🔒 Guided

Pre-launch preview. Authorised access only.

Incorrect code

Guided by A Guide to Cloud
Explore AB-900 AI-901
Guided AB-731 Domain 1
Domain 1 — Module 2 of 11 18%
2 of 27 overall

AB-731 Study Guide

Domain 1: Identify the Business Value of Generative AI Solutions

  • Generative AI vs Traditional AI: What's the Difference?
  • Choosing the Right AI Solution for Your Business
  • AI Models: Pretrained vs Fine-Tuned
  • AI Cost Drivers and ROI: Tokens, Pricing, and Business Cases
  • Challenges of Generative AI: Fabrications, Bias & Reliability
  • When Generative AI Creates Real Business Value
  • Prompt Engineering: The Skill That Multiplies AI Value
  • RAG and Grounding: Making AI Use YOUR Data
  • Data Quality: The Make-or-Break Factor for AI
  • When Traditional Machine Learning Adds Value
  • Securing AI Systems: From Application to Data

Domain 2: Identify Benefits, Capabilities, and Opportunities for Microsoft AI Apps and Services

  • Mapping Business Needs to Microsoft AI Solutions
  • Copilot Versions: Free, Business, M365, and Beyond
  • Copilot Chat: Web, Mobile & Work Experiences
  • Copilot in M365 Apps: Word, Excel, Teams & More
  • Copilot Studio & Microsoft Graph: Building Smarter Solutions
  • Researcher & Analyst: Copilot's Power Agents
  • Build, Buy, or Extend: The AI Decision Framework
  • Microsoft Foundry: Your AI Platform
  • Azure AI Services: Vision, Search & Beyond
  • Matching the Right AI Model to Your Business Need

Domain 3: Identify an Implementation and Adoption Strategy

  • Responsible AI and Governance: Principles That Protect Your Business Free
  • Setting Up an AI Council: Strategy, Oversight & Alignment Free
  • Building Your AI Adoption Team Free
  • AI Champions: Your Secret Weapon for Adoption Free
  • Data, Security, Privacy & Cost: The Four Pillars of AI Readiness Free
  • Copilot & Azure AI Licensing: Every Option Explained Free

AB-731 Study Guide

Domain 1: Identify the Business Value of Generative AI Solutions

  • Generative AI vs Traditional AI: What's the Difference?
  • Choosing the Right AI Solution for Your Business
  • AI Models: Pretrained vs Fine-Tuned
  • AI Cost Drivers and ROI: Tokens, Pricing, and Business Cases
  • Challenges of Generative AI: Fabrications, Bias & Reliability
  • When Generative AI Creates Real Business Value
  • Prompt Engineering: The Skill That Multiplies AI Value
  • RAG and Grounding: Making AI Use YOUR Data
  • Data Quality: The Make-or-Break Factor for AI
  • When Traditional Machine Learning Adds Value
  • Securing AI Systems: From Application to Data

Domain 2: Identify Benefits, Capabilities, and Opportunities for Microsoft AI Apps and Services

  • Mapping Business Needs to Microsoft AI Solutions
  • Copilot Versions: Free, Business, M365, and Beyond
  • Copilot Chat: Web, Mobile & Work Experiences
  • Copilot in M365 Apps: Word, Excel, Teams & More
  • Copilot Studio & Microsoft Graph: Building Smarter Solutions
  • Researcher & Analyst: Copilot's Power Agents
  • Build, Buy, or Extend: The AI Decision Framework
  • Microsoft Foundry: Your AI Platform
  • Azure AI Services: Vision, Search & Beyond
  • Matching the Right AI Model to Your Business Need

Domain 3: Identify an Implementation and Adoption Strategy

  • Responsible AI and Governance: Principles That Protect Your Business Free
  • Setting Up an AI Council: Strategy, Oversight & Alignment Free
  • Building Your AI Adoption Team Free
  • AI Champions: Your Secret Weapon for Adoption Free
  • Data, Security, Privacy & Cost: The Four Pillars of AI Readiness Free
  • Copilot & Azure AI Licensing: Every Option Explained Free
Domain 1: Identify the Business Value of Generative AI Solutions Premium ⏱ ~12 min read

Choosing the Right AI Solution for Your Business

Not every problem needs generative AI. Learn to match the right type of AI — from predictive models to Copilot to custom Foundry solutions — to the business problem in front of you.

How do you choose the right AI solution?

☕ Simple explanation

Think of AI solutions like tools in a workshop. You wouldn’t use a hammer to cut wood.

Some business problems need prediction (“Who will churn?”). Some need automation (“Route tickets to the right team”). Some need creation (“Draft a proposal from our last 10 winning bids”).

Choosing the right AI solution starts with the problem, not the technology. Once you know what you’re solving, the right tool becomes obvious.

Selecting the right generative AI solution requires mapping business requirements to solution categories. The key dimensions are:

  • Problem type: Prediction, classification, generation, or augmentation?
  • Data requirements: Does the solution need access to proprietary data? How sensitive is it?
  • User audience: Is this for knowledge workers, customers, or automated workflows?
  • Customisation level: Off-the-shelf, configured, or custom-built?
  • Scale and cost: How many users? What’s the per-interaction cost tolerance?

For generative AI specifically, the spectrum ranges from ready-made productivity tools (Microsoft 365 Copilot) to fully custom applications (built with Microsoft Foundry and Azure OpenAI Service).

