Power Your Documents with AI – Project Syntex


I recently wrote a blog about the power of creating and adding Content Types to your SharePoint document libraries. Content Types, in SharePoint, have been around for well over a decade, and have served people reliably for all of those years, but as with everything in technology, things  are constantly evolving, but every now and then you come across a revolution. Well, for document libraries, this is it. Project Syntex. 

SharePoint Syntex, at its core, utilises AI and Machine Learning in partnership to scan through documents in your libraries, and tag/extract content automatically so you never have to do it manually again. For example, a contract processing expert can teach SharePoint Syntex to get the contract’s value, along with the expiration date and key terms and conditions. Project Syntex is taught by people, not large datasets, so you can train it to do whatever you want.  

Below is an example of Project Syntex identifying the cost within a contractual agreement: 

project syntex

The cool thing about Project Syntex is that you only have to give the AI a few examples of the documents you want it to go away and scan after you’ve trained it, and it’s away at the races. It involves no code or prior development knowledge. 

Once SharePoint Syntex is away at the races, it can then automatically suggest or create metadata, invoke custom Power Automate workflows, and attach compliance labels to enforce retention or record management policies, all automatically and at an instant! 

Project Syntex isn’t all about tagging metadata though. For those of you who are information publishers out there who handle purchase orders, invoices or other forms, I’m sure you’re tired of spending time extracting information, transferring it to Excel or a database and passing the information forward in a business process.  

SharePoint Syntex also includes a powerful form processing engine, based on AI Builder (found on the PowerPlatform), that lets you automatically recognize and extract common values from these forms, such as dates, figures, names, or addresses.  

The brain behind Syntex only requires a few examples of a form that you are working with (like I mentioned earlier with the documents), then it’s able to start recognising these forms and apply it’s training to search for all the information you want it to scan for.  

Below is an example of a business expense being scanned by Project Syntex, and picking up all the key data: 

petty cash project syntex

Once the data is extracted, Project Syntex can then trigger a PowerAutomate workflow to log this data in the Business Expense list, and notify the relevant reviewers that it needs to be approved. All done autonomously. 

Project Syntex streamlines so many information management and document management procedures, and It has to be one of the most exciting SharePoint revolutions I have seen since I started working with SharePoint. I cannot wait until Syntex becomes available globally, so we can see more people using AI to make their work processes easier to follow, and more efficient.