De-duplicate a signup table in pandas

Question

A signup export has duplicate emails (the same person submitted the form twice) and a few rows with a missing name.

Write a function `clean_signups(df)` that returns a cleaned DataFrame: drop rows where `name` is null, then keep only the **first** row for each `email`. Reset the index so it runs 0.

Accepted Answer

Step 1: `dropna(subset=["name"])` removes rows missing a name without touching nulls in other columns — always scope `dropna` with `subset`. Step 2: Order matters. Dropping null names *before* de-duplicating means a null-name row can never win the `keep="first"` race for its email. Step 3: `drop_duplicates(subset=["email"], keep="first")` keeps the earliest signup per email. Step 4: `reset_index(drop=True)` renumbers the surviving rows; `drop=True` stops the old index becoming a column. Step 5: Every step returns a new DataFrame, so the caller's original data is untouched.

De-duplicate a signup table in pandas

More Pandas — Cleaning questions

Asked at Zoho