Blank rows and empty cells
Separate completely blank rows from missing values in otherwise populated records. Review whether those fields are required.
Check CSV Data Quality Online
Find blank rows, duplicate records, whitespace and missing values in your CSV. Review what needs attention before import, directly in your browser.
CSV data quality checker
Drop your CSV here
Scan a local CSV for factual findings before import.
UTF-8 CSV only. Up to 25 MB for this lightweight tool.
Raw CSV data stays in this browser. Findings are not an import guarantee. Nothing is cleaned automatically, and your source file is never overwritten.
Check before cleanup
What is a CSV data quality check?
A CSV data quality check inspects the contents of a parsed file for specific, observable problems. It helps you find CSV errors and cleanup candidates before preparing a business-system import.
This checker reports counts, columns and affected rows instead of an invented health score. Some findings, such as empty optional fields or intentional uppercase codes, may be acceptable in your workflow.
Common findings
Common CSV data quality issues
Specific checks with visible rules, not inferred data types or AI risk ratings.
Whitespace and placeholders
Find surrounding spaces and configured tokens such as NULL or N/A. A placeholder is reported, never replaced automatically.
Exact repeated rows
Identify later copies of identical whole rows. For removal, use the free exact duplicate remover.
Observed casing variants
Review values such as Active and active in the same column. Different capitalization is a review candidate, not an automatic error.
High missing-value rates
Find columns at or above your chosen missing percentage. The actual count and percentage remain visible.
Basic email-format findings
Select an email column to check non-missing values with a basic format rule. No external email-verification service is called.
Transparent rules
How RowDesk detects findings
Exact duplicates compare every original parsed cell, including capitalization and whitespace. Blank rows contain only empty or whitespace-only cells. Casing checks compare trimmed text within each column and report observed case variants.
Missing-value rates combine empty cells and your configured placeholders without counting a cell twice. Column warnings use the displayed threshold, not a guessed statistical risk level. The email check runs only where you explicitly enable it.
Malformed quotes or inconsistent field counts are reported as parsing errors instead of silently discarding cells. This is a focused CSV validation tool, not a validator for every destination-system requirement.
Before CRM import
Know what needs attention first
Missing identifiers, accidental spaces and repeated records can make an import harder to review or cause downstream duplicates. Checking the incoming file first gives you a concrete list of issues to investigate.
After reviewing the scan, use the free CSV cleaner for basic cleanup or RowDesk Clean Data for the connected preparation workflow. To check incoming records against an existing export, compare two CSV files.
Always verify required fields, allowed values and matching rules for the system receiving the data. Read the guide to cleaning CSV data before import for a practical preparation process.
A practical checklist
Check the file. Then choose the fix.
Use the findings to decide what to review, rather than treating every flag as an error.
- Confirm the structure
Check the delimiter, header, row count and column count against the expected export.
- Review missing data
Configure placeholders and inspect incomplete columns, especially required import fields.
- Inspect actual rows
Filter by issue and column. Decide whether duplicates, casing and whitespace need changes.
- Prepare a separate copy
Clean only the issues you understand, then verify the resulting file against the import requirements.
Browser-local privacy
Scan locally. Keep your source unchanged.
Parsing, scanning, filtering and summary generation happen in a browser worker. Raw file contents and cell samples are not sent to a server or analytics by this checker.
The optional email request is separate from the dataset. The summary download contains aggregate counts, column names and settings, not raw data rows. Closing the page clears the working dataset. Learn about RowDesk local processing and source-file safety.
Frequently asked questions
Data quality checks, explained.
Does the checker fix my CSV automatically?
No. It scans and reports findings without changing your source. You decide what needs cleanup. The optional download is a summary of findings and settings, not a cleaned dataset or a transformation audit report.
Why do issue counts add up to more than affected rows?
One row can contain empty cells, whitespace and several other findings. Counts use their stated units: rows, cells or columns. The unique affected-row total counts each row once, regardless of how many findings apply.
How are text casing issues detected?
The scan looks for the same trimmed text occurring with different capitalization within one column, then flags every occurrence in that variant group. It does not assume that all names should be title case or that uppercase values are wrong.
What counts as a missing value?
Empty and whitespace-only cells count as empty. Values equal to your configured placeholder tokens after trimming also count as missing. Tokens are case-sensitive. The default warning threshold is 50% missing values in a column; you can change it from 1% to 100%.
Does RowDesk verify email addresses?
The optional check uses the site's existing basic email-format validator on one column you select. It skips missing values and checks a simple address shape and length. It does not check inbox existence, DNS or deliverability, and a passing value is not proof of a valid address.
Can this validate CSV before import into my CRM?
It can find common formatting and missing-value issues before import, but it does not know your CRM schema, required fields, accepted values or business rules. Check those requirements separately. Findings are not a guarantee of import success.
Do I need an email address to see the findings?
No. The scan, column breakdown and affected-row pages are available before any email request. Only the optional summary download uses the shared first-export email gate, with optional unchecked marketing consent and no account requirement. Successful completion is remembered across all four free tools for 30 days in this browser.
What are the file and processing limits?
This lightweight tool accepts UTF-8 CSV files up to 25 MB, within a budget of 2 million cells and 500 columns. It scans all accepted rows locally, while displaying 20 affected rows and 6 data columns at a time. Device memory can also limit processing.
Related workflows