Go When

  • AI usage is occasional.
  • Organizations are testing AI.
  • Departments need specialized agents.
  • Seasonal usage is expected.
  • Budget flexibility is acceptable.

Examples:

  • HR help desk agent
  • Legal document agent
  • IT support chatbot
  • SharePoint knowledge assistant

Administrative Considerations

Administrators should evaluate:

  • Expected AI usage
  • Number of users
  • Cost predictability
  • Department requirements
  • Governance policies
  • Licensing strategy
  • Agent deployment plans

Security Remains the Same

Regardless of licensing model:

  • Microsoft Entra ID authentication is used.
  • Microsoft Graph permissions are enforced.
  • Microsoft Purview policies apply.
  • Data Loss Prevention (DLP) policies remain active.
  • Sensitivity labels continue protecting content.
  • Microsoft Defender protections remain in effect.

Licensing changes how organizations pay for AI—not how Microsoft secures organizational data.

Best Practices

Microsoft recommends that organizations:

  • License frequent users with Microsoft 365 Copilot subscriptions.
  • Use pay-as-you-go for occasional AI usage.
  • Monitor AI adoption and consumption.
  • Start with pilot deployments.
  • Evaluate SharePoint agents for departmental knowledge scenarios.
  • Review licensing regularly as adoption increases.

Exam Tips

For the AB-900 exam, remember these key points:

  • Microsoft 365 Copilot is commonly licensed per user with a monthly subscription.
  • Pay-as-you-go bills organizations based on AI usage.
  • Monthly licensing provides predictable costs.
  • Pay-as-you-go offers flexibility for occasional or specialized AI use.
  • SharePoint agents can be deployed using consumption-based licensing in supported scenarios.
  • Licensing affects billing—not security or permissions.
  • Microsoft Graph, Microsoft Purview, and Microsoft Entra ID protections apply regardless of licensing model.
  • Heavy AI users are generally better suited to monthly licensing.
  • Departmental or pilot AI deployments often benefit from pay-as-you-go.

Practice Exam Questions

Question 1

Which licensing model provides users with a predictable monthly cost for Microsoft 365 Copilot?

A. Pay-as-you-go
B. Monthly per-user license
C. Azure consumption credits
D. SharePoint storage licensing

Correct Answer: B

Explanation: A monthly per-user license provides continuous access to Microsoft 365 Copilot for a fixed monthly subscription.

Question 2

What is the primary advantage of the pay-as-you-go licensing model?

A. Users receive unlimited AI usage regardless of activity.
B. Organizations pay only for actual AI usage.
C. Every employee automatically receives Microsoft 365 Copilot.
D. It disables Microsoft Graph integration.

Correct Answer: B

Explanation: Pay-as-you-go charges based on consumption, making it suitable for occasional or specialized AI usage.

Question 3

Which type of user is generally the best candidate for a Microsoft 365 Copilot monthly license?

A. An employee who rarely uses Microsoft 365 applications
B. A seasonal contractor who accesses AI once a month
C. A knowledge worker who uses Copilot throughout the workday
D. A visitor with guest access to SharePoint

Correct Answer: C

Explanation: Heavy or daily users benefit from the predictable costs and continuous access provided by the monthly licensing model.

Question 4

An organization wants to deploy an HR SharePoint agent that employees will use occasionally. Which licensing model is often the better fit?

A. Monthly Copilot license for every employee
B. Windows Enterprise licensing
C. Exchange Online licensing
D. Pay-as-you-go

Correct Answer: D

Explanation: Pay-as-you-go is well suited for departmental agents with occasional usage, allowing organizations to pay based on consumption.

Question 5

Which statement about Microsoft 365 Copilot monthly licensing is correct?

A. It charges only when AI is used.
B. It is assigned to individual users as a subscription.
C. It replaces Microsoft Entra ID.
D. It is available only for SharePoint.

Correct Answer: B

Explanation: The traditional Microsoft 365 Copilot model is licensed per user through a recurring subscription.

Question 6

Which capability is commonly associated with SharePoint agents?

A. Managing Windows updates
B. Replacing Microsoft Graph
C. Answering questions using SharePoint content and document libraries
D. Creating Azure virtual machines

Correct Answer: C

Explanation: SharePoint agents are grounded in SharePoint content and help users locate and understand organizational knowledge.

Question 7

How do Microsoft Purview policies behave when an organization switches from monthly licensing to pay-as-you-go?

A. They are automatically disabled.
B. They apply only to SharePoint documents.
C. They require users to purchase additional licenses before functioning.
D. They continue to protect data regardless of the licensing model.

Correct Answer: D

Explanation: Security and compliance controls such as Microsoft Purview continue to protect data regardless of how AI services are licensed.

Question 8

Which licensing model generally provides the most predictable monthly budgeting?

A. Pay-as-you-go
B. Monthly per-user licensing
C. Azure Reserved Instances
D. SharePoint storage quotas

Correct Answer: B

Explanation: Monthly licensing offers a fixed recurring cost, simplifying budgeting and financial planning.

Question 9

What is a potential disadvantage of pay-as-you-go licensing?

A. It cannot be used with agents.
B. It prevents users from accessing SharePoint.
C. Monthly costs may vary depending on AI usage.
D. It disables Microsoft Graph permissions.

Correct Answer: C

Explanation: Consumption-based billing means costs fluctuate according to actual usage, making budgeting less predictable.

Question 10

Which statement best summarizes the difference between Microsoft 365 Copilot monthly licensing and pay-as-you-go?

A. Monthly licensing is subscription-based, while pay-as-you-go is consumption-based.
B. Monthly licensing does not include Microsoft Graph.
C. Pay-as-you-go removes Microsoft Purview protections.
D. Monthly licensing is only available for SharePoint.

Correct Answer: A

Explanation: The fundamental difference is the billing model: monthly licensing charges a fixed subscription per user, whereas pay-as-you-go charges based on actual AI service consumption.

Go to the AB-900 Exam Prep Hub main page