CSV data cleaning checklist
Build a repeatable cleaning sequence: profile the file, normalize structure, clean values, validate output, and keep a change record.
CSV cleaning guides and workflow playbooks
These guides are focused on real files people need to clean: spreadsheet exports, CRM lists, order reports, bank transactions, survey responses, and recurring Python pipelines. Each guide ends with a local-first workflow you can test in the browser tool.
Start here
For one-off cleanup
Use the checklist, Excel, date/number, and schema guides before importing a file into a spreadsheet or BI tool.
For business exports
Use the CRM, ecommerce, bank, and survey guides when the CSV comes from a specific workflow with domain-specific risks.
For recurring jobs
Use the Python automation guide to turn the same cleaning rules into repeatable scripts with reviewable output.
Core guide set
Excel CSV import pitfalls and fixes
Handle semicolon exports, quoted commas, date conversion, leading zero identifiers, and values Excel tries to reinterpret.
CSV schema validation before analysis
Define expected headers, duplicate-header rules, type checks, required fields, and exception handling before loading data.
CSV date and number cleaning
Standardize date shapes, numeric separators, null values, and ambiguous locale formats before calculating metrics.
Clean CRM contact exports
Normalize contact fields, lifecycle stages, owner names, blank values, and duplicate contacts before import or campaign use.
Prepare ecommerce order CSV files
Protect order IDs and SKUs while cleaning totals, status labels, fulfillment dates, and duplicate rows.
Normalize bank transaction CSV files
Prepare transaction exports for reconciliation with date, amount, reference, and audit-trail checks.
Clean survey CSV results
Preserve open-text answers while cleaning respondent IDs, submitted dates, ratings, segment labels, and duplicates.
Python CSV automation
Move from manual cleanup to repeatable scripts with deterministic transforms, logs, validation, and local QA.
Editorial standards
We keep the guide set intentionally narrow. Pages are included here only when they explain a distinct workflow, name concrete failure modes, and connect back to a testable cleaning routine in the local tool.