The AI solution spectrum

Not all generative AI solutions are created equal. They range from ready-to-use to fully custom:

The generative AI solution spectrum — from ready-made to fully custom. Prices shown are approximate list prices and may vary by agreement and region.
FeatureCustomisationCostTime to deployBest for
Microsoft 365 CopilotLow — use as-is with your M365 data$30/user/month add-onDays (enable licences)Productivity: drafting, summarising, analysing across M365 apps
Copilot Studio agentsMedium — custom agents with your data sourcesIncluded with Copilot or PAYGWeeksSpecific workflows: HR FAQ bot, policy lookup, onboarding helper
Azure OpenAI ServiceHigh — custom prompts, fine-tuning, RAGPay-per-tokenWeeks to monthsCustom applications: customer-facing chatbots, document processors
Microsoft FoundryFull — end-to-end AI app developmentPay-per-use + computeMonthsComplex solutions: multi-model orchestration, custom agents, enterprise AI apps

Decision framework: Four questions to ask

Before recommending a solution, ask these four questions:

1. Who is the user?

User TypeBest Starting PointWhy
Knowledge workers (internal)Microsoft 365 CopilotAlready embedded in their daily tools
Customers (external)Azure OpenAI + custom appNeed branded, controlled experience
Specific teams with specific dataCopilot Studio agentsGrounded in team-specific knowledge sources
Developers building AI productsMicrosoft FoundryFull control over models, data, and deployment

2. What data does the AI need?

  • M365 data only (emails, documents, meetings) → Microsoft 365 Copilot
  • Specific knowledge bases (policies, product catalogues) → Copilot Studio agents with connectors
  • External or proprietary data → Azure OpenAI with RAG pattern
  • Multiple data sources, complex orchestration → Microsoft Foundry

3. How much control do you need?

  • Minimal — just want AI in existing workflows → Copilot
  • Some — custom prompts, specific knowledge sources → Copilot Studio
  • Full — control every aspect of the AI experience → Foundry + Azure OpenAI

4. What’s the budget model?

  • Per-user monthly → Copilot (~$30/user/month)
  • Per-use consumption → Azure OpenAI (tokens), Copilot Studio (messages)
  • Mixed → Enterprise agreements combining per-user and consumption
💡 Scenario: Elena's firm chooses three AI solutions

Elena’s consulting firm (Meridian Consulting) doesn’t pick just one — they pick the right solution for each problem:

  1. Proposal drafting → Microsoft 365 Copilot (works inside Word with existing documents)
  2. New hire onboarding FAQ → Copilot Studio agent (grounded in the HR SharePoint site)
  3. Client-facing research portal → Azure OpenAI + custom app (branded experience, external users, proprietary data)

This is a common pattern: most organisations use multiple AI solutions for different use cases.

When generative AI is NOT the answer

Not every problem needs generative AI. Sometimes traditional AI or even simple automation is better:

ScenarioBetter AlternativeWhy Not Gen AI?
Predicting equipment failureTraditional ML (predictive maintenance)You need precise numerical predictions, not generated content
Routing support ticketsClassification modelCategorisation is faster and cheaper with traditional ML
Processing invoices with fixed formatsRule-based automation (RPA)Structured extraction doesn’t need the flexibility of an LLM
Real-time fraud detectionAnomaly detection modelSpeed and precision matter more than language generation
💡 Exam tip: 'Select a generative AI solution' questions

When the exam says “select a generative AI solution to meet a business need,” it’s testing whether you can:

  1. Recognise when gen AI is appropriate (vs traditional AI or automation)
  2. Choose the right Microsoft gen AI product (Copilot vs Studio vs Azure OpenAI vs Foundry)
  3. Match the solution to the constraints (budget, users, data, customisation needs)

The wrong answer often suggests gen AI for a problem that’s better solved by traditional ML or simple automation.

Key flashcards

Question

When should you choose Microsoft 365 Copilot over Azure OpenAI Service?

Click or press Enter to reveal answer

Answer

Choose Copilot when the users are internal knowledge workers who need AI embedded in their existing M365 apps (Word, Excel, Teams). Choose Azure OpenAI when you need custom applications, external-facing AI, or fine-grained control.

Click to flip back

Question

What are the four key questions for selecting an AI solution?

Click or press Enter to reveal answer

Answer

1. Who is the user? 2. What data does the AI need? 3. How much control do you need? 4. What's the budget model?

Click to flip back

Question

When is generative AI NOT the right choice?

Click or press Enter to reveal answer

Answer

When you need precise numerical predictions (use ML), fast categorisation (use classification), structured data extraction from fixed formats (use RPA), or real-time detection (use anomaly detection).

Click to flip back

Knowledge check

Knowledge Check

Ravi's company needs a customer-facing chatbot that answers questions about their product catalogue. The chatbot must be branded and integrated into their website. Which solution should he choose?

Knowledge Check

Tomás oversees 5,000 workers at PacificSteel Manufacturing. His maintenance team wants AI to predict when conveyor belt motors will fail before they cause downtime. What type of AI solution is most appropriate?

Knowledge Check

Elena wants her consultants to use AI to summarise client meeting notes in Microsoft Teams. Budget is a concern. What's the most cost-effective solution?

🎬 Video coming soon

Next up: AI Models: Pretrained vs Fine-Tuned — understanding the building blocks of generative AI.

← Previous

Generative AI vs Traditional AI: What's the Difference?

Next →

AI Models: Pretrained vs Fine-Tuned

Guided

I learn, I simplify, I share.

A Guide to Cloud YouTube Feedback

© 2026 Sutheesh. All rights reserved.

Guided is an independent study resource and is not affiliated with, endorsed by, or officially connected to Microsoft. Microsoft, Azure, and related trademarks are property of Microsoft Corporation. Always verify information against Microsoft Learn.