Excel Copilot: 10 Game-Changing Features

Key insights

• COPILOT function: Microsoft embeds a large language model directly into the grid so Copilot can analyze text and generate content inside cells.
  Type "=" or a natural-language prompt and Copilot will autocomplete and propose workable formulas.
• Explain formulas & track changes: Copilot explains complex formulas in-place so you see how calculations work without leaving the sheet.
  It also plans edits, asks clarifying questions, and records every AI-made change for traceability.
• Python in Excel: Copilot runs Python code inside the workbook to support forecasting, risk analysis, machine learning, and richer visuals.
  This keeps advanced analysis and results within the same Excel file.
• Charts, PivotTables and visual summaries: Copilot builds and edits charts, PivotTables and dashboard-style outputs from your data.
  It highlights trends, outliers, and key insights and links visuals back to the source ranges.
• Agent Mode and Skills: Agent Mode enables multi-step, agentic actions and you can switch between Chat, Plan and Edit modes for safer workflows.
  Custom skills let you automate repeatable tasks and apply workbook-specific rules and personalization.
• Cross-workbook, document import and connectors: Copilot can combine files from your computer, OneDrive or SharePoint and extract data from PDFs and documents.
  Expanded connectors (including finance connectors) let you pull trusted market and private data into analyses.

Overview of the video

Kenji Farré (Kenji Explains) [MVP] published a concise video titled "Top 10 Excel Copilot Features You Need to Know," aimed at analysts and power users. In the video, he demonstrates recent updates that make Copilot in Excel more than a chat helper by showing it generate formulas, explain calculations, and automate workflows. Moreover, Farré highlights how these tools work directly inside workbooks, which reduces context switching and speeds up common tasks. As a result, viewers get a practical sense of how the features behave in real spreadsheets and which ones matter most for data work.

Key features highlighted

Farré walks through headline features such as the COPILOT function, natural-language formula generation, and the ability to explain formulas in-place. He also showcases multi-step automation via Agent Mode, the integration of Python for advanced analysis, and the new concept of reusable Skills for repeatable tasks. Additionally, cross-workbook merges, document import from PDFs, and model selection options are shown to expand Copilot's reach across data sources. Consequently, the video frames these capabilities as practical tools that reduce manual work while enabling richer analysis.

How Copilot changes everyday workflows

First, Farré emphasizes that automation can speed up routine tasks, such as building charts, PivotTables, and dashboards from raw tables without complex menu navigation. Then, he shows how generating formulas from plain language and having Copilot explain them improves understanding and reduces errors for less-experienced users. Furthermore, running Python inside Excel opens advanced forecasting and visualization directly in the workbook, which benefits analysts who previously moved between tools. Therefore, Copilot often shortens workflows and concentrates analysis inside a single file.

Tradeoffs and user control

However, Farré also points out several tradeoffs that teams should consider before depending on Copilot. For example, automation increases speed but can obscure the exact transformation steps unless users rely on the new planning and change-tracking features to maintain transparency. Moreover, choosing between different AI models presents a balance between cost, speed, and accuracy, and organizations must decide which option matches their risk tolerance. Finally, integrating external data and running code within workbooks raises governance and privacy questions that demand clear policies and testing.

Challenges and practical limitations

Farré demonstrates real limits, such as occasional formula suggestions that require adjustments and the need to verify AI-generated outputs before trusting them in reports. Additionally, cross-workbook assembly and PDF imports work well for many cases but can struggle with inconsistent formatting or very large datasets, which means manual cleanup still plays a role. He also notes that some advanced features may depend on subscription level or platform version, so compatibility can be a practical hurdle for mixed environments. Consequently, teams should plan pilots and user training to manage expectations and maintain data integrity.

Advice, best practices, and takeaways

To balance benefits and risks, Farré advises enabling planning and change-tracking, so users can review Copilot’s intended edits before applying them, which improves trust and auditability. He also recommends combining Copilot's quick suggestions with human review and versioning, thereby keeping control while gaining speed. Finally, creating reusable Skills and using model selection thoughtfully can standardize workflows and reduce surprises across teams. Overall, the video presents a pragmatic view: Copilot in Excel offers notable productivity gains, but success depends on governance, verification, and sensible adoption strategies.

Keywords

Excel Copilot features, Excel AI Copilot tips, Excel Copilot tutorial, Excel Copilot functions list, Excel Copilot data analysis, Excel Copilot shortcuts, Microsoft 365 Copilot Excel, Excel Copilot automation