De-duplicate CRM leads and keep the latest touch

Question

The CRM exported the same lead several times because the email casing differs and every touchpoint was logged.

Write `dedupe_leads(df)` that lower-cases and strips `email`, keeps only the most recent row per email by `touched_at`, and returns the result sorted by email with a reset index.

Accepted Answer

Step 1: Normalise before de-duplicating. ' Asha@mail.com' and 'asha@mail.com' are the same person, but not to `drop_duplicates`. Step 2: Sort by `touched_at` descending so the most recent touch is the first row within each email. Step 3: `drop_duplicates('email', keep='first')` then keeps exactly that row. Step 4: Re-sort by email and `reset_index(drop=True)` so the output is stable and the stale index does not leak downstream.

De-duplicate CRM leads and keep the latest touch

More Pandas — Cleaning questions