Columns shift unexpectedly
A quoted field can legally contain a comma or a newline. Check for unmatched quote characters and confirm that escaped quotes are doubled. Compare the declared delimiter against what the destination expects. Use CSV Structure Validator before editing the file.
Accented letters become symbols
Check whether the importer expects UTF-8 or a legacy encoding, and look for a byte-order mark. Do not assume that replacing visible replacement characters can restore original bytes. See how to diagnose damaged characters.
Leading zeros disappear
A spreadsheet may interpret identifiers such as 00123 as numbers. Import those columns as text, not numeric values. Use mixed-type audit to identify columns where inferred types could be unsafe.
Dates swap day and month
Dates such as 04/05/2026 are ambiguous. Specify the source locale and use ISO-style dates when appropriate; do not silently guess. See date normalization guidance.
Rows fail a target system's rules
Check missing required values, duplicate keys, unusual delimiters and schema-specific constraints. Record the exact row numbers from the target application's error report. Compare a known-good single-row import before trying bulk transformations.
CSV formulas or spreadsheet warnings
Cells beginning with =, +, - or @ can be interpreted as formulas by some spreadsheet software. DataCleaner enables a formula guard for CSV output, but protections vary across importers. Follow the CSV formula safety guide.
Keep a source backup and compare row counts, identifiers and a sample of special-case records. Never assume a file is repaired solely because its parser returns no error.
Related resources
Browse data tools · Data cleaning checklist · How our parser works