TL;DR
FabCon Europe 2026 is underway in Barcelona from September 28 to October 1. The event brings Microsoft Fabric closer to Copilot, agents, and databases. Key announcements include Fabric IQ inside Microsoft Copilot, agentic app creation in Power BI, a Data Engineering Agent for migrations, and IQ Sharing in OneLake. Several of these capabilities are in preview, so it is worth confirming the status of each before planning around it. Organizations already on Fabric can begin preparing their semantic models and governance now.
Key Takeaways Fabric IQ business context now flows into Microsoft Copilot, and the integration is generally available. Power BI Desktop is getting agentic app creation, currently in preview for Pro, Premium Per User, and Fabric capacity customers. The new Data Engineering Agent in Fabric (preview) can plan and run migrations, ETL, and performance work under engineer-defined guardrails. IQ Sharing in OneLake (preview) lets teams share governed data and business context with partners and customers. A Database Hub, Hyperscale updates, and generally available vector search bring the database estate into the Fabric story. Migration, semantic model quality, and funding eligibility are the three planning items for the next quarter. Watch on YouTube
Inside FabCon 2026 and Kanerika’s Fabric Innovations
A walk through the FabCon 2026 floor with Kanerika, covering where Microsoft Fabric is heading and what the team is building on it.
Key Highlights From FabCon Europe 2026 Nearly 5,000 data professionals are in Barcelona this week for FabCon Europe and SQLCon. The program runs more than 130 sessions, with tutorials, community meetups, and roundtables on the side. Here are the key updates and highlights from the event.
Announcement Status What It Does Fabric IQ in Microsoft Copilot Generally available Brings business context and Power BI semantic model logic into Copilot Agentic app creation in Power BI Desktop Preview, coming in weeks Builds apps from a natural language description, using trusted semantic models Fabric Apps updates Available since the June 2026 launch Adds PostgreSQL support, TypeScript backend functions, and a Secret Store IQ Sharing in OneLake Preview Shares governed tables, files, agent instructions, and ontologies across organizations Data Engineering Agent Preview Plans and runs migrations, ETL, and performance tuning within guardrails Observability and operations agents Preview Centralizes workspace telemetry, alerts, and root cause analysis Database Hub in Fabric Preview Manages SQL Server, Azure SQL, PostgreSQL, and Cosmos DB from one place Vector search on Azure SQL and Fabric SQL Generally available Adds DiskANN vector indexes for semantic search and RAG Google BigQuery mirroring Generally available Mirrors BigQuery data into OneLake
Microsoft also shared scale figures. Fabric has more than 40,000 customers, and Power BI has more than 425,000. Levi Strauss, H&M, and Thomson Reuters were among the featured customers.
1. Fabric IQ Now Feeds Microsoft Copilot The headline item is the integration between Fabric IQ and Microsoft Copilot, now generally available. Business definitions, metrics, and organizational knowledge from trusted Power BI semantic models flow into Copilot. Microsoft says Copilot users get these Fabric insights without additional token costs.
Copilot answers are only as good as the definitions behind them. If two departments calculate margin differently, Copilot will pick one and sound confident about it.
What This Means for Your Team Semantic model quality becomes an AI issue, and no longer only a reporting issue. Clean measures, clear naming, and one owner per metric now decide how Copilot behaves for every user in the company.
A short audit of your top ten semantic models is a sensible first step. Look for duplicate measures, undocumented filters, and metrics with no named owner.
2. Power BI Learns to Build Apps Agentic app creation arrives in Power BI Desktop as a preview over the coming weeks. It is available to Power BI Pro, Premium Per User, and Fabric capacity customers, and Microsoft lists no extra cost for Pro and Premium Per User.
A user describes the app in plain language. The system builds it from a trusted semantic model, shows a live preview, and publishes it to Fabric. Apps can take input, write data back, and keep shared state. Each one gets its own SQL database in Fabric with authentication and security built in.
What This Means for Your Team Analysts will be able to create small operational tools, such as approval screens or forecast adjustments, without waiting for a development sprint. That speed is useful, and it also creates sprawl if nobody sets rules.
Decide early who can publish, where apps live, and when an analyst’s app must move to IT. Microsoft says apps can be handed to developers without a rebuild, which helps once the governance path exists.
