CSV Cleaning

CSV files are one of the most common ways prospect data moves between tools, and one of the easiest places for bad data to hide. This category covers how to clean CSV files before importing into CRMs like HubSpot and Salesforce, how to remove duplicate contacts, how to standardise formatting, and how to validate emails and phone numbers before outbound.

Whether you're preparing a list for import, merging exports from different tools, or fixing a messy file before it reaches your CRM, these guides walk through the practical steps.

Why CSVs break on import

Most import failures are not caused by dirty data in the obvious sense. They are caused by structure: a delimiter appearing inside an unquoted field, an encoding mismatch that turns accented characters into replacement glyphs, a leading zero stripped from a phone number or postcode because a spreadsheet decided the column was numeric, or a date read as US format when it was written in UK format.

Deduplication is where the real judgement sits. Exact-match dedupe on email address is easy and catches the least. The duplicates that cost you are the near-matches: the same company as 'Acme Ltd', 'Acme Limited', and 'ACME LTD.', or the same person at two email formats. Normalising company names and email domains before you dedupe finds substantially more than deduping the raw values.

Order matters more than tooling. Standardise formats first, then deduplicate, then validate, then map fields to the destination. Running dedupe before standardisation misses the near-matches. Running validation before dedupe means paying to validate records you are about to delete. Mapping fields first means redoing the mapping every time the cleaning step changes a column.

Where to start

DataFixr CSV cleaning workflows

See how DataFixr supports CSV cleaning, deduplication, and HubSpot import preparation.

Guides in this topic