Section 4: Excel/Google Sheets, data analysis and reporting

Scenario 35: Creating a verifiable chart from a CSV sales export

Level: Intermediate · Tools: ChatGPT Data Analysis

Who and in what situation

Onur, a client reporting specialist at a digital agency in İzmir, reviews the date, campaign, spend, clicks and conversions fields in a CSV from an advertising platform. The same campaign may have different names.

How it is done today

Onur filters the CSV in Excel, enters the rates and prepares a line chart for a presentation. If the chart and the source total do not match, he retraces his steps.

How to do it with AI

  1. He attaches the CSV or XLSX file to a ChatGPT conversation. These formats are supported; file and plan limits may vary by account. [71] [79]
  2. He first asks it to report column types, missing values, duplicates and differences in campaign names. After uploading, he can inspect rows and columns in the interactive table view. [75]
  3. He asks it to calculate total spend and conversions by campaign and week, list the columns it used and create a line chart. ChatGPT can perform Python-based calculations and create charts. [71]
  4. He reviews the Python code, output and assumptions, and manually recalculates a few rows. If he finds the result satisfactory, he downloads the table or chart. [71] [75]

Example prompt

Review {the CSV file I uploaded} for data quality. List column types, missing values, duplicate rows and spelling differences in campaign names. Then calculate spend, conversions and cost per conversion by week and campaign. Show the filters, formula or Python code, and present the results in a line chart.

Evidence and real-world example

No published independent case study was found for this agency scenario. OpenAI’s official documentation describes CSV/XLSX uploads, interactive tables, code-assisted calculations, data quality checks and chart creation. [71] [75]

Points to consider

ChatGPT may choose an unsuitable method or chart on the first attempt; specify the method and columns explicitly. [71] On a personal plan, check your organisation’s policy; on Business or Enterprise, check the terms regarding exclusion from model training. [79] [80]

Sources for this scenario

  1. Data analysis with ChatGPT (help.openai.com)
  2. Extracting insights with ChatGPT data analysis (help.openai.com)
  3. Uploading files and audio to ChatGPT (help.openai.com)
  4. Business data privacy security and compliance (openai.com)