Ways I Use ChatGPT at Work (with examples!)

Margaret Efron
Learning Data
10 min readJun 5, 2024

--

Watch out. Your cat is learning ChatGPT. Photo by Catherine Heath on Unsplash

I use ChatGPT daily for work and personal questions. I think of it as my assistant. It’s “forgetful,” because my chats time out, and it’s not always correct, because it’s subject to hallucinations.

However, ChatGPT is still useful for work, including answering questions about Excel and SQL, and brainstorming ideas when my mind is blank.

Read on for my favorite ways to use ChatGPT at work!

Ways I Use ChatGPT at Work

1. To ask questions about Excel.

The most common way I use ChatGPT at work is to learn Excel.

I ask ChatGPT to troubleshoot Excel errors, explain Excel formulas, and determine which formula to use.

For example, I asked ChatGPT the difference between COUNT() and COUNTA() formulas.

My ChatGPT prompt: “What’s the difference between the COUNT formula and the COUNTA formula in Excel?”

I asked ChatGPT the difference between the COUNT and COUNTA formulas in Excel.

As with any ChatGPT output, you must know Excel fundamentals to understand the output and correct it, if necessary. You should also test out the formulas in Excel to see if they work as ChatGPT expects.

2. To brainstorm article ideas.

When your mind is blank but you need to draft an email, an article, or a headline for an upcoming event, using ChatGPT is better than staring at a blank page. Not all ChatGPT ideas will be good, so use your discretion!

My ChatGPT prompt: “I want to write articles to help aspiring data analysts learn how to use Excel. Can you help me generate article ideas?”

I asked ChatGPT to help me generate article ideas about learning Excel.

3. To brainstorm article content.

Now that you’ve used ChatGPT to generate article ideas, you need content. Luckily, ChatGPT can draft an article outline, as well!

My ChatGPT prompt: “If I’m writing an article to data analysts about how to use Excel pivot tables, what should I focus on? This is for beginners who don’t have experience with Excel pivot tables.”

ChatGPT also advised:

Use screenshots and visuals to illustrate each step of creating and working with pivot tables.

Keep the language simple and avoid overwhelming beginners with too much technical detail.

Provide downloadable sample datasets for readers to follow along with this tutorial.

All in all, this is a good article outline. However, it is still general and high-level. I would go back and forth with ChatGPT for specific tips and practical examples of what to include.

Also, there is no way I could include ALL this information in a simple Medium article — it would be too long.

As with all ChatGPT output, review it with a critical eye. If I were writing this article, I would include the first few bullet points: intro to pivot tables, prepping data, and creating a pivot table.

4. To proofread and generate simple SQL code.

I use ChatGPT to check my SQL syntax, troubleshoot SQL errors, and comment SQL code.

For example, if the SQL code is not working as expected, I may type a prompt like the one below.

My ChatGPT prompt: “Why isn’t this code working? SELECT*

WHERE[First Name] = “Dawson”

FROM table;”

I used ChatGPT to explain why my SQL code wasn’t working.

I also use ChatGPT to generate simple SQL code. I include the relevant columns and context for the question I’m trying to solve.

My ChatGPT prompt: “What is the correct SQL syntax if I want to sort a table by base_salary_final_amount in ascending order?”

I used ChatGPT to generate SQL code to sort data.

A few cautionary notes when using ChatGPT with SQL:

  • Be careful that it doesn’t make up SQL functions that exist in other programming languages, but not SQL.
  • SQL is a standard but has numerous distinct “flavors” — e.g. MySQL, SQL Server, Postgres. Check that the output is the “flavor” of SQL you need.
  • As you should when using AI with any technology, you must understand SQL fundamentals, so you can tell if the ChatGPT output is accurate.

5. To summarize technical concepts in a way that non-technical coworkers can understand.

As a data analyst, you often have to explain technical concepts to people without a technical background. To do this effectively, you must simplify and use as little jargon as possible. I use ChatGPT to explain technical concepts to people without technical backgrounds.

