Growing businesses are at a unique crossroads with their data. Leadership wants faster answers, consolidated views across entities, and a credible path to AI. Meanwhile, the finance and IT teams who deliver those answers already have full plates. The data needed to make better decisions almost always exists. Getting it out of the ERP in a usable, timely form is the hard part.
Dynamics 365 Business Central customers have traditionally filled that gap with workarounds. Teams built reports on API pulls that compete with day-to-day transactions. In addition, nightly exports landed in spreadsheets. Similarly, open-source connectors fell to whoever on the team understood them best. Traditional ERP reporting was never designed to feed a modern analytics platform, so every new question tended to become a small integration project.
That is changing. At FabCon Europe in Barcelona at the end of September, Microsoft announced native data mirroring from Business Central into Microsoft Fabric. It arrives as part of Business Central 2026 release wave 2. Instead of building and babysitting pipelines, companies choose the Business Central companies and tables they care about. Mirroring then replicates that data into OneLake, Fabric's unified data lake, within minutes of each posting. Microsoft's roadmap lists the public preview for November 2026, with general availability targeted for April 2027.
In this blog, we'll explore the data challenges growing businesses face today and how Business Central mirroring in Microsoft Fabric addresses each one. We'll walk through practical scenarios that show how companies can put it to work. Finally, we'll outline a phased approach for getting ready while the feature moves from preview to general availability.
The Challenges Growing Businesses Face With ERP Data
When we talk with finance leaders and IT directors about reporting, three problems surface again and again.
1. Fragile, Hand-Built Data Pipelines
Most Business Central reporting outside the ERP depends on custom extractions: OData feeds, API calls, scheduled exports, or community tools that someone configured years ago. Each one works until a field changes, someone adds an extension, or the person who built it moves on. Then reports quietly break and nobody notices until a number looks wrong in a board meeting.
These pipelines also carry a performance cost. API-based extraction runs against the same environment your users are posting transactions in, and it counts against service limits. As data volumes grow, the reporting workload starts to compete with the business it should support.
2. Multi-Company Reporting That Lives in Spreadsheets
Many mid-sized organizations run several legal entities in Business Central. Consolidated views of revenue, margin, inventory, or cash usually mean exporting from each company and stitching the results together in Excel. The process is slow, hard to audit, and different every time a different analyst does it.
Legacy approaches treat each company as its own island. When leadership asks for a cross-entity view, the answer takes days rather than minutes.
3. An AI Strategy Without a Data Foundation
Executives are under pressure to show progress with AI, but AI is only as useful as the data underneath it. When ERP data sits in scattered exports with only partial connections to CRM, eCommerce, or operational systems, nothing provides a trustworthy foundation. Forecasting, anomaly detection, and Copilot experiences have no single source to build on.
Traditional ERP reporting answers yesterday's question about one system. It was never designed to serve as the governed, organization-wide data layer that modern analytics and AI require.
The Role of Business Central Mirroring in Microsoft Fabric in Overcoming These Challenges
Business Central mirroring, powered by Microsoft Fabric and OneLake, offers a Microsoft-supported way to make ERP data available for analytics without the usual plumbing. Here's how.
1. Automation Without Heavy Overhead
Setup runs through an assisted setup guide inside Business Central. First, an administrator installs the Microsoft Fabric Mirroring extension and registers an app in Microsoft Entra ID. Next, they create a mirrored database in a Fabric workspace. The guide then connects Business Central to that database. Finally, the administrator selects which companies and tables to replicate. From there, the service keeps OneLake current on its own. An initial historical load comes first, followed by incremental updates that typically arrive within a few minutes.
Mirroring uses a Microsoft-managed replication process instead of the public Business Central APIs. As a result, it does not draw down your API limits, and early testers report no noticeable effect on production performance. It covers both standard tables and tables added by extensions, so customizations and ISV add-ons come along without separate integration work. The data lands in OneLake in Delta Parquet format, where you can query it with SQL or use it directly in Power BI.
- Replace scheduled API exports with continuous replication that a Business Central administrator manages from a single setup page.
- Bring extension and add-on tables into analytics without writing new connectors each time you install a solution.
- Assign dedicated permission sets so only designated administrators can configure or change the Fabric connection.
Example scenario: a professional services firm has relied on a community export tool for years. It can test native mirroring in a sandbox and compare the results side by side. Once the numbers reconcile, it can retire the custom pipeline. Fewer moving parts means lower maintenance cost and fewer surprises at month end.
2. Faster, Consolidated Decision-Making
Mirroring works across companies, which makes multi-entity reporting a configuration choice rather than a spreadsheet exercise. Microsoft is also rebuilding its nine standard Business Central Power BI apps to read from Fabric, offering both single-company and multi-company versions. The multi-company versions also add the ability to filter by company.
