web
You’re offline. This is a read only version of the page.
close
Skip to main content

Announcements

No record found.

News and Announcements icon
Community site session details

Community site session details

Session Id :
Small and medium business | Business Central, N...
Suggested Answer

How do organizations use AI, automation & integrations in Dynamics 365 Business Central to scale?

(6) ShareShare
ReportReport
Posted on by 20
I’m exploring how modern organizations are using Microsoft Dynamics 365 Business Central to drive scalable and resilient finance and supply chain operations.
Specifically, I’d love insights from those with real-world experience on:
 
  • How are you leveraging AI-driven automation (e.g., Copilot, workflows) to streamline finance or operations?
  • What role do embedded analytics and real-time reporting play in decision-making?
  • Which integration strategies (e.g., CRM, eCommerce, warehouse systems) have worked best without creating upgrade or maintenance challenges?
  • What are your best practices for maintaining performance, governance, and long-term upgradeability?
Also, if you’ve faced challenges like technical debt, slow performance, or complex customizations—how did you overcome them?
Looking forward to learning from the community’s experiences and recommendations!
I have the same question (0)
  • Suggested answer
    Nimsara Jayathilaka. Profile Picture
    5,100 Super User 2026 Season 1 on at
    Hi
     
    To drive scalable operations in Business Central without creating technical debt, organizations leverage Copilot for automated bank reconciliations and inventory forecasting alongside Power Automate for robust approval workflows. Real-time decision-making is powered by embedding Power BI dashboards directly into Role Centers, while seamless integrations with CRM and eCommerce are best achieved using standard APIs, Dataverse, and certified AppSource connectors like the native Shopify integration to avoid upgrade challenges. The ultimate best practice for maintaining performance, governance, and seamless automatic wave updates is to completely avoid modifying base application code by relying strictly on AL extension events and custom codeunits, while actively monitoring system health and troubleshooting slow performance using Azure Application Insights telemetry.
  • Suggested answer
    OussamaSabbouh Profile Picture
    15,536 Super User 2026 Season 1 on at
    Hello,
    from my experience the scalable BC implementations are the ones that stay as close as possible to standard and use extensions/integrations only where they add clear value: use Copilot/AI for focused tasks like bank reconciliation, payables/e-document matching, product text, and exception handling rather than trying to “AI everything”; use Power BI and Analysis Mode for operational visibility, with Power BI for cross-company/cross-system dashboards and BC analysis pages for live day-to-day checks; for integrations, prefer standard connectors first like Dataverse/Dynamics 365 Sales, Shopify, Power Automate, APIs/webhooks/business events, and avoid file drops or direct table-style coupling where possible; for performance and upgradeability, keep custom AL small, event-based, and well tested, avoid heavy logic on pages/reports/web services, use API pages instead of exposing UI pages, monitor telemetry, remove unused extensions, clean orphaned extension data, and review customizations before every major update. Most technical debt I’ve seen comes from over-customizing posting, warehouse, pricing, or reporting logic; the fix is usually to move logic into clean extensions, replace custom reports with Power BI where practical, archive/remove unused apps, and challenge every customization with “can standard BC + workflow/configuration do this now?”
    Regards,
    Oussama Sabbouh
  • Suggested answer
    Jainam M. Kothari Profile Picture
    16,237 Super User 2026 Season 1 on at
    Hello,
     
    Modern organizations are increasingly focusing on cloud-first, scalable architectures by leveraging AI-driven automation, embedded analytics, and low-customization integration strategies.

    Many companies are using Copilot, approval workflows, OCR invoice processing, and Power Automate to reduce manual finance and operational tasks, improve data accuracy, and accelerate month-end activities. Embedded analytics through Power BI and real-time dashboards play a critical role in management decision-making by providing live visibility into cash flow, inventory, sales, purchasing, and operational KPIs.
     
    For integrations, organizations are achieving better long-term upgradeability by using standard APIs, web services, Dataverse, and middleware platforms instead of direct database customizations, especially for CRM, eCommerce, WMS, and third-party logistics systems.
     
    Best practices for governance and performance include minimizing customizations, following standard business processes wherever possible, using extensions instead of base code modifications, archiving historical data, monitoring telemetry, and implementing role-based security and approval controls.
     