For example, I once had a coworker who wanted to report on the 90th percentile, but when I provided them with the 90th percentile, they looked confused. They asked, “What does this number mean?”

I used ChatGPT to explain the 90th percentile to them.

My ChatGPT Prompt: “What does it mean if a salary is in the 90th percentile? Explain in terms that a person without a technical background would understand.”

I asked ChatGPT to explain the 90th percentile.

ChatGPT says: “If a salary is in the 90th percentile, it means that the salary value is greater than or equal to 90% of the salaries in a particular dataset. In other words, it indicates that the individual’s salary is higher than the salaries of 90% of the people in the reference group.”

Now I can email my coworker back with this simpler explanation of the 90th percentile. If I want more context, I can ask ChatGPT for examples as well.

6. To ask basic questions, like in a Google search.

Use ChatGPT like a Google search engine. As long as you don’t need the responses to be super accurate or up-to-date, ChatGPT provides good general knowledge.

My ChatGPT prompt: “What are some good stretches I can do at my desk at work? Provide a list of specific examples.”

I asked ChatGPT for stretches I can do at my desk.
ChatGPT gave me a list of stretches I can do at my work desk.

ChatGPT listed 7 stretches for my workday, including neck stretches, shoulder and upper back stretches, chest openers, ankle circles, and hip and leg stretches. All of these would be easy to do at my desk!

7. To ask general office etiquette questions.

I ask ChatGPT how to handle certain problems at work, including keeping my meetings on track and discussing communication preferences with my manager.

My ChatGPT prompt: “I’m a data analyst and I have weekly 1-on-1 meetings with my manager. I find that these meetings sometimes get off track. What are some ways I can ensure that these meetings stay productive and focused on the agenda at hand?”

I asked ChatGPT for help in keeping my meetings on track.
ChatGPT listed practical suggestions to keep my meetings focused.

ChatGPT offered general guidance to increase the productivity of my meetings, including sharing an agenda with my manager before the meeting, having specific time slots for each item, and using visual aids like charts, graphs, or slides to keep the discussion focused.

My ChatGPT prompt: “What are some ways you can ask your boss about their preferred communication styles? This is for an office job.”

I asked ChatGPT how I should talk to my boss about their preferred communication style.
ChatGPT suggested ways I can talk to my boss about their communication style.

ChatGPT gave suggestions that hadn’t occurred to me, like discussing frequency and timing and clarifying times/dates that are best for discussion.

8. To ask for time productivity tips.

Sometimes I have trouble focusing on work projects or finishing them on time. ChatGPT can give me suggestions on how to improve my productivity.

My ChatGPT prompt: “What are some tips for maximizing time productivity at my job as a data analyst? Be as specific and practical as possible. Give me a list of ways to improve my time productivity.”

I asked ChatGPT for tips on how to increase my productivity.
ChatGPT gave advice on how to increase my productivity.

ChatGPT listed ways to increase my productivity at work. In particular, I’ve found the following to be helpful: minimizing meetings and asking coworkers to send me concise emails instead when possible, taking short breaks between tasks to recharge, and delegating tasks.

If you don’t like ChatGPT’s initial list of suggestions, you can say so, using prompts like: “I don’t like the Pomodoro technique for time management. Are there other methods of time management I should use instead?” or “What are some specific ways I can minimize meetings at work?”

9. To interpret emails from the IT desk.

Sometimes, I can’t understand emails from the IT desk, either because it has jargon I don’t know, or it is weirdly phrased. When this happens, I copy and paste the text from the email into ChatGPT and ask for a summary (as long as the information is not confidential.)

My ChatGPT prompt: “Can you simplify this language for a layperson? The language is in quotation marks.”

I asked ChatGPT to summarize emails from the IT desk.

10. To list words or phrases in alphabetical order.

I ask ChatGPT to list words, phrases, or names in alphabetical order.

My ChatGPT prompt: “Put these words in alphabetical order [insert list]”

I asked ChatGPT to list a series of words in alphabetical order.
ChatGPT listed the words in alphabetical order.

