Just as many believed MS Excel was becoming obsolete, Microsoft introduced the general availability of Copilot in Excel. This feature enables users to interact with Excel using natural language.
Interestingly, this development comes after Microsoft recently released a research paper SpreadSheet LLM, a framework that enables LLMs to effectively process and analyse complex spreadsheet data.
“Microsoft released a copilot in Excel, and it has officially caught up with Google in the AI race,” posted a user on X. “Fantastic to watch”
Meanwhile, Google also recently integrated Gemini into Google Sheets. Its key features include the ability to generate content based on user needs. Whether summarising spreadsheets, creating tables, formulating custom formulas, or simplifying data explanations, Gemini is designed to address a wide range of data analysis needs.
The demand for Copilot grew as professionals began shifting away from Excel in favor of alternatives like Equals, Tomat AI, and Julius AI which had AI features.
However, Microsoft Excel remains the world’s most popular spreadsheet software. As of 2024, over 175,723 companies globally continue to rely on Excel as a key productivity tool. In fact, four out of five Fortune 500 companies use Excel, and more than two billion people around the world work with spreadsheets.
Why Copilot in Excel?
Excel has hundreds of formulas and the struggle is real to remember them. Now, instead of searching online for formulas, one can simply ask Copilot directly within the spreadsheet, making it easier to expand formula usage.
Moreover, Copilot is now capable of reasoning over structured data, beyond just tables, and can perform tasks like adding filters or splitting text. It also assists in highlighting key data points using formula- based conditional formatting.
For instance, Copilot makes it easier to flag important details, such as identifying when spending exceeds revenue in a sales report. Not only that, users can now work with unformatted data, without needing to structure it as a table.
The company has also introduced new features to fully utilise Excel’s capabilities, including support for more advanced formulas like XLOOKUP and SUMIF, better conditional formatting, and the ability to refine visualisations such as charts and PivotTables with Copilot.
The Death Knell for Business Intelligence Tools
With Copilot in Excel, it would be very easy to make data driven decisions. As a part of Copilot analysis, it now suggests the most effective visual formats, such as bar charts, line graphs, or PivotTables, and configures the appropriate fields, layouts, and filters for users.
This allows users to make full use of Excel’s powerful features without needing deep expertise. For example, a sales consultant can quickly understand the relationship between ad spend and campaign reach, as Copilot generates a PivotTable to assist in planning the budget for the next quarter.
“Microsoft demoing Copilot in Excel- basically just an automated version of classic BI,” posted a user on X. This feature is likely to challenge popular BI tools such as Tableau, Qlik Sense, Looker, and Zoho Analytics.
Not all analysis involves numerical data, and Copilot’s capabilities now include text analysis as well. It can process raw text and convert it into actionable insights, enabling detailed examination of both numerical and textual data with the same accuracy and efficiency. For example, Copilot can now analyse all the customer feedback from the past quarter and surface the top three concerns.
Copilot in Excel with Python
Excel spreadsheets just got a major overhaul. Microsoft has announced a public preview of Python in Excel. It allows users to perform advanced data analysis directly within Excel by integrating Python. This feature enables users to use natural language to describe the type of analysis they want, and Copilot will automatically generate, explain, and insert the necessary Python code into the spreadsheet.
It allows users to carry out complex tasks like forecasting, risk analysis, machine learning, and data visualisation using natural language, eliminating the need for coding. It’s as if you have an expert data analyst on hand.
To offer a complete experience to users, Microsoft also partnered with enterprise Python repository Anaconda, which includes libraries such as pandas, statsmodels, seaborn, and Matplotlib.
Copilot provides transparency by explaining the generated Python code and creating a dedicated analysis sheet. This sheet serves as a sandbox for users to interact with Copilot, while the original data remains untouched. Users can also refresh the analysis as the data changes.
In the past, many enterprises that have been hesitant to use ChatGPT for Excel-related tasks can now take advantage of Copilot, which is expected to be more secure and not use enterprise data.
Interestingly, when Copilot in Excel with Python was announced last year, there were concerns about its local availability. To address these privacy issues, Microsoft has assured users that they are providing ‘enterprise-level security.’ This means the code will run in cloud-isolated containers with no network access, though some customers may still find this approach concerning.
Copilot in Excel with Python is currently being made available to Windows Insiders, who are users participating in early testing of new software features. To use this feature, a Microsoft 365 Copilot license is required.
Once this feature is activated, users can initiate advanced data analysis by either clicking on the ‘Advanced analysis’ prompt provided by Copilot or by entering a custom prompt asking Copilot to ‘analyse with Python.’
The post How to Excel in Excel with Copilot appeared first on AIM.