3. Fabric Apps Add Developer Tooling Fabric Apps launched in June 2026, and Microsoft says thousands of developers already use it through the open-source Rayfin SDK and CLI. This week’s additions include direct connections to warehouses, SQL databases, lakehouses, and semantic models without copying data.
Developers also get TypeScript backend functions, a Secret Store, PostgreSQL support, private-by-default access, and built-in usage and performance metrics. Together these make Fabric a place to run applications as well as analytics.
4. IQ Sharing and the Wider OneLake Reach IQ Sharing in OneLake lets organizations share governed data and business context with other teams, customers, and partners. The preview supports governed tables, files, Markdown agent instructions, and RDF ontologies. Native support for Fabric IQ ontologies and Power BI semantic models is on the roadmap. Early adopters include EY, S&P Global, TomTom, Financial Fabric, and xMentium.
OneLake also connects to more sources. Bidirectional integration with Salesforce Data Cloud 360 is in preview, and Google BigQuery mirroring is generally available. Microsoft also named Atlan, dbt, and ClickHouse as partners, and announced deeper integration with Dynamics 365 Business Central.
What This Means for Your Team Sharing context along with data is a new idea for most companies. A supplier or customer can receive the tables and the definitions that explain them.
That raises the bar on documentation and access control. Teams that plan to share data externally should classify sensitive columns, set retention rules, and confirm ownership before turning any sharing on.
Listen on Spotify
Why Most Fabric Deployments Fail at Scale
5. The Data Engineering Agent Targets Migrations For anyone planning a move to Fabric, this is the item to watch. The Data Engineering Agent is in preview and draws on technology from Microsoft’s January 2026 acquisition of Osmos. It handles platform migrations, lakehouse modernization, performance tuning, complex ETL, and data preparation.
Engineers define the outcome and set guardrails. The agent plans the work, runs it, checks the results, and adjusts. Tasks inherit workspace permissions, and authorized teammates can follow progress and answer questions. Microsoft says it works across Fabric, GitHub Copilot, Visual Studio Code, Codex, and Claude Code.
What This Means for Your Team Autonomous migration sounds simple, but real migrations rarely fail on code conversion. They fail on undocumented business logic, hidden dependencies between jobs, and reports that nobody remembered to list.
An agent works from the inputs it receives. A complete inventory of sources, pipelines, and consumers still has to exist first, and someone has to define what a correct result looks like. Treat the agent as a fast pair of hands that needs a clear brief.
“Enterprises want partners who move them onto modern data platforms faster, with less risk. FLIP automates the hardest parts of a migration, from assessment through to execution, so our clients reach Power BI or Microsoft Fabric in weeks instead of quarters.”
Amit Chandak, Chief Analytics Officer, Kanerika
6. Real-Time Intelligence, Observability, and Billing Real-Time Intelligence gets new Copilot help. Users can create real-time dashboards from a text description, and Copilot assists with eventhouse tables, policies, and schemas. An operations agent adds root cause analysis and pre-approved actions in preview.
Observability is also in preview. Workspace monitoring can gather telemetry from many workspaces into one eventhouse, the monitor hub shows deeper job and capacity detail, and Activator sends alerts for job failures and successes. A new capacity view helps track consumption.
On cost, Microsoft is expanding on-demand billing. Teams pay only for the compute they use, and resources scale up and down with demand. A zero-provisioned option lets a team start without an upfront capacity commitment.
What This Means for Your Team Consumption billing lowers the barrier to a pilot. It also makes cost monitoring a daily habit, since a runaway job now appears directly on the bill.
7. Databases Join the Fabric Story SQLCon shared the stage with FabCon this year, and the database announcements were significant.
Database Hub in Fabric (preview) manages SQL Server, Azure SQL, PostgreSQL, Cosmos DB, and SQL database in Fabric from one view. Database agents for SQL and PostgreSQL on Azure are coming to preview, watching signals and helping with performance troubleshooting. Azure SQL Database Hyperscale gets serverless auto-pause (preview), elastic pools of up to 50 databases, and 192 vCore compute in general availability. Vector search with DiskANN indexes is generally available on Hyperscale, SQL Managed Instance, and SQL database in Fabric. SQL Server on Azure Local and its disconnected operations option are generally available, and Fabric is expanding to GCC High. The vector search item deserves attention. Teams building retrieval-augmented generation can now keep vectors inside the SQL database they already run, and skip a separate vector store.