    Common challenges such as technical debt, slow performance, and complex legacy customizations are typically resolved through periodic customization reviews, moving heavy processes to background jobs, adopting event-driven development, eliminating unused extensions, and aligning implementations with Microsoft’s standard roadmap to ensure smoother upgrades and long-term system stability.
  • Suggested answer
    Gregory Mavrogeorgis Profile Picture
    1,339 Super User 2026 Season 1 on at
    Hi,

    From the projects I have seen lately, BC is moving fast, but the value is not always where the marketing slides put it. A few honest notes.

    On AI and Copilot, the useful parts so far are the small things, bank reconciliation suggestions, sales line descriptions, the side-pane to ask questions on data. The big "agent" promises are still early. We use it mainly to save time on data entry and drafting, not for decisions. For workflows, standard approvals do the job, and Power Automate covers the rest. The trap is huge Power Automate chains that hit BC every minute and then everyone complains about performance. Better to keep the heavy logic inside BC with AL events.

    On reporting, real-time works when the data is clean. Standard analytics views, Power BI on BC APIs, and the new analysis mode on lists cover most cases. Finance still ends up in Excel half the time, this is why Jet Reports is still very common.

    For integrations, the rule is, always APIs and webhooks, never direct table reads. CRM through Dataverse, Shopify or Sana for eCommerce, Tasklet for handheld. Custom middleware that writes directly into BC tables is technical debt waiting to explode at the next upgrade.

    On performance and upgradeability, boring advice but works. Customisations as small extensions, event-based, no objects in the wrong range, telemetry on, sandbox refreshed often. Most slow performance is bad SQL or a loop that should be a filter, the Performance Profiler shows it in five minutes.

    On technical debt, usually it comes from old C/AL converted to AL and never cleaned up. The fix is to split the monolith into smaller extensions, replace table modifications with event subscribers, and retire legacy code slowly. Not glamorous, but every upgrade after that is painless.

    Short story, BC is solid if you treat it with respect. Stay close to standard, integrate through APIs, keep customisations small and event-based, and use AI where it actually saves time.

     Tick the checkbox below to mark the answer as verified, if it helped resolve your question.
     
    Regards
    Gregory Mavrogeorgis
     
  • Suggested answer
    Aman Kakkar Profile Picture
    3,465 Super User 2026 Season 1 on at
    Hi,
     
    If you search for "Copilot & agent capabilities", you will see a list of AI features that Microsoft has already released in Business Central itself, that does not require any external integrations. Some of them are billed but majority of them can be used if you have a Business Central license.
     
     
     
     
     
    For creating your Custom AI Agents, you can watch this video - How to create agents in Business Central
     
    For integrations with any LLM models, you need to deploy the LLM Model in the AI Foundry in Azure and then can integrate the LLM model via API calls. For more info - How to Set Up Azure OpenAI: A Step-by-Step Guide
     
    Do mark as verified if this helps.
    Aman K
  • Suggested answer
    YUN ZHU Profile Picture
    100,552 Super User 2026 Season 1 on at
    Hi, hope the following helps.
    Dynamics 365 Business Central: All Copilot and agent capabilities
     
    Thanks.
    ZHU
  • Gerardo Rentería García Profile Picture
    26,483 Most Valuable Professional on at
    Hi, good day
    I hope this can help you, and give you some hints.

Under review

Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.

Helpful resources

Quick Links

Introducing the 2026 Season 1 community Super Users

Congratulations to our 2026 Super Stars!

Meet the Microsoft Dynamics 365 Contact Center Champions

We are thrilled to have these Champions in our Community!

Congratulations to the April Top 10 Community Leaders

These are the community rock stars!

Leaderboard > Small and medium business | Business Central, NAV, RMS

#1
OussamaSabbouh Profile Picture

OussamaSabbouh 2,184 Super User 2026 Season 1

#2
YUN ZHU Profile Picture

YUN ZHU 1,148 Super User 2026 Season 1

#3
AndrewThomas81 Profile Picture

AndrewThomas81 677

Last 30 days Overall leaderboard

Featured topics

Product updates

Dynamics 365 release plans