---
name: data-cleaning-plan
description: "Clean it in the right order, or you'll do half of it twice. Use when the user wants help with Data Cleaning Plan. Related topics: Data, Technical, Productivity."
---

# Data Cleaning Plan

Clean it in the right order, or you'll do half of it twice.

## How to use this skill

1. Ask the user for each input listed below in a single message, skipping any they have already given.
2. Where options are listed, offer them, but accept other answers too.
3. Fill the answers into the prompt template, then follow the completed prompt to produce the result.

## Inputs

- **The data** (longer free text)
- **Where it came from** (longer free text)
- **Problems I've spotted** (choose any: Duplicates, Missing values, Inconsistent formats, Outliers, Wrong types, Free text where categories should be, Encoding mess)
- **What it feeds** (longer free text)
- **Tools** (choose one: Spreadsheet, Python, R, SQL)
- **Rows** (free text, e.g. "Roughly")

## Prompt template

## Cleaning plan

- The data: {{The data}}
- Where it came from: {{Where it came from}}
- Problems I've spotted: {{Problems I've spotted}}
- What it feeds: {{What it feeds}}
- Tools: {{Tools}}
- Rows: {{Rows}}

Give me the sequence, ordered to avoid rework. Deduplicating before standardising formats misses duplicates, so order matters more than technique here.

Per problem: how to detect it, how to fix it, and what to do when the right answer is genuinely ambiguous. Be specific about missing values, since dropping and imputing lead to very different conclusions.

End with the checks that prove the cleaned data is sound.

---

Source: [Data Cleaning Plan on PromptBuild](https://promptbuild.io/prompt/WrGooHazX0YtGSXyZkEG/data-cleaning-plan)