What the Announcements Mean If You Use Fabric or Plan to Move The common thread is context. Microsoft is betting that agents need governed data, shared definitions, and clear ownership before they can act safely. The table below turns that into a starting checklist for three common situations.
Your Situation What to Do Next Why It Matters Now Already running on Fabric Audit semantic models, name metric owners, review capacity and billing settings Copilot and agents will use your definitions as they are today Planning a move to Fabric Inventory sources and reports, pick a first workload, check funding eligibility The migration agent and on-demand billing lower the cost of starting Still on Informatica, SSIS, Alteryx, or legacy BI Map each tool to a Fabric target and rank workloads by risk The gap between legacy stacks and Fabric AI features keeps widening
Check the Status Before You Commit Many of this week’s items are previews. Preview features can change, and they often lack the service commitments that production workloads need.
Use them in pilots and sandbox workspaces. Move to production once the feature reaches general availability and your own testing supports it.
How Kanerika Helps Teams Move to Fabric As a Microsoft Solutions Partner for Data and AI, the company holds an Analytics Specialization. It also holds the Advanced Specialization for Data Warehouse Migration to Microsoft Azure, earned in December 2025, and is a Microsoft Fabric Featured Partner. At FabCon 2026 in March, the team debuted Karl and the Azure to Fabric Migration Accelerator as native Fabric workloads . Everest Group also placed Kanerika among the Major Contenders in its Microsoft Azure Services PEAK Matrix Assessment 2026.
The team designs each engagement around the workloads a client runs today. The steps are usually the same. Assess the current estate, design the target on Fabric, migrate in waves, govern the result, and train the people who use it.
Case Studies From Real Fabric Projects Two Microsoft Fabric engagements show what that looks like in practice.
FoodPharma had six operating systems, including NetSuite, RedZone, and UpKeep, feeding separate reports. The Kanerika team unified them on Microsoft Fabric, consolidating more than 50 tables and roughly 1 TB of history in a seven-week implementation. Cross-functional reporting dropped from two business days to 90 minutes, and the BI team recovered about 15 hours a week. Microsoft published the result as a customer story on FoodPharma and Microsoft Fabric .
Southern States TOYOTAlift (SSMH) had data split across SQL Server and SharePoint, with inconsistent KPIs and little real-time visibility. The team built a lakehouse that joined the sources, cleaned and validated the data, and set up role-specific reporting. The case study reports an 85% gain in operational visibility and 90% better data accuracy. Delano Gordon, CIO of SSMH, has credited Kanerika’s flexibility in aligning Microsoft Fabric with the business. The full story is in the SSMH Fabric and Power BI case study .
Case Study
Southern States Material Handling Unifies Its Data on Microsoft Fabric
SQL Server and SharePoint data moved into one lakehouse, with an 85% gain in operational visibility and 90% better data accuracy.
Read the Case Study →
Migration Accelerators for Fabric The FLIP platform from Kanerika automates migrations from legacy tools into Fabric and Power BI. It is also listed on Azure Marketplace. For Azure to Fabric moves, the accelerator announcement cites 80% faster timelines, 50% lower costs, and 65% fewer resources. The current paths into Fabric and Power BI include the following.
On-Demand Webinar
Migrate to Microsoft Fabric 5X Faster With FLIP
See how automated assessment and conversion shorten the path from legacy tools to Fabric, and what the team checks before each wave.
Watch the Webinar →
The Data Engineering Agent and FLIP address the same pain from different angles. Whatever tooling runs the conversion, the assessment, dependency mapping, and validation around it decide whether a migration lands cleanly. That surrounding work is where delivery experience counts.
Microsoft Funding Can Lower the Cost of Starting Microsoft offers partner-led funding programs, including ECIF, Azure Migrate, and Azure Innovate, that can offset part of the cost of qualifying migration and build projects. Eligibility depends on the customer, the workload, and Microsoft’s current program terms.
The team helps clients check eligibility and prepare the paperwork before a project starts. The Microsoft funding support datasheet explains how that works.
Watch on YouTube
Reduce Your Fabric Migration Costs With Microsoft Funding
A Migration Office Hours session on which Microsoft programs can offset migration cost and how to prepare before a project starts.
