Main Features within ChatGPT Plus

Function What it's good for
Default Writing copy, summaries, viewing (and analyzing) pictures
Advanced Data Analysis Data work (summaries, cleaning, creating), writing code
Bing Browsing the web
DALL-E 3 Image generation
Plugins Mix of plugins from external companies. Most are in early phases

General use cases

  • Clean and relabeling data
  • Summarize data
  • Create dummy data
  • Prep for technical interviews
  • Learn how to use any software tool faster
  • Learn how to edit HTML code on your website
  • Help with querying SQL
  • Write and debug code. Replit is an easy place to run code
  • Summarize any pdf, website, other document fast
  • Create images for presentations, websites, etc.
  • Upload a picture of something, and ask it a question

What it’s not consistently good at yet:

  • Building a model from scratch: It can get most of it right, but not consistently so would need to be able to check the work
  • Checking a model: Not consistent in beign able to check a model, unless very precise about what you want it to check. Even then, it may not be able to pinpoint the exact issue in a way you can immediately fix it
  • Converting tables from PDF to Excel: Sometimes works. Depends on the pdf and can take a decent amount of instruction to get the data in the desired format

Example Use Cases

  • Cleaning and relabeling data

    Example prompt: I'm concerned the customer names are inconsistent. Can you help me relabel them?

    Easily clean, relabel and recategorize data using ChatGPT Plus. See our blog post Use AI to Clean Up Messy Data for a step by step overview

  • Summarizing data

    Simple - Example Prompt: Please help me summarize revenue by customer in the attached sales data

    Quickly run simple data summaries by uploading your excel file. You can then ask ChatGPT to insert the summaries back into the Excel worksheet.

    Advanced - Example Prompt: Please run a customer cohort analysis. Start with revenue by cohort by month. Cohort is defined as first month of purchase

    When running advanced analysis (such as a customer cohort analysis), it is helpful to ask ChatGPT to run the anaysis step by step. Once a step is complete, ask ChatGPT to check that the analysis is correct before moving on to the next step.

  • Creating dummy data

    Prompt: I need to make up some data for an excel class I’m teaching. Can you help me make up some fake employee data – every row should be one employee with their name, annual salary, team, location, US-based tech company. I’d like 100 rows and once we take a look at it, we can put in excel

    In the prompt to create data, provide enough detail of the type of data you would like to create and any relevant parameters. See our blog post Create Dummy Data with ChatGPT for a step by step overview

  • Prepping for technical interviews

    Prompt: I’m prepping for IB interviews. The question is you have $30 of depreciation and a 30% tax rate, walk through the 3 financial statements. Please help me practice a concise and correct answer. Please answer in order of IS then CF then BS

    In your prompt, it will be helpful to include any preferences or specific things you are looking for in the interview answer. Use ChatGPT Plus on your phone to practice answering questions and get real-time feedback.

💡Pro Tips

Above all else, EXPERIMENT! (while respecting privacy considerations)

Experiment with what tool is best for any given problem. This will continue to evolve, but for now…

  • Cutting data – ChatGPT Plus: Advanced Data Analysis
  • Writing code – ChatGPT Plus: Advanced Data Analysis
  • Summarizing a model or financial statements:
    • PDF – Claude, ChatGPT: Advanced Data Analysis
    • Image of key model inputs/outputs – ChatGPT: Default

For prompts:

  • When prompting ChatGPT, it’s great to give ChatGPT a little context (e.g. “I need to make up some data for an Excel class I’m teaching” and “assume US-based tech company”) to help tailor results
  • Think of working with ChatGPT as a conversation. You don’t have to get everything right on the first try.

When uploading / downloading files: 

  • When uploading Excel files try not to have unnecessary empty rows & columns or coach it on where to start
  • Experiment with Excel vs. PDF vs. image vs. link to the webpage to see which format will return best results
  • If your session times out, links and file downloads may not work.  If so, ask ChatGPT to recreate the relevant content

When conducting data analysis:

  • Always check your work to help confirm accuracy of analysis
  • For more advanced analysis, it helps to go step-by-step and check work at each step

 

Additional resources

To improve your Excel and Google Sheets skills, get better at double-checking with sumifs, countifs or pivot tables, join our Spreadsheet Fundamentals courses. 

You can also stay up-to-date on upcoming events as part of our AI and Data Analysis Learning Series by subscribing to our calendar.

Learn more about partnering with Recalc on AI training for your organization

Scroll to Top