How to Use Copilot Natural Language Search in Dynamics 365

By Carrie Marshall | July 6, 2026

In Part 1 of this blog series, we explored what Copilot Natural Language Search is and how it works. In this follow-up blog and video below, we'll focus on practical applications, how different teams within your organization can use Natural Language Search, and best practices for getting started.

How Natural Language Search Helps You Find Information Faster

Before exploring specific examples, it's important to understand how Natural Language Search simplifies the process of finding and working with data in Dynamics 365.

The shift to a better data retrieval system is simple: use Copilot and AI directly embedded into your Dynamics 365 systems to simply ask where data is, like you would ask a human. It’s a relatively simple process:

  1. Type your request or inquiry into the search bar like you would to a human who manages a certain dataset.
  2. Wait for Copilot to interpret your prompt and go into your system to find the data without knowing the field names or schema details.
  3. Take the results and apply filters to them, visualize them, or do whatever you want

Picture showing how Copilot processes prompts and gives a result

Essentially, it is a conversational layer on top of views and grids that are already in your system. Microsoft Copilot will interpret your sentence and build the grid filters for you. You can then see the criteria, edit them, or remove them. It’s not a silver bullet to solve all your data needs, but it is much faster and more efficient. Plus, it still requires human involvement, as you can still control filters and validate logic at any time, and traditional views still work.

Copilot can help streamline your day, while still giving you full control over your search criteria and results.

Natural Language Search in Practice: Finding Incomplete Records

One common use of Natural Language Search is identifying incomplete or missing data. In this example, we are working with the All Accounts view and notice that the “Address 1: City” field is empty for two records.

Screenshot showing natural language search prompt to find missing fields

We’ll use this as our example to show you this natural language search capability in action. Once we’re in this view, asking Copilot for the information is simple.

In the search bar near the top, type:

“Show me the accounts without a city.”

Copilot will search through the Accounts list and pull up the records that match the search. It will also give you the filter criteria, in this case:

“Address 1: City: Does not contain data.”

Screenshot showing natural language search prompt to find missing addresses

You can then amend, edit, or remove that criteria as you see fit.

Why You Can Trust the Results

While Natural Language Search can significantly speed up the process of finding information, users still maintain full visibility and control over the results. Human oversight remains important, and Copilot makes it easy to review, edit, and validate the filters it applies.

Rather than replacing traditional filtering, Copilot helps users build queries faster while keeping them in control.

Users can:

  • Gain clear insights into the logic: Filters are visible to you as tags and/or criteria.
  • Edit capabilities any time: You can open and edit filters for verification.
  • Cleanup in one-click: If you want to remove or clear all conditions, you can do so with one simple click.

Nothing you do will cause harmful system changes or problems, ensuring that natural language Copilot search keeps power users comfortable while allowing new users to move faster and more efficiently. This leads to faster and more effective AI adoption.

Pro Tip: Turn Frequently Used Searches into Reusable Views

One feature that a lot of people miss when using natural language search in Dynamics 365 is the ability to turn your search results into reusable views. Once Copilot produces the view for you, you can save it as a new view, then customize columns and share it like any other.

This reduces admin dependency because users can self-serve and see what is valuable without having to go to someone higher up.

Going back to our example, you can save the view easily by clicking on the name of the newly created view and selecting “Save as new view” from the drop-down menu.

Screenshot of Dynamics 365 and where you can select "save as view" to save prompt

On the next screen:

  • Give your view a name that works for you or your team
  • Enter a description if necessary
  • Hit Save

Screenshot showing field to save prompt in Dynamics 365

This feature is a huge time-saver and empowers users to save views that are highly relevant to their jobs.

Getting Started with Natural Language Search

Admins who have control of these settings can do this by following these steps:

  1. Go to the Power Platform Admin Center
  2. Go to the specific environment settings. In this case, Dynamics 365 Sales.
  3. In the environment drop-down menu, go to Product > Features.
  4. Then, find the Natural Language Grid and View Search. From there, you can
    • Enable it for all users by selecting “All Users”
    • Enable it for users as the feature becomes available. This auto-enables a feature when it becomes generally available.
    • Disable the feature for everyone by selecting “No One”

Diagram showing how setting options for natural language search

How to Roll It Out

In terms of rolling this feature, or any AI features, to your organization, there is a pretty simple but effective method:

  1. Pilot the features to some users: Start small with a group of highly experienced AI users who can work with a few common scenarios.
  2. Develop a prompt playbook: Give 8-10 example prompts for each role. This will also help you develop a library of prompts that people can use in the future.
  3. Keep Governance top of mind: Make sure you keep predefined views for controlled reporting and use Copilot for exploration.

Additionally, you may want to temporarily disable the feature as well. While this may sound counterproductive, the reasons you would want to do this could include:

  • User not knowing what to type or how to prompt effectively
  • Preference for structured access, which you will need to develop
  • Strict data governance for pre-defined views that need to be put in place
  • Waiting for maturity of the feature so it's more effective

At the end of the day, whatever you do will be what best aligns with your users, your system, and your environment.

Reach out to Stoneridge Software Today!

Whether your organization is completely new to AI, just starting to adopt it, or has already rolled it out to some or all of your team, we can help you at any stage.

Reach out to us today to optimize your Copilot experience!


Talk to a Stoneridge expert today CTA Button

Carrie Marshall
Our Verified Expert
Carrie Marshall
Read More from Carrie Marshall

Related Posts


Under the terms of this license, you are authorized to share and redistribute the content across various mediums, subject to adherence to the specified conditions: you must provide proper attribution to Stoneridge as the original creator in a manner that does not imply their endorsement of your use, the material is to be utilized solely for non-commercial purposes, and alterations, modifications, or derivative works based on the original material are strictly prohibited.

Responsibility rests with the licensee to ensure that their use of the material does not violate any other rights.

Start the Conversation

It’s our mission to help clients win. We’d love to talk to you about the right business solutions to help you achieve your goals.

Subscribe To Our Blog

Sign up to get periodic updates on the latest posts.

Thank you for subscribing!