CSV import fails in integration hub when special characters present

CRM Zoho CRM
, , , , , , , ,
1

I’m running into a critical issue with CSV imports through the Integration Hub in Zoho CRM 2021. When importing customer data containing special characters like ñ, é, ü, and Asian characters, the import either fails completely or corrupts the data.

The problem seems related to UTF-8 encoding validation during the import process. I’ve verified the CSV file is UTF-8 encoded, but characters still get mangled. For example, “José García” becomes “Jos� Garc�a” after import.

Additionally, field mapping accuracy is affected - some fields with special characters don’t map correctly to the target fields. The character normalization doesn’t seem to be working as expected.

Has anyone successfully imported CSVs with international characters? What encoding settings or pre-processing steps are needed to ensure data integrity?


Import Error Log:
Row 47: Character encoding mismatch
Field 'Company Name': Expected UTF-8, received ISO-8859-1
Status: Import halted at 46/500 records
2

I had similar issues last year. The problem is that even if your CSV is UTF-8, Excel sometimes saves it as UTF-8 with BOM which Zoho doesn’t handle well. Try opening your CSV in Notepad++ and converting it to UTF-8 without BOM. That fixed about 70% of my encoding problems with special characters.

3

Another thing to check - are you using the correct delimiter? If your data contains commas and you’re using comma as delimiter, that could be causing field mapping issues that appear as encoding problems. Try using pipe (|) or tab delimiters instead. Also, wrap all fields in double quotes to prevent special characters from being interpreted as delimiters or control characters.

4

Let me address all three key aspects of your import issue systematically.

UTF-8 Encoding Validation: The Integration Hub in Zoho 2021 requires explicit encoding declaration. First, ensure your CSV is genuinely UTF-8 without BOM (Byte Order Mark). Use a hex editor or specialized tool to verify - many text editors claim UTF-8 but add invisible characters. In the Integration Hub import configuration, navigate to Advanced Settings and explicitly set:


File Encoding: UTF-8 (strict)
Character Set Validation: Enabled
Invalid Character Handling: Replace with ?

Special Character Normalization: The import engine needs preprocessing for proper normalization. Before importing, use DataPrep to create a transformation workflow:

  1. Add a “Normalize Text” step for all text columns
  2. Apply Unicode NFC (Canonical Decomposition followed by Canonical Composition) normalization
  3. Set fallback encoding to UTF-8 for any unrecognized characters

This ensures characters like ñ, é, ü are stored in their composed form rather than decomposed (base character + combining diacritic), which prevents display issues.

Field Mapping Accuracy: Your field mapping problems stem from character encoding affecting field boundary detection. Implement these fixes:

  1. Always enclose field values in double quotes in your CSV: “José García”,“Müller GmbH”
  2. Escape any internal quotes by doubling them: “Company ““Special”” Name”
  3. In the mapping configuration, set explicit data types for each field - don’t use auto-detect
  4. For fields with international characters, map to “Text (Multilingual)” field type, not standard “Text”

Additional Validation Step: Before running the full import, test with a 10-row sample containing your most problematic characters. Check the preview in Integration Hub - if characters display correctly in preview but fail on import, the issue is in the commit phase, not the read phase. In that case, you may need to contact Zoho support to check database-level character set settings for your instance.

Workaround for Persistent Issues: If the above doesn’t resolve it completely, consider using the Zoho CRM API for imports instead of CSV. The API handles UTF-8 natively and bypasses the CSV parser entirely. You can write a simple script to read your CSV and push records via REST API with proper encoding headers.

This combination of proper encoding declaration, normalization preprocessing, and explicit field type mapping should resolve your import issues while maintaining data integrity for international characters.