Over the past six months, AI has gone from being a distant acquaintance to a trusted coworker. It has made it possible for me to do things I would never have been able to do, to do things faster, and produce better quality documentation. Here are some areas where I’ve leveraged AI, with specific examples.
Note: my company uses Copilot, which helped me write this article 😛
Reviewing Documents
This was the first one that opened my eyes to the possibilities of AI. Let’s face it – we rarely validate our content against our full style guide. But Copilot helped me take key content from our style guides and create a detailed validation prompt that output any deviations from the guide. It’s taken our style guide from a document we read once and hope we remember, to a living tool that actively improves the quality of our writing.
SharePoint
We use SharePoint for document storage and delivery, and I’ve been able to push the limits of what it can do, thanks to the help of Copilot. We’ve…
- Designed libraries for SOPs and Work Instructions.
- Collected and stored document metadata.
- Developed custom approval flows.
Power Automate
I’m somewhat familiar with Power Automate. I can build basic workflows, but I cannot build some of the more complex flows without doing way more googling than I’d like – and even then, I sometimes can’t do what I want. I’ve asked Copilot to give me step by step instructions for creating flows that:
- Approve and review documents, including automated reminders.
- Notifying writers of metadata they need to review upon uploading a document.
- Notifying writers of additional tasks the need to do after creating or modifying a button.
- Taking information from a spreadsheet and adding that information as metadata in SharePoint.
- Pulled visits data for documents on a monthly basis.
Power BI Dashboard
Like Power Automate, I know a bit about Power BI, but not near enough to make really useful dashboards. But, with a great deal of hand-holding from Copilot, I was able to produce a Documentation Governance dashboard that included multiple data sources, custom columns and measures, substantial data transformation, and KPIs. This is a task I would never have been able to do on my own – it really requires a great deal of knowledge about database queries. Instead, Copilot generated the queries/measures/etc., and I just copied and pasted them. Here’s more details about this process.