Getting Started with Dataflow Gen2: Simple, Practical, Real

#fabcon2026 #session #learning #microsoft #fabcon2026-session

Description

Want to get started with Dataflow Gen2 without stress?
This session explains what DFG2 is, why it matters, and how it fits into Microsoft Fabric.
We’ll build a dataflow from scratch, connect multiple data sources and apply proven Power Query patterns to clean messy data, consolidate files, reuse logic with parameters, publish to a Lakehouse, and configure a simple, reliable refresh.

Key Takeaways

What Dataflow Gen2 is, where it fits in Fabric, and when to use it.
Append, Folder Connector, Parameters, Data Destination — reusable from Monday.
From messy source files → clean Lakehouse table, scheduled and repeatable.
The 5 most common mistakes — so you don't have to learn them the hard way.
Lakehouse / Warehouse
Why Fabric & DFG2 Now?

My Notes

Action Items

Resources & Links

Slides

📥 Download Slides

2026-03-06
Getting Started with
Dataflow Gen2:
Simple, Practical, Real
Cristian Angyal
About me
o Electrical Engineering background
o Former Project Manager (PMP®, PMI-ACP®)
o Microsoft Certified Professional
(DP-600, DP-700, PL-300, MCT, MCSA, MCSE, MOS Master)
Cristian Angyal
o Romania PBI and Modern Excel UG Founder
https://www.meetup.com/romaniapug
/cristian-angyal
https://linktr.ee/cristiangyal
o Microsoft MVP (Excel & Power BI)
o Husband (+25y) and father (+9y)
#NeverStopLearning
o Love to ski, travel and learn
What You'll Leave With
By the end of this session:
• Mental model
• What Dataflow Gen2 is, where it fits in Fabric, and when to use it.
• Working patterns
• Append, Folder Connector, Parameters, Data Destination — reusable from Monday.
• A full ETL flow
• From messy source files → clean Lakehouse table, scheduled and repeatable.
• Gotchas to avoid
• The 5 most common mistakes — so you don't have to learn them the hard way.
What is Dataflow Gen2?
"A cloud-hosted Power Query experience inside Microsoft Fabric that lets you ingest, transform, and
load data into Fabric items — without writing code "
.
Cloud-Hosted
Power Query
Fabric-Native
No Code
Runs in Fabric
no local PQ engine
Same editor as
Power BI Desktop
Direct output to
Lakehouse / Warehouse
Point-and-click
ETL for analysts
• Why Fabric & DFG2 Now?
• Data Destinations are first-class Fabric items
• Automatic staging — no intermediate tables
• VNet data gateway support for enterprise connectivity
Gen1 vs Gen2 — What Changed?
Feature
Dataflow Gen1
Dataflow Gen2
Platform
Power BI (Premium/PPU)
Microsoft Fabric (any F-SKU)
Output
Dataset (semantic model)
Lakehouse, Warehouse, KQL DB
Staging
Manual intermediate steps
Automatic — built-in
Data Gateway
On-prem gateway
VNet gateway supported
Performance
Standard Mashup
Enhanced engine + scale-out
Incremental
Limited
Native incremental refresh
If you know Power Query, you're already 80% of the way there. Gen2 is just Power Query with Fabric superpowers.
Where DFG2 Fits in Fabric
Destinations
Sources
SharePoint
CSV / ERP
SQL Server
REST API
→
Dataflow
Gen2
Lakehouse
→
Warehouse
KQL DB
When to use DFG2 vs other approaches
• Dataflow Gen2
• Business analyst or data engineer who knows PQ, needs governed repeatable ETL
• Pipelines
• Orchestration, complex sequencing, dependencies, trigger-based workflows
• Shortcuts
• Zero-copy access to existing data — no transformation needed
• Notebooks
• Complex code-first transformations, ML preprocessing, Spark scale
Demo
5 Things to Take Away
• DFG2 = Power Query + Fabric-native destinations
If you know PQ, you're already productive.
• Start small, publish early
Don't build the perfect dataflow: build one that works, then iterate.
• Use parameters from day one
Avoid hardcoded paths and dates: your future self will thank you.
• The Folder Connector is your best friend
For monthly/incremental file loads, it's unbeatable.
• Set your refresh schedule before you walk away
A dataflow no one runs is just a draft.
Common Gotchas to Avoid
➢
No Data Destination set
Query runs successfully — but data goes nowhere. Always set destination before publishing.
➢
Mismatched column names
Source and destination schema must align. Use the column mapping UI to verify before publish.
➢
Wrong refresh mode
Using Replace on a large table when you needed Append loses all historical data. Check first.
➢
No gateway for on-premises sources
Plan connectivity early. VNet/on-prem gateway setup takes time: don't leave it for demo day.
Resources & Next Steps
• Dataflow Gen2 Documentation
learn.microsoft.com/fabric/data-factory/dataflows-gen2-overview
• Power Query M Function Reference
learn.microsoft.com/powerquery-m/power-query-m-function-reference
• Microsoft Fabric Community
community.fabric.microsoft.com
• YouTube, Blogs, Larger Community, etc
"Append, Folder Connector, Parameters, Lakehouse Destination — once you know these building blocks,
you can handle nearly any ETL scenario without a single line of code."
Sound off.
The mic is all yours.
Influence the product roadmap.
Join the Fabric User Panel
Join the SQL User Panel
Share your feedback directly with our
Fabric product group and researchers.
Influence our SQL roadmap and ensure
it meets your real-life needs
https://aka.ms/JoinFabricUserPanel
https://aka.ms/JoinSQLUserPanel
How was
the session?
Complete Session Surveys in
for your chance to WIN
PRIZES!