Session Abstract

Copilot in SharePoint analyses document libraries and suggests metadata columns for each file. But it creates columns per library, it doesn't set your taxonomy, build site columns, or create content types. Without a prepared architecture, the AI-generated columns could reinforce fragmentation rather than resolve it. The session shows how this works through live demos. You'll see the difference between metadata you can use and metadata that just adds more to clean up. We walk through the architecture decisions you have to make first, what works and what doesn't, and where you must step in manually even when the technology works as intended. You leave knowing what's your job, what AI can take off your hands, and how to prioritise when you can't clean up the whole tenant at once.

Mats Warnolf

I love learning new things and solving old problems. I started my days as an admin back in 1990 and have had many roles since. I became a full time Microsoft Certified Trainer in 1999. Today I work as an independent consultant specializing in Microsoft 365 productivity. Microsoft 365 MVP. I produce a bi-weekly podcast in Swedish called Office 365-podden.