AB-731, AI Governance, AI Strategy, Microsoft Certification

Identify when to build, buy, or extend AI solutions (AB-731 Exam Prep)

 
This post is a part of the AB-731: AI Transformation Leader Exam Prep Hub.
This topic falls under these sections:
Identify benefits, capabilities, and opportunities for Microsoft’s AI apps and services (35–40%)
   --> Identify benefits and capabilities of Microsoft 365 Copilot and Microsoft Copilot
      --> Identify when to build, buy, or extend AI solutions

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 most important responsibilities of an AI Transformation Leader is deciding how an AI capability should be delivered. Organizations generally have three choices:

  1. Buy an existing AI solution.
  2. Extend an existing Microsoft AI solution.
  3. Build a custom AI solution.

Selecting the correct approach affects cost, time-to-value, risk, maintenance requirements, and long-term flexibility.

Why This Decision Matters

Not every business problem requires a custom AI application.

Many organizations already have access to AI capabilities through:

  • Microsoft 365 Copilot
  • Microsoft Copilot Chat
  • Microsoft Copilot Studio
  • Dynamics 365 Copilot experiences
  • Power Platform
  • Azure AI services

Building a custom solution when an existing capability already meets the requirement can increase:

  • Cost
  • Development effort
  • Security risk
  • Maintenance burden
  • Adoption challenges

The goal is to achieve maximum business value with minimum complexity.

The Three Approaches

Buy

Buy means adopting a ready-made Microsoft AI solution.

Examples include:

  • Microsoft 365 Copilot
  • Microsoft Copilot Chat
  • Dynamics 365 Copilot
  • GitHub Copilot
  • Security Copilot
  • Power BI Copilot

Advantages

  • Fast deployment
  • Lower risk
  • Minimal development effort
  • Built-in security and governance
  • Microsoft-managed updates

Best Use Cases

  • Common productivity scenarios
  • Meeting summaries
  • Email drafting
  • Document creation
  • Data analysis
  • Standard customer service scenarios

Example

A company wants employees to summarize meetings, draft emails, and create presentations.

Best approach: Buy Microsoft 365 Copilot.

Extend

Extend means enhancing an existing Microsoft AI solution with organization-specific capabilities.

This approach provides:

  • Faster implementation than building from scratch.
  • Customization without recreating core AI functionality.
  • Access to enterprise data and business systems.

Examples

  • Connecting Copilot to Salesforce.
  • Adding custom actions.
  • Integrating ServiceNow.
  • Creating custom agents.
  • Using plugins and connectors.
  • Adding knowledge sources.

Advantages

  • Faster time-to-value.
  • Lower cost than custom development.
  • Leverages Microsoft’s security and orchestration.
  • Preserves existing investments.

Best Use Cases

  • Existing AI tools satisfy most requirements.
  • Additional business processes must be incorporated.
  • Integration with enterprise systems is needed.

Build

Build means creating a completely custom AI application.

Organizations typically use:

  • Azure AI Foundry
  • Azure OpenAI Service
  • Azure AI Search
  • Azure AI Services
  • Custom machine learning models

Advantages

  • Maximum flexibility.
  • Full control.
  • Highly specialized experiences.

Disadvantages

  • Highest cost.
  • Longer implementation times.
  • Increased maintenance responsibilities.
  • Greater governance requirements.

Best Use Cases

  • Unique competitive differentiators.
  • Industry-specific requirements.
  • Specialized workflows unavailable in existing products.

Example

A medical research company creates a proprietary clinical-analysis assistant trained on internal datasets.

Best approach: Build.

Decision Framework

Ask the following questions:

1. Does Microsoft already provide the capability?

If yes, prefer Buy.

2. Does an existing Copilot solve most of the problem?

If yes, consider Extend.

3. Is the requirement unique or strategic?

If yes, consider Build.

4. How quickly must value be delivered?

  • Buy → fastest
  • Extend → moderate
  • Build → longest

5. What level of maintenance is acceptable?

  • Buy → minimal maintenance
  • Extend → moderate maintenance
  • Build → highest maintenance

Comparison of Build, Buy, and Extend

FactorBuyExtendBuild
Time to deployFastestModerateSlowest
CostLowestMediumHighest
CustomizationLimitedModerateHighest
MaintenanceLowMediumHigh
Security managementMostly MicrosoftSharedOrganization responsibility
Best forStandard scenariosBusiness-specific enhancementsUnique solutions

Understanding Microsoft 365 Copilot Extensibility

Microsoft designed Microsoft 365 Copilot to be extensible rather than isolated.

Organizations can enhance Copilot without replacing it.

The extensibility framework allows businesses to:

  • Connect external systems.
  • Create custom agents.
  • Add specialized skills.
  • Access organizational knowledge.
  • Execute business actions.

This enables organizations to keep the productivity benefits of Microsoft 365 Copilot while tailoring experiences to their own processes.

Components of the Microsoft 365 Copilot Extensibility Framework

1. Copilot Studio

Copilot Studio enables organizations to:

  • Create custom copilots.
  • Build agents with low-code tools.
  • Connect to enterprise systems.
  • Define conversation flows.
  • Add automation.

Example

An HR department builds an onboarding agent that answers company-specific questions.

2. Connectors

Connectors allow Copilot to access external information.

Examples:

  • ServiceNow
  • Salesforce
  • SAP
  • Jira
  • Internal databases

This helps Copilot use information beyond Microsoft 365 content.

3. Graph Connectors

Graph connectors bring external content into Microsoft Graph.

