Skip to main content
CSV & Spreadsheet Data Quality

How to Validate a CSV File Before Importing It Into Any System

Importing a bad CSV creates problems that are 10x harder to fix than catching them before the import. Here's a pre-import validation checklist for any system.

You imported a CSV into your CRM. It looked fine. Three days later you discovered it created 1,800 duplicate records, overwrote 200 existing customer records with wrong data, and populated the phone column with values that were actually email addresses. The import was a disaster that took two weeks to clean up.

All of this was preventable with a pre-import validation. Here's how to do it.

The Pre-Import Validation Checklist

1. Check row and column counts Before importing, confirm: how many rows does the file contain? How many columns? Does this match what you expect? A file that should have 5,000 contacts but contains 500 or 50,000 is wrong before you've looked at any data.

Sohovi validates your dataset before it enters the warehouse — catching format errors, nulls, and duplicates at the source.

2. Confirm column headers match the target system Your CRM expects "email" but your file has "Email Address". Your system expects "company_name" but the file has "Account". Mismatched headers cause wrong-column imports — data going into the wrong field.

3. Validate the email column If you're importing contact data, check the email column for: completeness (how many rows have values?), format validity (do they all look like real email addresses?), and uniqueness (are there duplicates?).

4. Check date columns for consistent formatting Date columns with mixed formats (some MM/DD/YYYY, some YYYY-MM-DD, some written out as "January 5, 2024") will import incorrectly into most systems. Standardize before importing.

5. Verify unique identifier values If you're importing into a system that matches records by a unique field (customer ID, email address), check that field for: no nulls, no duplicates, and values that won't conflict with existing records.

6. Spot-check 20–30 random rows Automated checks catch systematic problems. A random spot-check catches unexpected issues — a row where columns appear shifted, encoded characters in text fields, or data that looks wrong on inspection.

7. Test with a small sample first Before importing the full file, import 10–20 rows into a test environment or a sandbox. Verify the results look correct before running the full import.

Sohovi profiles CSV files in seconds — showing completeness rates, email validity, duplicate counts, and format patterns — making steps 1–5 automatic before you run the import.

An hour of validation prevents days of cleanup. Build it into your import process permanently.

Selva Santosh

Data quality, for people who ship

Selva writes practical guides on data quality, profiling, and governance to help teams ship better data.

Start for free

Stop guessing. Start knowing your data quality.

Sohovi profiles your datasets in minutes — surfacing completeness gaps, type mismatches, and duplicate patterns before they reach production.

No credit card required · Free forever plan