Excel has been one of the most important business tools for decades. What is changing now is the way people interact with the spreadsheet. Instead of relying entirely on formulas, PivotTables, manual analysis and years of Excel experience, users can increasingly describe what they want in natural language and use AI to help build, analyze and explain the workbook.
Microsoft Copilot in Excel can help users analyze data, generate formulas, create charts and PivotTables, transform information, apply formatting and make changes to workbooks. Microsoft also provides natural-language data analysis capabilities that can surface trends, comparisons, outliers and other insights. Microsoft’s documentation on Copilot in Excel recommends reviewing and verifying AI-generated results before relying on them.
For organizations already using Microsoft 365, this creates an important opportunity: Excel does not have to be replaced to take advantage of AI. Instead, AI can become another layer on top of an existing business tool.
What Can AI Actually Do in Excel?
AI in Excel is most useful when it helps a person understand data, perform repetitive work, create a first draft of an analysis or turn a business question into an Excel operation.
- Explain trends in a dataset
- Identify unusual values and outliers
- Generate or improve formulas
- Create charts and PivotTables
- Summarize large tables
- Classify or transform data
- Highlight important information
- Suggest ways to analyze a dataset
- Help clean and organize information
- Perform deeper analysis using Python in Excel where available
Microsoft’s current Copilot experience can work in chat, planning and editing modes, depending on the user’s environment and licensing. That means AI can be used simply to ask questions about a workbook, or it can be used to help plan and carry out changes. Microsoft’s Copilot in Excel FAQ documents these capabilities and their availability.
Tip #1: Stop Asking Generic Questions
One of the easiest ways to improve AI results is to stop using vague prompts.
Instead of asking:
“Analyze this spreadsheet.”
Try something like:
“Analyze the sales table. Compare Q1 and Q2 revenue by region, identify the three regions with the largest percentage increase, identify any unusual declines, and create a chart showing the comparison.”
Microsoft recommends being specific, naming the columns or ranges involved, starting broad and refining the result with follow-up questions. Microsoft’s Copilot in Excel tips provides similar prompting guidance.
Tip #2: Ask AI to Explain the Formula
Excel users often inherit workbooks containing formulas they did not create. Instead of spending 20 minutes trying to understand a complicated formula, ask Copilot to explain it in plain English.
For example:
“Explain this formula in plain English. Identify what each reference does, what conditions are being evaluated, and what result the formula returns.”
This is especially useful when an organization has spreadsheets that have become critical business systems but were never formally documented.
Tip #3: Use AI to Create Formulas From Business Requirements
You do not always need to know the exact Excel function before asking for help. Describe the business rule.
For example:
“Create an Excel formula that marks an employee record as ‘Review Required’ when the training date is more than 12 months old, the employee is active, and the department is not blank.”
AI can provide a starting formula that can then be tested against known examples.
This is one of the most practical uses of AI in Excel because the user can describe the desired business logic instead of translating the requirement into Excel syntax first.
Tip #4: Turn Questions Into Charts
Sometimes the problem is not calculating the answer. The problem is seeing the answer.
For example:
“Show monthly revenue for the last 24 months as a line chart and identify the months with the largest increases and decreases.”
Copilot can create charts and PivotTables from appropriate Excel data. Microsoft specifically documents using Copilot to create visualizations, highlight information, sort and filter data, and surface important patterns. See Microsoft’s guidance on visualizing data with Copilot in Excel.
Tip #5: Ask Follow-Up Questions
AI analysis does not have to be a one-question interaction.
Start with:
“What are the most important trends in this dataset?”
Then ask:
- “Why is that category changing?”
- “Show me the underlying numbers.”
- “Compare that result with last year.”
- “Which locations are driving the change?”
- “Show me the outliers.”
- “Create a chart for management.”
Microsoft documents this conversational approach for Copilot in Excel and notes that follow-up questions can build on the context of the previous question.
Real Business Application #1: Budget Analysis
Finance teams can use AI to accelerate the first pass of budget analysis.
For example, a manager could ask:
“Compare actual spending against budget by department. Identify departments more than 10 percent over budget, summarize the largest variances, and create a management summary.”
The result should still be reviewed by the appropriate finance professional. AI can accelerate analysis, but it should not replace financial controls or human approval.
Real Business Application #2: Sales Analysis
A sales team could use AI to identify revenue patterns across customers, products or territories.
A useful prompt might be:
“Analyze revenue by customer and product. Identify the fastest-growing products, customers with declining revenue and any unusual changes from the previous period.”
The same workbook can then be used to create charts or management reports.
Real Business Application #3: Employee and HR Data
Excel is frequently used for employee tracking, onboarding lists, training records, staffing analysis and other operational tasks.
AI can help identify missing information, classify records, summarize trends or create formulas for business rules.
For example:
“Review the employee table for missing required fields. Identify incomplete records and create a column explaining which information is missing.”
Organizations should be particularly careful with sensitive employee information. Access controls, appropriate data handling and organizational AI policies should be established before employees use AI with sensitive business data.
