how to automate the variance commentary in D365

Noruwa O. Izevbaye 0 Reputation points
2026-08-07T15:15:03.9566667+00:00

Our Variance commentary is produced manually by each Vertical which has a business impact of Inconsistent quality and timeliness of commentary; significant manual effort duplicated.

Microsoft 365 and Office | Development | Other
0 comments

2 answers

Sort by: Most helpful
Most helpful Newest Oldest
  1. Ruby Nguyen 820 Reputation points Independent Advisor
    2026-08-07T16:08:23.52+00:00

    Good day Noruwa, 

    Thank you for sharing your concern.

    Currently, the Dynamics 365 Finance provides functionality for variance analysis and comment capture, it doesn't natively deliver a fully automated end to end narrative generation process. To automate and standardize the workflow, I recommend leveraging Power Automate alongside Dynamics 365 Finance:

    Option 1: Centralized commentary tracking and approval 

    If the main goal is to improve ownership, visibility, and approval tracking, you can use a SharePoint or Microsoft List as a central repository and automate the process with Power Automate.  

    Commentary requests can be automatically assigned to the right business owners, notifications can be sent through Teams or Outlook, approvals can be managed within the workflow and reminders can be triggered for overdue items. This helps improve accountability, reduce manual follow up and provide clear visibility into submission and approval status. 

    Option 2: Automated commentary requests from Dynamics 365 Finance 

    If you want commentary requests to be generated directly from financial data, you can leverage Dynamics 365 Finance business events and Power Automate.  

    Based on predefined rules such as variance thresholds, specific accounts, or business units, the system can automatically create commentary tasks, route them to the appropriate owners, send notifications, and initiate approval workflows. This reduces manual effort, improves consistency and provides a scalable process across the organization. 

    Additional information is available in the following resources: 

    Create and test an approval workflow with Power Automate - Power Automate | Microsoft Learn 

    Application Connector - Finance & Operations | Dynamics 365 | Microsoft Learn 

    If you want to find more steps about Power Automate, I recommend posting your question in the Microsoft Power Platform Community Forum. In this forum, you'll get the most qualified group of respondents and other partners who read the forums regularly can share their knowledge or learn from your interaction.   

    I hope this helps clarify the available options and provides a starting point for designing a more efficient and standardized variance commentary process. Please let me know if you have any questions. 

    If the answer is helpful, please click "Yes" and kindly upvote it. If you have any extra questions about this answer, please click "Comment".            

    Note: Please follow the steps in the forum documentation to enable e-mail notifications if you want to receive the related email notification for this thread.  

    Was this answer helpful?

    0 comments

  2. AI answer

    2026-08-07T15:15:22.99+00:00

    To automate variance commentary in D365-related finance workflows, use the Variance analysis feature in Finance in Microsoft 365 Copilot and, for full automation, create an AI action based on a template.

    A practical path is:

    1. Prepare the workbook for variance analysis
      • Open an Excel workbook that contains the financial data.
      • Keep the pivot table in one worksheet and the source data in another worksheet in the same workbook.
      • In Excel, go to Home > Finance to open the Finance in Microsoft 365 Copilot sidecar.
    2. Generate variance insights and commentary
      • Use the Variance analysis feature to identify variances and analyze them.
      • Select Analyze for an individual variance or Analyze all for all listed variances.
      • Expand each variance result to view the generated insights.
    3. Standardize the commentary output
      • Use Parameters to guide the analysis toward the attributes that matter to each vertical, such as products, regions, departments, cost centers, or profit margins.
      • In Analysis options, set the summary length, tone, and type so the generated commentary is more consistent across teams.
      • This directly addresses inconsistent quality by applying the same analysis focus and output structure within a session.
    4. Allow controlled editing before publishing
      • The generated commentary can be edited in the sidecar.
      • Select Include in variance report > Edit analysis > Save.
      • This supports a governed process where Copilot drafts the commentary and finance users refine it only when needed.
    5. Publish commentary back into Excel
      • Use Publish as comment to write the variance summary as a native Excel cell comment on the first cell of the variance result.
      • This helps distribute commentary directly in the workbook used for review.
    6. Move from assisted to autonomous automation
      • Create a template first.
      • Then create a new AI action for Variance analysis.
      • In the AI action setup:
        • Go to the Finance Agent sidecar and select AI Actions.
        • Select + New AI Action > Variance analysis.
        • Enter a unique action name.
        • Select the template.
        • Set the Data source to the SharePoint folder for input files and output storage.
        • Choose a Trigger: Event or Schedule.
          • Event runs when a source file is added to the folder.
          • Schedule runs when the schedule criteria are met and the two source files are present.
        • Complete required recipient details and save.
    7. Use autonomous runs to reduce duplicated manual effort
      • AI actions can run variance analyses on behalf of the team and notify recipients when processing is complete.
      • For AI actions, pivot tables aren’t required in the source files being analyzed; the dataset only needs the flat raw table data.

    For the business problem described, the recommended operating model is:

    • Use templates to enforce a common variance-analysis approach.
    • Use parameters and analysis options to standardize commentary style.
    • Use AI actions with Event or Schedule triggers to automate recurring commentary generation.
    • Keep human review only for exceptions by editing the generated analysis before final reporting when necessary.
    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.