Free PL-900 Exam Flashcards

Power Apps lets business users and developers build custom apps with little or no code, replacing manual paper/Excel processes with structured, mobile-ready data collection and workflows tied to organizational data.

Power Automate reduces repetitive manual work by automating multi-step processes across apps and services (approvals, notifications, file processing), freeing employees to focus on higher-value tasks.

Dataverse gives organizations a secure, structured place to store and manage business data that Power Apps, Power Automate, and other Microsoft services can share, instead of siloed spreadsheets or disconnected lists.

Connectors let apps and flows integrate with hundreds of external services and data sources (e.g., SharePoint, Outlook, Salesforce) without writing custom integration code, so a single app can pull from many systems.

Power Pages lets organizations build secure, low-code external-facing websites so people outside the company (customers, partners, citizens) can view or submit data that flows back into Dataverse.

Unlike a raw database, Dataverse ships with built-in security roles, business logic, standard/custom tables, and native integration across Power Apps, Power Automate, and Dynamics 365 — reducing the setup work traditional databases require.

Tables store a specific type of data (like Rows in a spreadsheet); columns define the fields within a table (like text, number, or choice fields). Microsoft provides standard tables (e.g., Account, Contact) and lets you add custom tables/columns.

Relationships connect tables so related records stay linked, e.g., a one-to-many relationship lets one Account have many Contacts, while a many-to-many relationship lets records on both sides link to multiple records on the other.

Forms control how a single record's data is displayed and edited (used in model-driven apps); views control how a list of records is filtered, sorted, and displayed as a table.

Dataverse supports business rules (no-code validation/logic), Power Fx formula columns, business process flows, and workflows/plug-ins for more advanced automation — all without needing a full custom application.

Copilot-style AI assistance in Dataverse can generate tables, columns, and relationships from a natural-language description of the data you need, speeding up data modeling for non-technical makers.

An environment is a container that separates and secures data, apps, and flows for a specific purpose (e.g., Development, Test, Production). Every tenant gets a Default environment, and admins can create additional Sandbox, Production, Trial, or Developer environments.

Access to Dataverse data is controlled through security roles (bundles of permissions), business units (organizational divisions), and teams — allowing granular control over who can view, edit, or share records.

Admins use Data Loss Prevention (DLP) policies to control which connectors can be combined, and Power Platform integrates with tools like Microsoft Purview for compliance, while apps should follow accessibility guidelines for users with disabilities.

The Power Platform admin center provides analytics dashboards (app usage, flow runs, errors) so admins can monitor adoption, troubleshoot failures, and understand how apps and flows are performing across the organization.

Power Platform pipelines move solutions through Development, Test, and Production environments in a controlled, repeatable way — typically using managed solutions in later stages to prevent uncontrolled edits.

Canvas apps start from a blank canvas where the maker drags and drops controls to design the exact screen layout, ideal when the UI needs to be highly customized around a specific business process.

Canvas apps connect to 1,000+ data sources via connectors, use the Power Fx formula language for logic, and let makers freely position screens, galleries, forms, and controls pixel-by-pixel.

Model-driven apps are generated automatically from your Dataverse data model (tables, forms, views, relationships) — best when you need a structured, data-first app without designing every screen manually.

Model-driven apps are responsive by default, enforce Dataverse security roles and business rules automatically, and can incorporate business process flows to guide users step-by-step through a data entry process.

Canvas apps prioritize design flexibility and can start from little or no existing data; model-driven apps prioritize consistency and speed by generating the UI directly from a Dataverse data model.

Power Fx is the low-code, Excel-like formula language used across Power Apps (and Dataverse formula columns) to define logic such as calculations, validations, and conditional behavior without traditional code.

Plan Designer is an AI-assisted planning tool that helps makers describe an app idea in natural language and generates a structured plan (screens, data, logic) before the app is built — bridging idea to build.

Code apps let professional developers build apps using standard code (e.g., React) while still running within Power Platform's governance, security, and Dataverse connectivity — bridging pro-code and low-code development.