Real Business Application #4: Operations and Inventory
Operations teams can use Excel and AI to look for inventory patterns, recurring shortages, unusual demand and supplier issues.
Try:
“Identify products where average monthly demand increased while inventory levels decreased. Rank them by potential supply risk and create a summary table.”
This turns a broad business question into an analysis that can be investigated further.
Real Business Application #5: Project Management
Project managers often maintain Excel files containing tasks, owners, dates, budgets and status information.
AI can help turn that information into a management view.
“Review the project task table. Identify overdue tasks, tasks due within 14 days, owners with multiple overdue tasks and dependencies that could affect the target completion date.”
That can provide a useful first-pass project health analysis without requiring the manager to manually filter multiple columns.
AI + Excel + Power Automate
The bigger opportunity comes when Excel is not treated as an isolated spreadsheet.
Microsoft 365 organizations can connect Excel with SharePoint, Power Automate, Power Apps and other Microsoft services. That creates an opportunity to move from “AI helps me analyze a spreadsheet” toward “AI and automation help operate a business process.”
For example, an organization could collect a request through a form, place structured information into a Microsoft 365 data source, apply business rules through Power Automate consulting, and use Excel for analysis and reporting.
That is where AI becomes part of a broader business architecture rather than simply another productivity feature.
Python in Excel: Taking Analysis Further
For organizations that need more advanced analysis, Python in Excel can extend what can be done inside the workbook. Microsoft documents Python in Excel as a way to use Python directly in Excel cells and work with Excel ranges, tables and other worksheet objects.
Python in Excel can be useful for more sophisticated statistical analysis, data preparation and visualization. Availability depends on the Microsoft 365 environment, Excel version, update channel and licensing.
Microsoft also documents advanced Copilot analysis that can use Python-based analysis for more complex questions. This creates an interesting bridge between traditional spreadsheet analysis and more advanced data science workflows.
One Important Excel AI Change in 2026
If you have seen articles recommending the Excel COPILOT() worksheet function, be careful with older information.
Microsoft states that the COPILOT worksheet function was discontinued on September 14, 2026. Existing cached results can remain in workbooks, but recalculation of that function can return a #NAME? error. Microsoft recommends using Copilot in Excel instead.
This is a good example of why organizations should verify AI and Microsoft 365 guidance against current Microsoft documentation rather than relying on older AI-generated tutorials.
The AI Excel Hack: Ask for the Business Answer, Not the Excel Function
One of the biggest changes AI introduces is that users can begin with the business question instead of the technical implementation.
Instead of thinking:
“Which Excel formula do I need?”
Start with:
“What business question am I trying to answer?”
Then tell Copilot what the data represents, identify the relevant columns, define the desired output and ask it to show or explain the work.
This approach can make Excel more accessible to people who understand their business but do not consider themselves advanced spreadsheet users.
AI in Excel Still Requires Human Review
AI-generated formulas, summaries and analysis can contain errors. Microsoft explicitly recommends reviewing, editing and verifying AI-generated content.
That is particularly important when Excel is being used for financial reporting, compliance, legal decisions, employee decisions, operational controls or other high-impact business processes.
A good operating model is simple:
- Ask AI to produce a first result.
- Review the logic.
- Test the result against known data.
- Check formulas and assumptions.
- Have the appropriate business owner approve important outputs.
- Document repeatable AI-assisted processes.
How AI in Excel Fits Into a Microsoft 365 Strategy
AI in Excel should not be viewed separately from the rest of the Microsoft environment. Identity, data, permissions, governance, collaboration and automation all affect how AI can be used safely and effectively.
Organizations considering broader Microsoft 365 modernization can start with Microsoft 365 consulting to evaluate their current environment and identify opportunities across Microsoft 365.
For organizations evaluating AI more broadly, an AI readiness assessment can help examine technology, data, processes, security and organizational readiness.
An organization developing a broader roadmap can also explore AI strategy consulting and AI governance consulting.
Practical Copilot Prompts for Excel
- Trend analysis: “Identify the most important trends in this dataset and explain what changed.”
- Outliers: “Find unusual values or outliers and explain why they may deserve attention.”
- Formula: “Create a formula that applies this business rule and explain how the formula works.”
- Management report: “Summarize the most important findings for an executive audience.”
- Chart: “Create the most appropriate chart to compare these categories over time.”
- Data quality: “Identify missing, inconsistent or unusual values in this table.”
- Follow-up: “Now explain the biggest driver behind that result.”
What This Means for Businesses
AI is not making Excel obsolete. It is changing what people can ask Excel to do.
The most valuable use cases are not necessarily flashy. They are often practical improvements to the everyday work organizations already perform: analyzing budgets, reviewing sales, identifying exceptions, preparing management reports, checking data quality, documenting formulas and turning repetitive spreadsheet work into structured processes.
For organizations already invested in Microsoft 365, this can be a natural starting point for AI adoption because employees are already familiar with Excel and the surrounding Microsoft ecosystem.
If your organization is looking beyond individual productivity improvements and wants to identify where AI can improve business processes, contact WS Tech Consulting to discuss your environment and objectives.