For finance leaders, that means a consolidated view of performance that refreshes throughout the day instead of at the end of a manual process. For operations leaders, it means inventory, sales, and purchasing data from every entity in one place.
- A distributor operating three regional entities mirrors item ledger entries and sales lines from each. Every morning, a single Power BI report compares fill rates and margin by region.
- A CFO reviews a consolidated G/L, customer ledger, and vendor ledger view during the close rather than waiting for each controller to submit an export.
- A manufacturer tracks production output and scrap across plants that run as separate Business Central companies. It spots a quality problem at one site before it reaches customers.
When everyone works from the same current numbers, decisions move faster. Meetings shift from debating whose spreadsheet is right to deciding what to do next.
3. A Governed Foundation for Analytics and AI
Once Business Central data sits in OneLake, you can combine it with data from other sources in the same Fabric environment. Microsoft also announced expanded OneLake connectivity, including Salesforce Data Cloud integration and Google BigQuery mirroring. Consequently, teams no longer have to analyze ERP data in isolation. Data engineering, data science, and Power BI all work against the same copy of the data.
That shared foundation is what makes AI practical. Forecasting models, anomaly detection, and natural-language reporting all depend on clean, current, connected data. Mirroring delivers the ERP portion of that foundation without a custom project.
- A wholesaler blends mirrored sales history with eCommerce order data to build a demand forecast that informs purchasing decisions in Business Central.
- A finance team flags unusual vendor payment patterns by running anomaly detection against mirrored vendor ledger entries.
- A sales leader joins Business Central invoicing data with CRM pipeline data to compare forecasted revenue against what actually shipped and billed.
A governed data layer turns AI from a pilot that never scales into a capability that improves decisions across the business.
The Phased Approach to Implementing Business Central Mirroring in Microsoft Fabric
At KTL Solutions, we help growing businesses adopt Business Central mirroring in Microsoft Fabric through a practical, phased approach. That way, they are ready to move quickly as the feature reaches general availability.
1. ERP Stabilization
We start by making sure your Business Central environment is on the 2026 release wave 2 update path and that your core data is clean. Mirroring replicates every field in the tables you select. Resolving duplicate customers, inconsistent dimensions, and outdated item records first pays off in every report downstream.
2. Data Governance
Next, we decide which companies and tables to mirror, who owns each dataset, and who can access it in Fabric. Most organizations begin with a focused set covering the general ledger, receivables, payables, inventory movements, and order documents, then expand from there.
3. System Integration
We then stand up a Fabric workspace in the same Microsoft Entra tenant as Business Central and size the capacity. Mirroring adds no charge on the Business Central side, so the budget question is Fabric capacity. We typically pilot on a Fabric trial capacity. The new zero-provisioned F0 option announced at FabCon may also fit, once Microsoft supports it for mirroring. Then we move to F2, the smallest fixed capacity suited to production mirroring. Where you already rely on custom exports or community connectors, we run both in parallel in a sandbox, reconcile the results, and plan a clean cutover.
4. AI Enablement
Finally, we build on the mirrored data with consolidated Power BI reporting and blended datasets from CRM and eCommerce. We also add targeted AI scenarios such as forecasting and anomaly detection. With a reliable foundation in place, these capabilities deliver measurable value rather than one-off demos.
Learn More About Fabric Mirroring from Microsoft
Business Central mirroring runs on Microsoft Fabric's open mirroring capability. Microsoft's Fabric documentation is the best place to learn how the underlying technology works. Microsoft has also published prerelease Business Central setup articles, which it will update as the preview rolls out. In the meantime, these official resources are a good starting point for IT and data teams:
- Microsoft's Business Central documentation for the feature, covering requirements, setup, table selection, and monitoring.
- Microsoft's overview of how mirroring keeps a continuously updated, analytics-ready copy of operational data in OneLake, the different mirroring approaches, and what gets created in your workspace.
- A walkthrough of creating an open mirrored database, the same type of item Business Central connects to during setup.
- How to start a free Fabric trial, the simplest way to explore mirroring before committing to paid capacity.
- The official roadmap entry (ID 574126), which tracks the preview and general availability dates as Microsoft updates them.
Conclusion
Growing businesses need faster, consolidated insight and a credible data foundation for AI, and they cannot afford to keep maintaining fragile pipelines to get there. Traditional ERP reporting, built on exports and API pulls, was never designed to carry that load.
Business Central mirroring in Microsoft Fabric gives organizations a Microsoft-supported path from transactional data to organization-wide analytics. Moreover, it adds no Business Central cost and requires far less maintenance. With the public preview arriving this fall, now is the right time to clean up your data, plan your scope, and pilot the integration. KTL Solutions can help you do all three. Because better decisions start with data you can actually reach.
Contact us:
Ready to migrate to Dynamics 365 Business Central? .