Working with Entities in Power Platform Copilot Studio

Introduction

If you’ve started exploring Copilot Studio in Microsoft Power Platform, you’ve probably come across the term “entities.” These entities are the backbone of natural language understanding in your chatbot. In this article, we’ll break down what Copilot Studio is, how entities work, and why custom entities can make your bot smarter and more context-aware.

What is Copilot Studio?

Copilot Studio is Microsoft’s low-code environment for building AI-powered chatbots that can be deployed across various platforms like Microsoft Teams, websites, and third-party channels. It’s part of the Power Platform ecosystem, which also includes Power Apps, Power Automate, and Power BI.

With Copilot Studio, even non-developers can build sophisticated conversational bots using a visual interface—yet it still provides enough flexibility for developers to extend the logic, integrate with APIs, and build custom scenarios.

Why Entities Matter in Copilot Studio?

At its core, a chatbot should understand what users are saying—and that’s where entities come in.

An entity in Copilot Studio helps the bot recognize key information (or data) from a user's input. Think of it like tagging important words or phrases that the bot can act on. Entities are used alongside topics and triggers to make conversations smarter and more personalized.

Go to the settings of your agents or bot and click settings.

Click settings

Common Use Cases for Entities

Extracting locations, dates, or names from questions
Recognizing product names, order numbers, or IDs
Creating guided interactions like booking forms, surveys, or service requests

Types of Entities in Copilot Studio

Copilot Studio provides two main types of entities.

1. Prebuilt Entities

These are ready-made entities provided by Microsoft to recognize common data types like.

Number
Dates and times
Emails and phone numbers
Currency
Age, temperature, URL, and more

These are perfect for common tasks and save time during development.

Development

2. Custom Entities

Custom entities are user-defined categories based on your specific business domain. You define the possible values and synonyms that the bot should recognize.

For example

ProductCategory: Mobile, Laptop, Tablet
SupportType: Billing Issue, Technical Issue, General Inquiry

How to Create a Custom Entity in Copilot Studio?

Creating a custom entity is straightforward and can greatly enhance the bot’s ability to understand your domain-specific vocabulary.

Step-by-Step Guide

Open Your Bot in Copilot Studio: Navigate to https://copilotstudio.microsoft.com and open your bot project.
Go to the Entities Section: In the left navigation panel, click on Entities.
Create a New Entity
- Click on + New Entity
- Give it a name, such as DeviceType
Select Closed List.
Add items/values under the entity (e.g., Laptop, Tablet, Smartphone)
And then click Save.

Step 1. Add Synonyms (Optional but Recommended).

For each value, you can add synonyms to make recognition more flexible.

Example

Value: Laptop
Synonyms: Notebook, Ultrabook, Desktop.

Step 2. Save the Entity.

Save the Entity

Step 3. Use the Entity in a Topic.

You can now use this entity in trigger phrases or questions within topics. For example.

“I need help with my Laptop.”
“My Notebook is not turning on.”

Copilot Studio will detect the value using your custom entity and store it as a variable in the conversation flow.

Purpose and Benefits of Using Entities

Here’s why entities are essential for your chatbot’s intelligence.

Improved Understanding: Bots can pick up specific pieces of information from vague or broad user input.
Dynamic Conversations: Extracted entity values can be used in conditions, actions, or API calls.
Personalization: Makes responses context-aware (e.g., “I see you’re having an issue with your Tablet.”)
Data Collection: Gather structured data from users for further processing or reporting.

Best Practices

Use prebuilt entities for standard data types to save time.
Add synonyms to make custom entities more flexible.
Keep entities domain-specific and avoid overlapping values.
Test entity recognition using the Test bot panel to ensure accurate matches.

Final Thoughts

Entities are the unsung heroes of conversational AI. By helping your bot extract meaningful data from user input, you make the entire interaction more useful, relevant, and engaging.

Whether you’re building a helpdesk bot, a product guide, or a scheduling assistant, leveraging custom and prebuilt entities in Power Platform Copilot Studio is a game-changer for building smarter conversations.