Power Apps Studio is the design environment where makers build and edit both canvas and model-driven apps — adding screens, controls, data connections, and formulas.

Makers can describe the app they want in natural language, and AI generates an initial canvas app — screens, data connections, and basic layout — that the maker then refines manually.

AI can help scaffold a model-driven app by suggesting or generating the underlying Dataverse tables, relationships, and forms based on a natural-language description of the business scenario.

The "vibe" app-creation experience lets a maker describe what they want conversationally and iteratively refine the app through further natural-language prompts, rather than manually configuring every screen and control.

Cloud flows are Power Automate workflows that run in the cloud and connect online services and data sources; the three main types are automated, instant, and scheduled flows.

An automated flow starts automatically in response to a trigger event, such as a new email arriving, a file being added to a folder, or a new record being created in Dataverse.

An instant flow is started manually by a user, typically by clicking a button in the Power Automate mobile app, a Power Apps app, or a Teams/SharePoint action — used for on-demand tasks.

A scheduled flow runs automatically at a specific time or on a recurring interval (e.g., every night at midnight) rather than in response to an event or manual click — useful for regular batch tasks like nightly reports.

Desktop flows use robotic process automation (RPA) to record and replay user actions on legacy desktop applications and websites that don't have modern APIs, automating tasks that cloud flows can't reach directly.

The Approvals capability lets flows route requests (like expense reports or document sign-off) to one or more approvers, who can approve or reject directly from email, Teams, or the Power Automate app.

Power Automate has deep prebuilt integration with Microsoft 365 apps — for example, saving Outlook attachments to SharePoint, posting Teams messages on form submission, or routing Microsoft Forms responses into an approval flow.

Power Automate can automate document-centric processes — extracting data from incoming documents (invoices, forms), routing them for review, and filing or archiving them — reducing manual document handling.

A trigger is the event that starts a cloud flow (e.g., "when a new item is created"); actions are the steps the flow performs afterward (e.g., "send an email," "create a record"). Every flow needs exactly one trigger and can have many actions.

Makers can describe the automation they want in natural language, and AI generates or edits the trigger and action steps of a cloud flow, reducing the need to manually configure each step.

AI assistance can help generate or adjust the recorded steps of a desktop (RPA) flow from a natural-language description, speeding up automation of legacy desktop and web-based processes.

Microsoft Copilot Studio is a low-code tool for building AI-powered conversational agents (chatbots/copilots) that can answer questions, complete tasks, and take actions — for scenarios like customer support or employee self-service.

Topics define specific conversation paths within an agent — each topic has trigger phrases that start it and a sequence of dialog steps or actions the agent follows to handle that particular request.

Knowledge sources (like SharePoint sites, public websites, uploaded files, or Dataverse) ground an agent's answers in your organization's actual content, letting it answer questions beyond its predefined topics.

Beyond scripted topics, an agent can use generative AI to compose answers dynamically from connected knowledge sources, handling questions that weren't explicitly anticipated by the maker.

Tools let an agent go beyond conversation and take real actions — such as looking up a record, updating data, or calling an external system — during a conversation with a user.

Model Context Protocol (MCP) servers are a standardized way to connect external tools and data sources to an agent, letting Copilot Studio agents call outside systems through a consistent, reusable interface.

Agent flows are Power Automate-style flows built to run as tools inside an agent, letting an agent trigger multi-step automated processes as part of answering or acting on a user's request.

After building an agent, makers publish it to one or more channels — such as Microsoft Teams, a website, or other messaging platforms — so end users can actually interact with it.

Microsoft Agent 365 provides organization-wide governance and management for AI agents — giving admins visibility, control, and security over the agents deployed across the company, similar to how Microsoft 365 manages users.

Copilot Studio provides analytics on how agents are used — conversation volume, resolution rates, and escalations — helping admins and makers understand adoption and identify where an agent needs improvement.

Agent evaluations test an agent's responses against expected outcomes before and after publishing, helping makers catch inaccurate or unhelpful answers and measure quality over time.