Understanding the Complete Data Journey: From Business Problem to Prediction in Microsoft Fabric
Full day training session for SQLBits 2026TL; DR
See how real business problems become insights and predictions in this hands on training day using Microsoft Fabric. Using a customer churn scenario, you’ll follow the full data lifecycle from ingestion to predictive ML, with no technical background required.
Session Details
This training session is designed for attendees who are new to data, haven’t found their niche, or decision makers who want to understand how a complete data solution is built from start to finish. It focuses on the full journey, from a business problem to insight and prediction, showing every step involved and explaining concepts clearly, without assuming prior technical experience.
Using a realistic business scenario, a company that is losing customers (customer churn), attendees will be taken on an end-to-end data journey using Microsoft Fabric. Starting with raw, imperfect data, we will use Fabric pipelines and dataflows to ingest, clean, and shape the data so we can identify where customers are leaving and the impact on the business. We will then model the data to support analysis and reporting before going further to explore why churn is happening and how predictive models built in Fabric notebooks can help identify which customers are likely to leave next.
The session is built around a continuous, practical demonstration, with each stage building on the last. There will be guided, introductory coding within Fabric notebooks as well as clicky-clicky-draggy-droppy pipeline creation. The aim is not to turn attendees into developers, but to demystify what the code is doing, why it exists, and how it fits into a wider data solution that supports both analytics and machine learning.
While the day is based on customer churn, the techniques and thinking shown apply to every data scenario, including sales performance, operational reporting, forecasting, and customer behaviour. Attendees will leave with a clear mental model of the entire data lifecycle and a better understanding of how tools like pipelines, dataflows, and notebooks work together within Fabric.
By attending this training day, you will:
-Understand how business problems translate into data and analytical solutions
-Learn why data quality, modelling, and preparation are critical foundations
-See how analytics and machine learning build on the same underlying data
-Gain confidence engaging with data solutions, even without a development background
-Leave with a transferable framework that can be applied to most real world data problems
This is a ‘zero to hero’ training day that shows how data, analytics, and machine learning come together in Microsoft Fabric to support better business decisions.
Using a realistic business scenario, a company that is losing customers (customer churn), attendees will be taken on an end-to-end data journey using Microsoft Fabric. Starting with raw, imperfect data, we will use Fabric pipelines and dataflows to ingest, clean, and shape the data so we can identify where customers are leaving and the impact on the business. We will then model the data to support analysis and reporting before going further to explore why churn is happening and how predictive models built in Fabric notebooks can help identify which customers are likely to leave next.
The session is built around a continuous, practical demonstration, with each stage building on the last. There will be guided, introductory coding within Fabric notebooks as well as clicky-clicky-draggy-droppy pipeline creation. The aim is not to turn attendees into developers, but to demystify what the code is doing, why it exists, and how it fits into a wider data solution that supports both analytics and machine learning.
While the day is based on customer churn, the techniques and thinking shown apply to every data scenario, including sales performance, operational reporting, forecasting, and customer behaviour. Attendees will leave with a clear mental model of the entire data lifecycle and a better understanding of how tools like pipelines, dataflows, and notebooks work together within Fabric.
By attending this training day, you will:
-Understand how business problems translate into data and analytical solutions
-Learn why data quality, modelling, and preparation are critical foundations
-See how analytics and machine learning build on the same underlying data
-Gain confidence engaging with data solutions, even without a development background
-Leave with a transferable framework that can be applied to most real world data problems
This is a ‘zero to hero’ training day that shows how data, analytics, and machine learning come together in Microsoft Fabric to support better business decisions.
3 things you'll get out of this session
-Understand how business problems translate into data and analytical solutions
-Learn why data quality, modelling, and preparation are critical foundations
-Leave with a transferable framework that can be applied to most real world data problems
-Learn why data quality, modelling, and preparation are critical foundations
-Leave with a transferable framework that can be applied to most real world data problems
Previous experience recommended
A laptop and access to the internet.
Speakers
Lewis Prince's other proposed sessions for 2026
AI Beyond the Hype; Practical Solutions for Real Businesses - 2026
Automating Multilingual Document Translation with Microsoft Fabric and Azure AI - 2026
Demystifying AutoML; Building Machine Learning Models with Azure - 2026
Responsible AI in Practice for the Microsoft Data Platform - 2026
Unlocking Insight from Text Using Azure AI Language Studio - 2026
Using Microsoft Copilot Studio to Build Chatbots That Answer Questions and Take Action - 2026
Lewis Prince's previous sessions
Fast and Efficent Text Analysis using Azure's Language Cognitive Service
We will go on a whirlwind tour of Azure Cognitive Services Language Studio, to show the wealth of tools available there, and then talking in particular about Key Word Extraction, Sentiment Analysis and Opinion Mining. We will apply these concepts through Language Studio APIs via Python on Customer Reviews data to then feed into a Power BI report to show how we can really squeeze so much more useful data out of chunks of text!
Copilot Studio: Guiding Your Chatbot Takeoff with AI Aviation!
I will talk you through how I created the Copilot chatbot for the SQLBits website.
The SQLBits Website ChatBot with Copilot Studio: How I built the chatbot for this events website)
I will talk you through how I created the Copilot chatbot for the SQLBits website.
Model Creation in Azure AutoML and Ingestion Through Excel
We will go through a brief introduction to what Machine Learning is and some of its applications. we will then explore why AutoML should be used by all; as a great starting point for anyone new to Machine Learning, as well as a time saving tool for those more experienced. Finally, we will expose a model as an endpoint and understand how we can use it. Specifically in this case how to use it in excel via VBA.
Technologies I will demonstrate are:
- Azure Machine Learning Studio
- VS code (with Azure Machine Learning Studio extension)
- Python
- VBA