Microsoft Power Platform Fundamentals

Apps and automations need data kept somewhere reliable. You learn how Dataverse stores data in tables, how columns and relationships model a business, and how standard and custom tables differ.

• Tables, rows and columns
• Standard and custom tables
• Column types and choices
• Relationships between tables
• Business rules for simple logic

You build A Dataverse table design for a small request-tracking process, with its relationships.

Where you build and what you connect to shape risk and reach. You learn what environments are, how connectors link to other services, and how administrators use policies and the admin center to keep builders in bounds.

• Environments and their purposes
• Connectors: standard, premium and custom
• Data policies that control which connectors mix
• The Power Platform admin center
• Solutions as containers for your work

You build An environment and connector plan for a team that builds both test and production apps.

Low-code does not mean low-risk. You learn how security roles, business units and field security control access to Dataverse data, and how the platform supports privacy and compliance obligations.

• Security roles and what they allow
• Business units and teams
• Field-level security
• Sign-in with Microsoft Entra ID
• Privacy and compliance support in the platform

You build A security design that gives three kinds of users the right access to one table.

A canvas app starts from a blank screen and a clear task. You learn how screens, controls and formulas work, how an app reads from data sources, and which problems suit this kind of app.

• Screens, controls and layout
• Power Fx formulas for beginners
• Connecting to data sources
• Galleries, forms and navigation
• Where canvas apps fit and where they do not

You build A simple canvas app that lists and adds records in a Dataverse table.

When an app is mostly about structured data and process, model-driven fits. You learn how forms, views and dashboards come from the data model, and how business process flows guide users through steps.

• How a model-driven app is generated from data
• Forms, views and charts
• Dashboards and site maps
• Business process flows
• Choosing canvas or model-driven

You build A model-driven app with a form, a view and a dashboard on the same Dataverse tables.

AI can now draft an app from a description. You use Copilot and the plan designer to turn an idea into a plan and a working draft, then check and adjust what it produced.

• Describing an app in plain language
• The plan designer and its output
• Reviewing generated tables and screens
• Improving a draft with follow-up prompts
• Knowing when to fix by hand

You build An app drafted from a plain-language description, with a list of what you corrected.

Automation removes the repetitive part of a process. You build cloud flows with triggers and actions, add approvals, and learn the difference between automated, instant and scheduled flows.

• Triggers and actions
• Automated, instant and scheduled flows
• Conditions, loops and simple expressions
• Approvals with people in the loop
• Run history and fixing a failed run

You build A cloud flow that sends a new request for approval and records the answer.

Some systems have no connector, only a screen. You learn how desktop flows automate those tasks, how attended and unattended runs differ, and how AI helps you draft flows from a description.

• What robotic process automation does
• Attended and unattended desktop flows
• Recording and editing actions
• Calling a desktop flow from a cloud flow
• Drafting flows with Copilot and checking the result

You build A desktop flow that copies data between two applications, triggered from a cloud flow.

An agent answers questions and takes action in a conversation. You learn how topics, knowledge sources and tools combine to make one, and where agents fit alongside apps and flows.

• What an agent is and what it can do
• Topics and how conversations are triggered
• Knowledge sources for grounded answers
• Tools and actions an agent can use
• Publishing to channels

You build A Copilot Studio agent that answers staff questions from a small set of documents.

An agent that nobody checks drifts. You read analytics, test an agent against sample questions, and learn how agent management across an organization, including Agent 365, supports oversight.

• Analytics for an agent’s use and answers
• Writing test questions and checking results
• Improving topics and knowledge from findings
• Agent 365 and oversight of agents across an organization
• Responsible use of agents

You build A test set and results sheet for an agent, with three improvements made from it.

One process joins the course. You choose a business process, model its data in Dataverse, build an app to capture it, a flow to move it along, and an agent to answer questions, and then review the whole.

• Choosing and describing a business process
• Modeling the data in Dataverse
• Building the app and the flow
• Adding an agent
• Reviewing the solution for security and value

You build A small working solution of one app, one flow and one agent around a single business process.