How to analyse files and spreadsheets with ChatGPT

Open a new chat

Click the + New Chat button – usually a plus‑sign at the top‑left of the ChatGPT homepage. You’ll see a blank conversation window with a faint line that says “Start typing…” at the bottom. The platform prepares a fresh context so the AI will focus only on the file you later provide. If you don’t see a plus‑sign, look for a button that says “New chat” or a similar icon in the upper corner.

You’ll know it worked when the chat pane clears and a blinking cursor appears ready for you to type.

Upload your file

Locate the paper‑clip icon or a button labelled “Upload a file” underneath the text box. Clicking it opens your computer’s file picker; navigate to the PDF or spreadsheet you saved and select it. After you confirm, a small thumbnail (for PDFs) or a file name (for spreadsheets) appears inside the chat area. This is the AI’s equivalent of handing a paper report to a colleague. If you don’t see a paper‑clip, look for a button that says “Attach” or a menu item called “Upload”.

You’ll know it worked when the file name or preview shows up right under the message box.

Ask for a plain‑language summary

Type a concise request such as:

Please give me a short summary of the main points in this file.

When you press Enter, ChatGPT scans the uploaded document and begins typing a paragraph that highlights headings, key figures and any conclusions it finds – much like a teammate giving you the gist of a meeting. The response may take a few seconds, especially for larger PDFs. If the reply looks generic, try adding “in plain language” to your prompt.

You’ll know it worked when the AI returns a short paragraph that mentions the document’s major sections or numbers.

Query specific data

Now ask a focussed question, for example:

What was the total revenue listed for Q3?

or, for a spreadsheet:

Which product category had the highest sales in the East region?

After you send the prompt, the model looks through the file, finds the relevant row or cell, and replies with the exact value and a brief context. If the answer seems unrelated, make sure you referenced the correct file – you can re‑upload it or specify “in the spreadsheet”. If the AI says it cannot find the data, try phrasing the request differently, such as “show me the cell that contains Q3 revenue”.

You’ll know it worked when the response includes a number (e.g., “$2.8 million”) and mentions the row or column where it was found.

Request a simple chart

Finally, ask the AI to visualise the data. A typical prompt looks like:

Create a bar chart of quarterly sales from the spreadsheet.

The assistant will generate an image of a bar chart and embed it directly in the chat. You can right‑click the image and choose “Save image as…” to download it for a report or presentation. Simple charts are supported on both free and paid tiers, but the free plan may limit how many images you can generate each day. If no chart appears, check that the file contains numeric sales data and try a more explicit request such as “draw a line chart of the sales column”.

You’ll know it worked when a picture of a chart shows up beneath your message, complete with labelled axes.