Wrapping Up FabCon Europe 2026 shows Microsoft tying Fabric to Copilot, agents, and databases. The general availability items are ready to plan around, and the previews are worth testing in a sandbox.
The common need is a clean, governed data foundation. Teams that fix semantic models, document ownership, and choose a migration path now will be ready when the previews mature.
Talk to Kanerika
Planning a Microsoft Fabric Migration?
Bring your current tools and workloads. The team will map them to a Fabric target and check which funding programs may apply.
Schedule a Demo →
Frequently Asked Questions
What Were the Biggest Announcements at FabCon Europe 2026? Microsoft announced Fabric IQ inside Microsoft Copilot, agentic app creation in Power BI Desktop, a Data Engineering Agent for migrations, and IQ Sharing in OneLake. It also introduced Database Hub in Fabric, observability tools, on-demand billing, and generally available vector search on Azure SQL. Several items are in preview, so check each one’s status.
When and Where Was FabCon Europe 2026 Held? FabCon Europe 2026 is taking place in Barcelona from September 28 to October 1, alongside SQLCon. Microsoft reports nearly 5,000 data professionals and more than 130 sessions. The program includes tutorials, community meetups, and roundtable discussions for Fabric, Power BI, and SQL practitioners, plus an exhibit floor where partners show their Fabric work.
What Is Fabric IQ and Why Does It Matter for Copilot? Fabric IQ holds business context such as definitions, metrics, and ontologies for data in Microsoft Fabric. Its integration with Microsoft Copilot is now generally available. Copilot can ground answers in trusted Power BI semantic models, which makes clean measures and clear metric ownership more important than before. Microsoft says Copilot users get these insights without extra token costs.
Is the Data Engineering Agent in Fabric Ready for Production? The Data Engineering Agent is in preview, and Microsoft has not set a general availability date in its announcement. It can plan and run migrations, ETL, and performance work under engineer-defined guardrails. Most teams should test it in a sandbox workspace first and keep human review on every result. Production use should wait for general availability.
Which Announced Features Are Generally Available Today? Generally available items include the Fabric IQ integration with Microsoft Copilot, Google BigQuery mirroring in OneLake, DiskANN vector search on Azure SQL Hyperscale, SQL Managed Instance, and SQL database in Fabric, 192 vCore Hyperscale compute, and SQL Server on Azure Local. Agentic app creation, IQ Sharing, and Database Hub remain in preview.
How Should Teams Prepare Semantic Models for Copilot and Agents? Start with the ten most used semantic models. Remove duplicate measures, document filters, and name one owner for each metric. Consistent definitions matter because Copilot and agents draw on them directly. A short review before turning on new AI features prevents conflicting answers across departments. Sensible owners are usually finance, sales, or operations leads.
Does Fabric Now Support Pay-As-You-Go Billing? Microsoft is expanding on-demand billing in Fabric, where teams pay only for compute they use and resources scale with demand. A zero-provisioned option lets teams start without an upfront capacity commitment. This lowers the cost of pilots, though usage monitoring becomes more important because heavy jobs appear directly on the bill.
How Can Kanerika Help With a Migration to Microsoft Fabric? Kanerika assesses the current estate, designs the Fabric target, migrates in waves, and sets up governance and training. Its FLIP platform automates migrations from tools such as Informatica, SSIS, Alteryx, Azure Data Factory, Tableau, and Cognos. FoodPharma and Southern States TOYOTAlift are two Fabric engagements completed by the team. The team also supports Power BI moves from legacy BI tools.
Is Microsoft Funding Available for Fabric Migration Projects? Microsoft runs partner-led programs, including ECIF, Azure Migrate, and Azure Innovate, that can offset part of the cost of qualifying projects. Eligibility depends on the customer, workload, and current program terms. Kanerika helps clients check eligibility and prepare the application before a project begins. The datasheet on the Kanerika site explains the process.
Where Can I Meet the Kanerika Team at FabCon Europe? The Kanerika team is exhibiting at Booth #38 in Barcelona through October 1. Visitors can discuss Fabric migrations, semantic model reviews, and Microsoft funding eligibility in person. Teams not attending can start a conversation through the Kanerika migration services page. Bring a migration question, a model you are unsure about, or a funding query.
