How to compare two spreadsheets with different column names
When column headers and row order differ, a line-by-line diff creates noise. Match records first, then compare equivalent fields.
Short answer
Choose a stable key (often a composite key such as fund + ISIN + valuation date). Write a mapping between differently named columns, normalise dates and numbers, set approved tolerances, and report missing, extra, duplicate and changed records. Never match solely by row position.
A practical example
A custodian export calls a field MktVal_GBP, while Finance calls it Market Value. Both are mapped to the same concept. The record key might be Fund + ISIN + AsOfDate. Check for duplicate keys on both sides before comparing values, or a VLOOKUP-style first match can hide a break.
Six checks to run
- Confirm both extracts use the same cut-off and population.
- Agree record identity: key columns and null-handling.
- Write field mappings and type conversions.
- Set currency, decimal and date tolerances explicitly.
- Separate missing, extra, duplicate and changed records.
- Keep source values, target values, rules version and run date with the report.
What about free CSV diff tools?
For a quick one-off comparison of ordinary CSVs, a browser-based keyed diff can be appropriate. For example, ReconFiles documents keyed CSV reconciliation and RowSlice documents keyed CSV comparisons. Review each tool's limits and data-handling model before using confidential information. For multi-format financial extracts with rules and repeatable evidence, a scoped reconciliation engagement may be a better fit.
Want to check two real extracts?
ShipDataFast's Recon Check offers a first break report at no charge. It is a service-led proof, not an unlimited do-it-yourself web app. Anonymised or sample files are fine. Agree a safe transfer method before sharing sensitive data.