Off-the-shelf AI gives everyone the same starting point. What makes it valuable to you is how you actually work: your process, your checks, what matters to your business. Copilot in Excel comes with built-in skills, but the more valuable ones are the ones you build yourself. Now that's a lot easier. Have a back-and-forth with Copilot to get to the right outcome, then ask it to turn that work into a skill, and it's yours to reuse.

For World Spreadsheet Day, we made Excel challenges inspired by finance legends. Solve one with Copilot for a shot at dinner with them. plsfixthnx.com

For the past year we've focused on helping Copilot understand your workbook. The next step is helping it take on more of the work. You can now turn workbook data into an interactive canvas that stays connected as the numbers change, teach Copilot the processes you repeat with skills, and spend a lot less time getting data ready.

Big addition to Excel... lists, arrays in cells, and nested arrays. Continuing to expand on the capabilities of the grid which makes Excel an even richer platform for you and your agents to build solutions on top of.

Almost every service that works with data has an "Export to Excel" button. But most of those exports are just snapshots. Whether the button creates a CSV or an XLSX file, the data is disconnected from its source as soon as the file is created. A better approach is a connected workbook that can refresh and stay up to date with the source data. That's exactly what Power Query is built for. This is why we created the open-source Connected Workbooks project on GitHub. It helps developers create refreshable Excel workbooks built around Power Query, while preserving the formulas, charts, and business logic people need to keep working with the data. Azure Data Explorer, Log Analytics, Datamart, and Viva Sales already use the library for their Excel integrations. Since the project is open source, any developer can use the same approach in their own product. github.com/microsoft/connect…

How does someone become an NFL Analyst? I met with Seahawks Analyst Brian Eayrs to learn about his career and how he uses Copilot in Excel. It all started with a genuine passion for both football and spreadsheets.

We're seeing instances of more than 1,200 people in a single Excel workbook at one time, including 500 making direct edits. Excel multi-player usage has more than tripled over the past four years. Tracking scenarios make up some of the largest categories of workbooks we see in Excel, which is exactly where multi-player is so critical. It works across desktop, web, and mobile, and we continue to invest in making these higher-scale use cases more robust. As agents do more work in Excel, people and agents will be working in the workbook together, pushing this foundation to completely new levels of scale.

Trusting your spreadsheet starts with being able to trace what changed and reverse it when needed, whether the changes came from AI or your coworkers. Most AI can help with what's in front of you. The new Change History skill helps Copilot in Excel understand what happened before. We've all come back from vacation to a spreadsheet that looks nothing like the one we left behind. Change History can understand changes made by both people and AI, summarize what matters, and answer questions about how the workbook evolved. In this example, I ask Copilot what changed, which edit broke a formula, and then restore the last working version. You can try it yourself in any workbook stored in SharePoint/OneDrive... just ask Excel Copilot: “Summarize the changes made in the last two weeks.”

A common way people collaborate in Excel is through a shared list. Everyone can add what they need, but over time the structure can drift. Excel Copilot can clean it up. If it happens regularly, turn the cleanup instructions into a skill and run it again whenever needed.