As with all output, it’s important to double-check ChatGPT’s response. I only see one error here: ChatGPT repeated the word “accordingly” twice.

11. To summarize text.

To summarize notes from a meeting or an email thread, copy and paste the information in ChatGPT and ask for a summary (given that the information is not confidential.)

My ChatGPT prompt: “Can you summarize a thread of emails for me? [then copy & paste the emails]”

I asked ChatGPT to summarize an email thread for me.
ChatGPT summarized the email thread.

This is useful for sending out meeting highlights, summarizing a long email thread, or understanding key points from a memo or article.

12. To ask questions about Outlook

When I have Outlook questions, I ask ChatGPT and test its response. For example, I received an email that I wanted to respond to in the future. I asked ChatGPT how to create an Outlook calendar event so I could remember to follow up at a certain time.

My ChatGPT prompt: “Is there a way in Outlook to set an email as a calendar event for yourself? For example, if I receive an email and want to remember to respond at a certain time.”

ChatGPT told me how to create an event from an Outlook email.

I tested the ChatGPT solution and it worked!

13. To categorize a list of things into logical groupings.

I often have lists of items that I need to sort into groups, but I’m not sure how. For example, I was writing an article on Excel keyboard shortcuts, and had a list of shortcuts: CTRL C is “copy,” CTRL A is “select the entire dataset,” CTRL 1 opens a format cells dialog box, and so on.

I wanted to group the Excel keyboard shortcuts into categories that would make sense for my article, so I copied & pasted the list into ChatGPT and asked it to categorize the shortcuts.

My ChatGPT prompt: “I am writing an article about Excel keyboard shortcuts. I have a list of keyboard shortcuts that I want to group into logical categories. Can you help me sort these shortcuts into logical groups? [insert list]”

ChatGPT sorted the keyboard shortcuts into the following categories: navigation shortcuts, selection shortcuts, editing shortcuts, formatting shortcuts, data shortcuts, worksheet management shortcuts, and search & replace shortcuts.

This saved me so much time and energy!

This is also useful for data analysts preparing for a presentation. Instead of throwing random data points on a PowerPoint slide, talk to ChatGPT and present your data points in a logical order.

Final Thoughts

ChatGPT is a great search engine if you don’t feel like using Google. Even better, you can communicate back and forth and tailor its responses.

However, if you are worried that ChatGPT will take over your data analyst job, don’t worry! AI isn’t at the point where it has domain knowledge or fully understands the nuances of the data and the stakeholder needs.

ChatGPT may make your job easier, but you still need to know the data analytics fundamentals: how to store, retrieve, collect, and transform data. You need to have these core skills to check the output for accuracy.

As with all ChatGPT output, use your discretion, because ChatGPT is subject to “hallucinations,” or making up information.

For more information, check out the Maven Analytics course on ChatGPT and Data Analytics, which gives tips about how to use ChatGPT for Excel, Google Sheets, Power BI, SQL, and Python.

I also recommend watching the Maven Analytics “Mavens of Data” interview with Alex the Analyst, “The Future of AI in Data.” Alex gave great tips about how he uses AI in work and its current shortcomings.

Further info:

Maven Analytics: ChatGPT and Data Analytics course

Maven Analytics: “The Future of AI in Data” Interview with Alex the Analyst

My Medium article: 5 Ways I Use ChatGPT to Level Up My SQL Skills

My Medium article: Top 5 Ways I Use ChatGPT to be an Excel Ninja

My Medium article: Breaking Down Barriers: Using ChatGPT to Explain Data to Your Client

The contents of external submissions are not necessarily reflective of the opinions or work of Maven Analytics or any of its team members.

We believe in fostering lifelong learning and our intent is to provide a platform for the data community to share their work and seek feedback from the Maven Analytics data fam.

Submit your own writing here if you’d like to become a contributor.

Happy learning!

-Team Maven

--

--

Margaret Efron
Learning Data

I love all things data and write about Excel, Power BI, and SQL. I currently work as a Business Systems Analyst at the Darden School of Business.