Understanding Numbers with AI Help

Chapter 1: What is Data Analysis and Why It Matters?

Imagine you run a small coffee shop. You sell espresso, cappuccino, and cold brew. At the end of the month, you look at your sales receipts and see that you made $12,000. Is that good? You have no idea. You don't know if that's up or down from last month, which drink sells best, or whether your busiest day is Tuesday or Saturday. You have data — thousands of numbers — but no understanding. That is the problem data analysis solves: it turns raw numbers into answers you can act on.

Data analysis is the process of inspecting, cleaning, transforming, and modeling data to discover useful information and support decisions. It is not about doing complex math in your head. It is about asking questions like "What is happening?", "Why is it happening?", and "What should I do next?" — and using numbers to answer them.

In this course, you will learn how to use AI tools like ChatGPT to do data analysis for you. You do not need to be a statistician or a programmer. You need to know what to ask and how to check the results. This chapter gives you the foundation: what data analysis actually is, why it matters, and how you can start doing it today with a free tool.

The Real Problem: You Are Drowning in Numbers

Every business, every website, every app generates data. A small online store might record every order: the product name, price, date, customer city, and payment method. That is easily 500 rows of data per month. A human cannot read 500 rows and spot patterns. You might notice that one product sold out, but you will miss that sales drop every time you run a discount on another product, or that customers from one city buy twice as much as customers from another.

Data analysis exists because human brains are terrible at processing large sets of numbers. We are excellent at recognizing patterns in stories and images, but we cannot hold 500 numbers in our working memory. AI tools, however, can process thousands of rows in seconds. The skill you need is not calculation — it is asking the right questions and interpreting the answers.

What Data Analysis Actually Does

At its core, data analysis involves four operations:

• Summarizing: Calculating averages, totals, minimums, and maximums. For example, "What is the average order value?" or "What was the total revenue in March?"
• Comparing: Looking at differences between groups. For example, "Do customers who buy coffee also buy pastries more often than customers who buy tea?"
• Finding trends: Identifying patterns over time. For example, "Are sales increasing every month, or is there a seasonal dip in summer?"
• Spotting anomalies: Finding values that do not fit the pattern. For example, "Why did sales triple on May 14th?"

When you ask ChatGPT to analyze data, it performs these operations for you. But you must understand what these operations mean so you can judge whether the answer makes sense.

Worked Example: Sales Data for a Small Bookstore

Let us work through a realistic example. You have a small online bookstore. You exported your sales for the first six months of the year. The data is in a CSV file — a plain text file where each line is a row and commas separate the columns. Here is what the first few rows look like:

date,book_title,category,price,quantity,city
2025-01-03,The Silent Patient,Mystery,14.99,2,New York
2025-01-05,Atomic Habits,Self-Help,16.50,5,Chicago
2025-01-07,Dune,Science Fiction,12.99,1,Los Angeles
2025-01-10,The Silent Patient,Mystery,14.99,3,Chicago
2025-01-12,Atomic Habits,Self-Help,16.50,2,New York

You have 180 rows like this. You want to know: What is my total revenue? Which category sells best? Is there a trend over time? Which city is my strongest market?

You could open this in a spreadsheet program and manually calculate each answer. That would take you 20 minutes and you might make a mistake. Instead, you will ask ChatGPT to do it.

Step-by-Step: Uploading Data to ChatGPT

This works with ChatGPT Plus (which includes the Code Interpreter feature, now called Advanced Data Analysis). If you have the free version, you can still follow along by pasting the data as text — the steps are slightly different, and I will note them.

Step 1: Prepare Your File

Make sure your data is in CSV format. If you exported from a spreadsheet, it likely already is. If you have an Excel file (.xlsx), you can still upload it — ChatGPT handles both. The file should have a header row (the first row with column names) and no empty rows.

Step 2: Upload the File

Open ChatGPT in your browser. Look at the bottom of the chat window — there is a paperclip icon (📎) or a plus sign (+) next to the message box. Click it and select "Upload file." Choose your CSV file from your computer. You will see the filename appear in the chat.

Step 3: Ask Your First Question

Type this exact prompt:

Analyze the attached CSV file. It contains sales data for my bookstore. 
Please:
1. Calculate total revenue (price multiplied by quantity for each row, then sum).
2. Show total revenue by category.
3. Show total revenue by month.
4. Show total revenue by city.
5. Identify any trends or anomalies you notice.
Present the results in a clear summary with numbers.

Press Enter. ChatGPT will read the file, perform the calculations, and give you a response. If you are using the paid version, it will write and run Python code behind the scenes to do this. If you are on the free version, it will do the calculations manually from the pasted text — which works for small datasets but is slower and less reliable for large ones.

Step 4: Ask for a Chart

After you get the summary, ask for a visualization:

Create a bar chart showing revenue by category, and a line chart showing revenue by month. 
Make sure the charts are labeled clearly.

If you are on the paid version, ChatGPT will generate the charts as images you can view and download. On the free version, it cannot generate images, but it can describe what the chart would look like or give you the data in a table format.

What the Results Mean

Suppose ChatGPT tells you that total revenue is $18,432.50, that Mystery books account for 42% of revenue, that January was your best month at $4,100, and that New York customers generate 55% of sales. Now you have actionable information:

• Mystery books are your strongest category — consider stocking more of them.
• January was strong — was there a holiday promotion? If so, repeat it.
• New York dominates — maybe you should focus marketing there, or investigate why other cities lag.

This is the value of data analysis: it converts raw numbers into decisions.

Common Mistakes Beginners Make

Expert Tip: Always Ask for the Method

Practice Task: Analyze Your Own Data

Create a small CSV file with 10 rows of data about something you care about. It could be your monthly expenses, the books you read this year, or the hours you spent on different activities each week. Use three columns: a date or label, a category, and a numeric value.

Then do the following:

1. Upload the file to ChatGPT (or paste it as text).
2. Ask for the total, the average, and the breakdown by category.
3. Ask for one trend or anomaly.
4. Ask ChatGPT to show the code or method it used.
5. Verify one number manually — pick one row, do the calculation yourself, and confirm it matches.

You should finish in under 15 minutes. If you can answer "What did I learn about my data that I did not know before?" then you have successfully completed your first data analysis.