Building Tables with Copilot and Data Management
Use Copilot to create Dataverse tables through natural language conversations, and learn the options for importing and exporting data across your Power Platform environment.
Creating tables with Copilot
Imagine telling a colleague: “I need a table to track vehicle inspections with the date, inspector, vehicle, result, and any notes.”
With Copilot in Power Apps, you do exactly that — in plain English. Copilot reads your description, suggests a table with the right columns and data types, and even generates sample data so you can see what it looks like.
You review the suggestion, tweak anything that is not right, and the table is created. No clicking through menus. No guessing which data type to pick.
The Copilot conversation flow
Here is how Tom at Summit Realty uses Copilot to create a property listings table:
Step 1 — Describe what you need
Tom opens Power Apps, clicks “Start with data”, and selects “Describe the new table”. He types:
“I need a table to track real estate property listings with the address, listing price, number of bedrooms, number of bathrooms, square footage, listing date, status, and the assigned agent.”
Step 2 — Review the suggestion
Copilot generates a table called “Property Listing” with columns:
| Column | Data Type | Notes |
|---|---|---|
| Property Name | Text | Primary name column (auto-generated) |
| Address | Text (single line) | Street address |
| Listing Price | Currency | Dollar amount |
| Bedrooms | Whole number | Integer count |
| Bathrooms | Whole number | Integer count |
| Square Footage | Whole number | Area in sq ft |
| Listing Date | Date only | When the property was listed |
| Status | Choice | Active, Pending, Sold, Withdrawn |
| Assigned Agent | Text | Agent name |
Copilot also generates five sample rows with realistic data.
Step 3 — Refine with follow-up conversation
Tom notices the Assigned Agent column is plain text — he wants it to link to a Users table instead. He types:
“Change Assigned Agent to a lookup to the User table.”
Copilot updates the column. Tom also adds:
“Add a Yes/No column called Featured Listing.”
The table updates instantly.
Step 4 — Create the table
Tom clicks “Create” and the Dataverse table is provisioned. He can now build a Power App on top of it.
What Copilot can and cannot do
Copilot CAN: Create tables, suggest columns and data types, generate sample data, add/remove/rename columns through conversation, and create a canvas app from the table.
Copilot CANNOT: Create relationships between tables automatically (you can ask it to change a column to a lookup, but complex relationship chains require manual setup). It also cannot create business rules, security roles, or virtual tables through conversation.
Think of Copilot as an accelerator for the initial table design. Fine-tuning still happens in the table designer.
Modifying tables through conversation
You can also modify existing tables using Copilot. From the table editor in Power Apps:
- “Add a column called Priority with values High, Medium, and Low” — creates a Choice column
- “Remove the Fax Number column” — deletes the column (if it is a custom column)
- “Rename Listing Price to Asking Price” — updates the display name
- “Change Square Footage to a decimal number” — updates the data type
Each change is previewed before it is applied, so you always review before committing.
Importing data into Dataverse
Getting data INTO Dataverse is a common task. The exam expects you to know these methods:
| Import Method | Best For | How It Works |
|---|---|---|
| Excel import | Small datasets, one-time loads | Upload an Excel file directly from the table view. Dataverse maps columns automatically. |
| CSV import | Flat data files | Upload a CSV file. Similar to Excel import but for comma-separated data. |
| Dataflows | Recurring imports from multiple sources | Visual ETL (Extract, Transform, Load) tool. Connect to 30+ sources, transform data, schedule refreshes. |
| Power Automate | Event-driven or conditional imports | Build a flow that writes data to Dataverse when triggered (e.g., when a form is submitted). |
| Power Query Online | Complex data shaping before import | Used within dataflows. Merge, filter, rename, and transform data before it lands in Dataverse. |
| Azure Data Factory | Enterprise-scale data pipelines | For large volumes and complex orchestration. Connects to hundreds of data sources. |
Priya imports her marketing contacts
Priya at Spark & Co has a spreadsheet of 500 marketing contacts from a trade show. She opens the Contact table in Dataverse, clicks “Import data from Excel”, uploads the file, and maps the columns. Five minutes later, all 500 contacts are in Dataverse and ready for her Power Apps campaign tracker.
What are dataflows?
Dataflows are a Power Platform feature that lets you connect to external data sources, transform the data using Power Query, and load it into Dataverse on a schedule.
Think of it as a pipeline: data flows from the source (SQL database, SharePoint, web API) through transformations (filter rows, rename columns, merge tables) and into a Dataverse table.
Dataflows are ideal when you need to import data regularly — daily, weekly, or hourly — without manual effort.
Exporting data from Dataverse
Sometimes you need to get data OUT of Dataverse:
| Export Method | Best For | How It Works |
|---|---|---|
| Export to Excel | Quick data extraction for analysis | Click “Export to Excel” from any table view. Opens in Excel Online or desktop. |
| Export to CSV | Flat file export for external systems | Similar to Excel export but produces a CSV file. |
| Power Automate | Automated data distribution | Build a flow that reads Dataverse rows and sends them to another system, email, or file. |
| Azure Data Factory | Enterprise data pipelines | Pull data from Dataverse into Azure SQL, data lakes, or data warehouses. |
| Dataverse Web API | Programmatic access | Use REST API calls to read data from any Dataverse table. Available to developers and external applications. |
Exam tip: Import vs export
The exam may describe a scenario and ask you to pick the right method. Key decision points:
- One-time, small dataset: Excel or CSV import/export
- Recurring, scheduled: Dataflows (import) or Power Automate (either direction)
- Complex transformation needed: Dataflows with Power Query
- Enterprise scale, many sources: Azure Data Factory
- Event-driven: Power Automate (when something happens, move data)
🎬 Video walkthrough
Flashcards
Knowledge Check
Tom wants to quickly create a Dataverse table for property listings by describing what he needs in plain English. Which feature should he use?
Priya needs to import marketing contacts from a trade show spreadsheet into Dataverse. This is a one-time import. What is the simplest method?
Aisha needs to automatically import shipping data from an external SQL database into Dataverse every night. Which method is most appropriate?
Next up: Environments and the Power Platform security model — how Aisha controls who can access what.