Dynamics 365 Business Central: documenting a business process in the AI era.

Like probably many of you, in the last months of the year I’m in the middle of some really huge go-lives and one of the tasks that usually you need to complete before a go-live of an ERP system is to document all the customized business processes.

This is usually a task delegated to application consultants and it’s usually time-consuming, so it’s sometimes overlooked (also because doing documentation is often deadly boring).
But things are changing with AI…

With the help of AI, you can create really great documentation for Dynamics365 Business Central processes in really few minutes (and also in a quite funny way). With this post I want to describe how I personally create documentation today for my customizations in Business Central. Probably it’s something that you also apply, but if not maybe this can be helpful also for you (this is something I also demoed during my vibe coding trainings in the last months).

Step 1: simulate your business process.

The first step is to simulate the business process you want to document in Business Central. For this task, there’s the marvellous Page Scripting tool that can help us. I use this tool to recreate the business process step by step like an end user.

In this example, with the Page Scripting tool I’ve registered a process for creating a purchase order, selecting a vendor, adding a line with a quantity, releasing the order, posting a receipt of a partial quantity.

At the end of the process, the tool gives me a step by step recording that you can save in YAML file:

This YAML file can also be shared with the customer (in order to see the recording of the process live in the system), but this is not our goal today…

Step 2: generate a process explanation document with AI.

As a second step, I open my AI tool (in this case I’m using GitHub Copilot) and I ask the LLM to analyze my YAML file and describe the process for an end user:

The AI model analyzes the YAML file and it produces a markdown document explaining the business process step by step:

This is the generated documentation file:

As you can see, this is a perfect document guideline explaining all the steps to perform the process for a basic user.

Step 3: generating a flow chart of the process.

One of the practice that I love to do is to add a diagram of the business process in the user manual.

To do that, I simply ask the AI model to generate a mermaid diagram accordingly to the YAML file:

and this is the result:

A flowchart of the process is created in my documentation folder and you can add it to your final documentation package:

Step 4: create a PDF of the business process documentation file for the customer.

The last step of my personal process is to generate a PDF file from the markdown file that describes my business process documentation. This will be the PDF file that I will send to the customer.

To do this, I again use AI:

To satisfy this task, the AI model prompts you to use tools like Pandoc. If you don’t have this tool installed on your machine, don’t worry… the AI will help you on this .

Final result: I have a new PDF file containing all the business process documentation that I can send to the end user:

And if I open the generated PDF file, it’s also nicely formatted:

Conclusion.

In conclusion, leveraging artificial intelligence to generate documentation for business processes represents a transformative shift in how organizations approach knowledge management and user support in general.

Key benefits of using this approach (at least for me) are the following:

• Speed and efficiency: AI can produce comprehensive documentation in minutes rather than hours or days, dramatically accelerating time-to-productivity.
• Consistency and Standardization: unlike human documentation efforts that may vary in quality and style, AI ensures uniform terminology, formatting, and depth across all materials, creating a cohesive user experience.
• Enhanced accuracy and errors reduction: AI minimizes the documentation errors that usually humans can do. Just give the AI model a good representation of the business process (YAML file recorded with Page Scripting tool) and results will be perfect.
• Cost optimization: organizations can significantly reduce documentation-related labor costs (time is precious).

Maybe you have a different approach on that, but if not I suggest to adopt this practice…

Original Post https://demiliani.com/2025/09/30/dynamics-365-business-central-documenting-a-business-process-in-the-ai-era/