Examples:

  • File repositories
  • CRM systems
  • Knowledge bases
  • SharePoint alternatives

This allows Copilot to retrieve and reason over additional organizational content.

4. Agents

Agents provide specialized experiences.

Examples:

IT Agent

Can:

  • Reset passwords.
  • Open tickets.
  • Provide troubleshooting instructions.

HR Agent

Can:

  • Explain policies.
  • Answer benefits questions.
  • Support onboarding.

Finance Agent

Can:

  • Retrieve budget information.
  • Explain expenses.
  • Generate reports.

5. Actions and Automations

Copilot can perform tasks, not just answer questions.

Examples:

  • Create tickets.
  • Submit forms.
  • Update records.
  • Trigger workflows.
  • Start Power Automate processes.

When to Extend Microsoft 365 Copilot

Extension is appropriate when:

✅ Microsoft 365 Copilot already solves most requirements.

✅ Business systems must be connected.

✅ Department-specific experiences are needed.

✅ Faster deployment is preferred.

✅ Customization is important but full development is unnecessary.

When to Build Instead of Extend

Building may be preferable when:

  • Requirements are highly unique.
  • Specialized models are required.
  • Proprietary intellectual property creates competitive advantage.
  • Regulatory requirements demand complete control.
  • Existing Copilot experiences cannot satisfy the scenario.

Example Scenarios

Scenario 1

Employees need help drafting emails and summarizing meetings.

Recommendation: Buy Microsoft 365 Copilot.

Scenario 2

Customer support employees need Microsoft 365 Copilot plus integration with ServiceNow.

Recommendation: Extend Microsoft 365 Copilot.

Scenario 3

A pharmaceutical company wants an AI system for proprietary drug research.

Recommendation: Build a custom AI solution.

Key Exam Points

Remember these principles:

  • Buy first whenever existing Microsoft solutions satisfy requirements.
  • Extend second when business-specific enhancements are needed.
  • Build last for highly specialized or differentiating scenarios.
  • Extending existing Copilot solutions often delivers faster ROI.
  • Microsoft 365 Copilot supports extensibility through:
    • Copilot Studio
    • Connectors
    • Graph connectors
    • Agents
    • Actions and automation
  • Custom development introduces greater cost and maintenance responsibilities.

Practice Exam Questions

Question 1

A company needs AI assistance for email drafting, meeting summaries, and presentation creation. No special requirements exist.

What is the best approach?

A. Build a custom AI application

B. Extend Microsoft 365 Copilot

C. Purchase Microsoft 365 Copilot

D. Create a machine learning model

Answer: C

Explanation: These are standard productivity scenarios already provided by Microsoft 365 Copilot. Buying provides the fastest and lowest-risk solution.

Question 2

Which approach generally requires the greatest development and maintenance effort?

A. Build

B. Buy

C. Extend

D. Use Copilot Chat only

Answer: A

Explanation: Custom-built solutions require ongoing development, infrastructure, monitoring, and governance.

Question 3

An organization already uses Microsoft 365 Copilot but wants employees to open ServiceNow tickets directly from Copilot.

Which approach is most appropriate?

A. Replace Copilot completely

B. Build a separate AI platform

C. Disable Copilot

D. Extend Microsoft 365 Copilot

Answer: D

Explanation: Since Copilot already satisfies most requirements, extending it with integrations provides the best value.

Question 4

Which factor most strongly favors the “buy” approach?

A. Need for proprietary AI models

B. Requirement for highly specialized algorithms

C. Desire for rapid time-to-value

D. Requirement for complete architectural control

Answer: C

Explanation: Purchased solutions provide the fastest deployment and quickest business value.

Question 5

Which Microsoft tool is primarily used to create custom agents and extend Copilot experiences?

A. Power BI

B. Microsoft Copilot Studio

C. Azure Virtual Machines

D. Microsoft Defender

Answer: B

Explanation: Copilot Studio enables low-code customization and agent development.

Question 6

A company’s AI capability represents a unique competitive advantage unavailable in commercial products.

Which strategy is usually most appropriate?

A. Buy

B. Extend

C. Outsource completely

D. Build

Answer: D

Explanation: Unique requirements often justify custom AI development.

Question 7

What is a major advantage of extending Microsoft 365 Copilot instead of building from scratch?

A. Eliminates governance requirements

B. Avoids all security concerns

C. Preserves existing Microsoft investments

D. Removes the need for connectors

Answer: C

Explanation: Extensions leverage Microsoft’s existing capabilities and infrastructure.

Question 8

Graph connectors primarily enable organizations to:

A. Train foundation models

B. Import external content into Microsoft Graph

C. Replace SharePoint

D. Eliminate data governance

Answer: B

Explanation: Graph connectors make external data available to Microsoft Graph and Copilot experiences.

Question 9

Which approach generally has the lowest operational burden?

A. Build

B. Extend

C. Hybrid custom development

D. Buy

Answer: D

Explanation: Microsoft manages most infrastructure, updates, and maintenance for purchased solutions.

Question 10

Which statement best describes the Microsoft 365 Copilot extensibility framework?

A. It allows organizations to enhance Copilot with agents, connectors, and actions.

B. It only supports custom machine learning models.

C. It replaces Microsoft Graph.

D. It requires organizations to build a new AI platform.

Answer: A

Explanation: The extensibility framework enables organizations to customize Copilot while retaining Microsoft’s core AI capabilities.

Go to the AB-731 Exam Prep Hub main page

Leave a comment