HomeCasa Analyze a CSV dataset and propose insights and charts
Analyze a CSV dataset: data quality checks, exploratory analysis, actionable insights and recommended charts with the code to build them. Designed for ChatGPT, Claude and Gemini. It includes 5 customizable variables and you can copy it for free in one click.
Intermediate · Text · Pending editorial review
Prompt
Act as a senior data analyst who explains results to non-technical audiences.
I am attaching (or pasting below) a CSV file about {{topic}}.
Column descriptions: {{columns}}.
Business question I want to answer: {{question}}.
Audience for the analysis: {{audience}}.
Tool I will use for the charts: {{tool}}.
Deliver, in this order:
1. Data quality check: row and column counts, detected types, missing values, duplicates, outliers and possible formatting errors. Explain how you would handle them before moving on.
2. Exploratory analysis: descriptive statistics for the key variables, distributions, relevant segments and trends over time if there are dates.
3. An answer to the business question with the numbers that support it and your level of confidence (what is correlation and what might be causation).
4. Three to five actionable insights, each with: finding, supporting data point, estimated impact and recommended action.
5. Four to six proposed charts: chart type, axis variables, why that chart, and the message its title should convey.
6. Code or step-by-step instructions to build those charts in {{tool}}.
7. Limitations of the analysis and additional data that would strengthen it.
Constraints:
- Work only with the data provided. Do not invent values or fill in missing data without saying so.
- If you cannot read the file or a column is ambiguous, ask me before continuing.
- Round numbers consistently and always state units.How to use it
- Upload the CSV to an assistant that can analyze files (ChatGPT or Claude with data analysis) or paste a representative sample.
- Describe each column and its units, even if they seem obvious.
- Ask one clear business question.
- Double-check the key numbers yourself before presenting the findings.
Usage example
Topic: online store sales in 2025 · Question: which categories and channels explain the Q2 sales drop?
Sample output
I can't see the attached file yet, so I won't give you any figures. Here's exactly what I'll do as soon as you share it:
1. Data quality check
| Check | Column | What I'm looking for |
|---|---|---|
| Types | date, amount_usd, units | Valid dates and numbers without "$" or stray commas |
| Duplicates | order_id | Repeated orders that would inflate sales |
| Missing | category, channel | Rows with no category: [to be checked] |
| Outliers | amount_usd | Negative amounts (returns?) or extreme values |
3. Answering the question (method)
I'll compare Q1 and Q2 by category and channel and calculate how much each combination contributes to the total drop.
import pandas as pd
df = pd.read_csv("sales_2025.csv", parse_dates=["date"])
df["quarter"] = df["date"].dt.quarter
t = df[df["quarter"].isin([1, 2])].pivot_table(
index=["category", "channel"], columns="quarter",
values="amount_usd", aggfunc="sum", fill_value=0)
t["change_usd"] = t[2] - t[1]
print(t.sort_values("change_usd").head(10).round(0))5. Proposed charts
| Chart | Axes | Title message |
|---|---|---|
| Waterfall | Contribution to the drop by category | "[Category X] explains most of the Q2 drop" |
| Grouped bars | Channel × quarter | Which channel fell the most |
Correlation isn't causation: a drop in one channel may come from price or ad-spend changes the CSV doesn't capture…
Key facts
- Category
- Data management · Data analysis
- Output type
- Text
- Level
- Intermediate
- Designed for
- ChatGPT, Claude and Gemini
- Verified in
- Pending editorial verification
- Variables
- Data topic, Columns, Business question, Audience and Charting tool
- Language
- English
- Price
- Free, no sign-up
- Updated
- Nov 7, 2026
Frequently asked questions
Is it safe to upload my data to an AI?
Remove or anonymize personal and confidential data before uploading, and review the privacy policy of the tool you use.
What if the file is too large?
Upload a representative sample or pre-aggregated data (for example, monthly totals by category) and ask the AI for code to rerun the full analysis on your own machine.
What is this prompt for?
Analyze a CSV dataset: data quality checks, exploratory analysis, actionable insights and recommended charts with the code to build them.
Which AI tools does it work with?
It is designed for ChatGPT, Claude and Gemini. We have not verified it in these tools yet, so results may vary between them.
How do I customize the prompt?
Click “Customize” and fill in the variables (Data topic, Columns, Business question, Audience and Charting tool). The text updates instantly and you can then copy it in one click. Values are stored only in your browser.
Do I need experience to use it?
It is intermediate: it helps to know the topic so you can provide good context and review the result critically.
Is this prompt free?
Yes. Every HomeCasa Prompts prompt is free, with no sign-up or subscription. You can also use them in professional projects.
Can I trust the AI output?
Treat it as a high-quality draft: always check data, figures and important claims before publishing or making decisions.