top of page

How to Use Copilot in Excel: 25 Prompts for Finance, HR, Sales and Operations (2026)

4 hours ago
4 min read
AI-generated illustration: finance professionals reviewing data together
AI-generated illustration. Not a photograph.

By Parikshit Khanna, corporate AI trainer, TEDx speaker and founder of Digital Training Jet | Published 22 October 2026


Quick answer

Copilot in Excel creates formula columns, explains formulas, builds charts and PivotTables, highlights outliers, cleans data and runs deeper analysis with Python in Excel. The 25 prompts below are grouped by team. Use your own column names, check every number against the source, and keep a human sign-off for anything that goes to management.


What is new in 2026

  • Core features: Microsoft lists formula columns with explanations, charts, PivotTables, summaries, trends and outliers, formatting, sorting and filtering, and advanced analysis through Python in Excel without needing Python skills. Microsoft Support

  • Agentic Copilot in Office apps: on 22 April 2026 Microsoft made Copilot's agentic capabilities generally available and the default in Word, Excel and PowerPoint for Microsoft 365 Copilot customers. Microsoft 365 Blog

  • June 2026 update: Plan with Copilot shows intended changes before editing, Show Changes marks edits made by Copilot, and skills and financial data connectors were added for Microsoft 365 Copilot customers. Microsoft 365 Blog, 25 June 2026

  • Retired: Microsoft says the COPILOT worksheet function is no longer available from 14 September 2026, so older workbooks that used it need another approach. Microsoft Support


25 Copilot in Excel prompts

#

Prompt

Team

1

Add a column that calculates margin percentage from Revenue and Cost, and explain the formula.

All

2

Explain what this formula does in plain language.

All

3

Fix the formula in column F; it returns errors for blank rows.

All

4

Highlight rows where Days Overdue is more than 30.

All

5

Clean this data: trim extra spaces, standardise city names and convert dates to DD-MM-YYYY.

All

6

Split the Full Name column into First Name and Last Name.

All

7

Create a summary sheet with the total, average and maximum of each numeric column.

All

8

Build a variance table of actual against budget by cost centre, with variance and variance percentage.

Finance

9

Show receivables ageing in 0 to 30, 31 to 60, 61 to 90 and over 90 day buckets.

Finance

10

Find duplicate invoice numbers and list them on a new sheet.

Finance

11

Add a column that flags expenses above the approval limit of Rs 50,000.

Finance

12

Summarise the key trends in this table in five bullets for the CFO.

Finance

13

Create a headcount summary by department, location and grade.

HR

14

Calculate monthly attrition from the joining and exit date columns.

HR

15

Show average training hours by department in a bar chart.

HR

16

Show total sales by region and month as a PivotTable.

Sales

17

Rank sales reps by quarterly revenue and add their target achievement percentage.

Sales

18

Show lead to order conversion rate by lead source.

Sales

19

Which five products had the biggest drop in units sold compared with last quarter?

Sales

20

Sort inventory by days of stock remaining and flag items below seven days.

Operations

21

Are there outliers in delivery times by warehouse? Show them.

Operations

22

Analyse using Python: forecast the next six months of sales and chart the forecast.

Analysis

23

Analyse using Python: group customers into segments by order value and frequency.

Analysis

24

Plan the changes first: I want this sheet turned into a monthly dashboard. Show the plan before you edit.

Review

25

List the assumptions behind this analysis and anything I should verify.

Review

Replace anything in brackets or generic wording with your own details. Prompts are written by Parikshit Khanna for training use.


AI-generated illustration: AI training for finance teams
AI-generated illustration. Not a photograph.

Tips that make these prompts work

  • Keep one header row and clear column names; Microsoft suggests adapting sample prompts to your own column names.

  • Describe the task in plain language, then ask Copilot to explain any formula it creates.

  • For complex analysis, ask Copilot to analyse using Python.

  • Use Show Changes and Plan with Copilot where available so you can see what will change.

  • Verify totals against the source before sharing, and keep payroll and personal data within your company's rules.


From Parikshit Khanna's training sessions

Photo from a Parikshit Khanna AI training session
Real photo from Parikshit Khanna's training sessions.
Parikshit Khanna teaching a Google Gemini workshop at IIT Guwahati
Real session photo: Google Gemini workshop at IIT Guwahati (Synapse, November 2025).
Photo from a Parikshit Khanna AI training session
Real photo from Parikshit Khanna's training sessions.
Photo from a Parikshit Khanna AI training session
Real photo from Parikshit Khanna's training sessions.
AI training session for the Aries Agro team
Real session photo: AI training for Aries Agro.
Participants at a Parikshit Khanna training session
Real photo: participants at one of Parikshit Khanna's training sessions.

These are real photographs from Parikshit's training work; captions name the session where it is known.


Train your team on Copilot in Excel

Parikshit Khanna runs hands-on workshops where teams practise these prompts on their own work, with review rules and a saved prompt library.

Step

What you share

1. WhatsApp or email

Organisation, city, team size and AI tools or licences

2. 20-minute call

Priority tasks and data rules

3. Written proposal

Agenda, format, dates and one fee; dates are blocked after written confirmation



About Parikshit Khanna

Achievement

Detail

TEDx speaker

Talk Redesigning Work with Artificial Intelligence, TEDxEicher School Faridabad Youth, 1 August 2026

Copilot and AI work, 2026

Copilot Chat workshop for GMR Airports' Delhi Duty Free; ChatGPT and Copilot workshop for vivo India; AI training for CP PLUS

Corporate clients

Emami, Arvind Fashions, Sheela Foam (Sleepwell), Godrej Properties, Gaursons, Hetero, Tinna

Government and academia

Prasar Bharati NABM programme; workshops at IIT Roorkee and IIT Guwahati; visiting faculty at GL Bajaj

International

Malabar Group, Dubai: Phase 1 delivered live online, July 2026

Reach

Trainer-reported reach of 3,50,000+ professionals since 2020

Credentials

Anthropic Claude Academy badges for Claude 101, Claude Cowork, and AI Capabilities and Limitations


Related guides


Frequently asked questions

Can I still use the =COPILOT function?

No. Microsoft says it is no longer available from 14 September 2026.

Does Copilot in Excel need a licence?

Microsoft's guide says Copilot in Excel needs a Microsoft 365 Copilot licence. Check with your IT team.

Can Parikshit train our team on these prompts?

Yes. Workshops run onsite in Indian cities on agreed dates or live online for teams in India, the UAE, the UK, Singapore and Indonesia, using your own tasks and sample data.

How do we book?

WhatsApp +91 99972 13177 or email pkhanna123@gmail.com. You get a written proposal with one recommended format and fee after a short call.


Note: product details are from official vendor pages as of October 2026 and change often; features depend on your licence and region. Reach is trainer-reported. Training is independent; no partnership with the vendors named is implied.

bottom of page