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?”

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?”

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 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?”

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. 

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.”

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.”

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?”

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.”

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

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!