AI Sales Engagement, Part 2: Building a Copilot Workflow with Dynamics 365 Sales
- Ryan Redmond 
- Jul 9, 2024
- 5 min read
Updated: Oct 14

This article is the second in our AI Sales Engagement series exploring how AI is transforming modern sales engagement.
In Part 1, we introduced the concept of AI Sales Engagement and showed how Microsoft Copilot Studio, Power Automate, and Dynamics 365 Sales form the foundation for a smarter, connected sales process.
In Part 2, we’ll bring those ideas to life through a practical example. We will build a Copilot-powered workflow that captures leads, automates follow-up, and feeds those insights directly into Dynamics 365 Sales.
Did you miss Part 1? AI Sales Engagement, Part 1: Transforming the Modern Sales Process.
Streamlining Sales Processes
For this demonstration, we’ll use a fictitious company named Fabrikam Ironworks, a B2C manufacturer of custom iron doors, gates, and railings. Their products come in a variety of styles including Classic, Traditional, and Modern, each made to order.
Fabrikam connects with customers through both online sales and in-home consultations. They want to deliver an on-demand, personalized experience for online shoppers that mirrors the quality of their in-person service and showcases how companies are evolving through modern sales engagement systems.
Their field sales teams need support too. These reps spend the day visiting customers’ homes and often lack the right brochures or materials to close the sale.
The solution is to build a custom Copilot Chatbot, accessible online, that can hold natural conversations with users, understand their requests, capture actionable data, and automatically send a personalized email with the correct product brochure attached.
It may sound complex, but with Microsoft Copilot Studio, the process is surprisingly straightforward.
We’ll call this solution the “Lead Catcher.”
This interactive experience provides users with immediate value, a personalized email and product-specific brochure, while also collecting contact details and buying-intent information that feed directly into Dynamics 365 Sales.
Lead Catcher: Work Breakdown
To build the Lead Catcher, let’s break the work into a few clear, repeatable steps:
1. Copilot Studio: Create a new Copilot
2. Topics: Design a friendly, guided conversation path
3. Power Automate: Trigger automated actions and workflows
4. Engage Customer: Confirm next steps and close out the conversation
Here’s how those steps come together inside Copilot Studio and Dynamics 365 Sales to create a connected experience:

Each block in this diagram represents a Topic within Copilot Studio—an individual part of the conversation where user input is captured and the Copilot performs an action or delivers a response.
For our flow, they are:
- Greeting: Initial message welcoming new user and guiding them into the conversation. 
- Product Interest: User selects which product type, style and size they are interested in. 
- Buying Intent: User indicates if they are looking to purchase now or just browsing. This helps us identify follow-up activities. 
- CRM Workflow: Once we have all of the inputs, this Topic hands the ball over to Power Automate. 
- Confirmation: Once all actions are complete, confirm that the conversation is finished and thank the customer. 
Now with the planning and work breakdown complete and once you have appropriate software license, create a new Copilot by navigating to https://copilotstudio.microsoft.com/ and login.
Once logged in, you simply click “+ New copilot” as shown below:

Following the work breakdown, create the topics to match the conversation flow from the block diagram above.
Note that Topics can be triggered to start a new conversation, or they can be chained together in a modular fashion.
In our example, each topic flows from one topic to the next.
One Topics redirects to another to form the entire conversation.
Each block in this diagram represents a Topic within Copilot Studio—an individual part of the conversation where user input is captured and the Copilot performs an action or delivers a response.

In addition to Topics, Entities play a key role in Copilot Studio. Entities are variables that can store parts of the conversation for later use.
In this example, we’ll use Entities to capture the customer’s preferred Product Type (Door, Gate, Railing) and Product Style (Modern, Traditional, Classical). These values are then passed to Power Automate and Dynamics 365 Sales. The screenshot below shows the ProductStyle entity definition.

The screenshot below shows how this Entity is used inside of the Product Selection topic to capture the user’s input.
Note: Using generative AI, this method accurately captures input whether the user types a single word (“Door”) or a full phrase (“I’d like to purchase a door”).

Next, define conditional logic based on user input. The screenshot below shows how branching rules guide the Copilot to the right follow-up Topic for each product type, ensuring a tailored conversation.

Once the Copilot has guided the user through the conversation and captured key details, it’s time to connect to Power Automate. This step might look advanced, but it’s built directly into Copilot Studio.
Within the CRM Workflow Topic, add a new node, choose “Call an Action,” and then click “Create a flow.”

This will automatically launch Power Automate and populate the Start and End nodes within Power Automate. By clicking on the first node in the flow, you can see the collection of variables that are passed into Power Automate from Copilot Studio as shown below.

Inside Power Automate, we’ll use the Microsoft Dataverse connector to connect directly to Dynamics 365 Sales. Earlier, our Copilot collected the user’s full name, but Dynamics stores first and last names separately. The screenshot below shows calculation nodes used to split and clean this data before creating the Lead record.

Power Automate then creates a Dynamics 365 Sales Lead using the chatbot’s captured data. This lead includes richer context than a traditional web form - making it far more actionable for the sales team.
Power Automate also uses the chatbot data to select the correct email template - matching the chosen product and style - from Dynamics 365 Sales, and then instantly sends a personalized email with the appropriate brochure attached.

The screen above shows the newly created Dynamics 365 Lead “Mike Johnson” along with the Personalized email and attachment.
Conclusion on Why Copilot Studio + Dynamics 365 Sales is Better Together
When Copilot Studio is paired with Power Automate and Dynamics 365 Sales, the result is a unified, AI-driven workflow that transforms how teams engage customers.
Generative AI handles natural conversation and data capture, Power Automate manages real-time orchestration, and Dynamics 365 Sales turns those insights into actionable leads.
This combination streamlines processes that once required multiple disconnected tools, freeing sales teams to focus on relationship-building and closing deals—exactly what AI Sales Engagement is meant to achieve.
Together, these Microsoft technologies deliver measurable value by reducing manual effort, improving data quality, and providing customers with faster, more personalized interactions.
At Optrua, we help organizations design, implement, and optimize solutions that connect AI and CRM systems in practical, revenue-driving ways.
Whether you’re exploring Copilot Studio for the first time or expanding your Dynamics 365 Sales environment, our team can help you build a smarter, scalable system that accelerates growth through the Optrua Care Plan.
What will you build with Copilot Studio?
Ready to take the next step? Contact Optrua to learn how we can help you implement AI-powered sales engagement in your business.
About the Author

Ryan Redmond is the founder of Optrua, specializing in CRM and business process optimization. Ryan channeled his passion for efficiency from lessons learned in the Navy to his work today.
He helps businesses streamline technology to improve employee and customer experiences and empower teams to work smarter, not harder, without unnecessary overhead.
Connect with Ryan on LinkedIn.





