Select one or more CSV files. Files are processed entirely in the browser and never leave the device. Headers are read from the first row of each file.
Choose CSV files:
Remove All Files
Each row below represents one column from one source file. Uncheck a row to drop the column from the merged output. Edit the Output Column Name to rename a column or to combine columns under a shared name (for example, rename First Name and first_name both to "first_name" so they merge into one output column).
Auto-Map by Name Include All Exclude All
Each row groups columns that share the same normalized original header across every uploaded file. Editing the Output Column Name here updates all matching columns at once, so a header repeated across 12 files only needs to be renamed in one place. Placeholder "(mixed)" means individual rows in the main table currently differ - type a new value to overwrite them all.
Exact match (fastest, identical values only)
Fuzzy match (similarity threshold below)
No deduplication (just merge)
Fuzzy similarity threshold (0 to 100):
Select which output columns identify a duplicate. If none are selected, every output column is used.
Process & Merge Reset Advanced Settings
Combine non-key column values from duplicates (semicolon separated)
Trim whitespace before comparing
Case-sensitive matching
Ignore punctuation when matching
Add source-file column to output
Preview row count:
Drop any row whose values contain one or more of the listed terms (substring match across all output columns). Enter one term per line or separate by commas, e.g. X, Y, Z. The Case-sensitive matching checkbox above applies. Blacklisted rows are removed before deduplication, and the count appears under Results.
Blacklist values:
Rewrite URL or path values to truncate at any of these path segments, then dedupe the result. Enter comma-separated segments (case-insensitive). example.com/blog/article-title/ and example.com/blog/page/2/ both collapse to example.com/blog/. Combine with the Blacklist above to drop the entire section. Applied to the same output columns as URL / Path Normalization below (or the dedupe key columns by default).
Collapse path segments:
Transform URL or path values into shorter dedupe-friendly keys. Applied to the selected output columns before deduplication, so values that differ only in protocol, leading www, domain, query string, or file extension can collapse together. With the defaults, "http://outlook.com/sitemap.xml", "https://www.kaltask.com/sitemap_index.xml", and "https://google.com/sitemap.php" reduce to short titles built from the final path segment. The transformed value replaces the original in the matching column, so the export shows the normalized form.
Enable URL / path normalization
Pick which output columns to normalize. Leave all unchecked to default to the dedupe key columns.
Strip protocol (http://, https://, ftp://, etc.)
Strip leading "www."
Strip domain (keep only the path)
Strip query string and fragment (?, #)
Strip file extension (.pdf, .html, .xml, .php, etc.)
Path depth limit (parent directories to keep before the final segment): (-1 = unlimited, 0 = final segment only, 1 = keep one parent directory, etc.)
Replace "-" and "_" with spaces
Replace last "/" with:
Apply Title Case to the final value
Export CSV
-
Status: No files loaded.
Total input rows: 0
Blacklisted rows removed: 0
Unique output rows: 0
Duplicates removed: 0
Output columns: 0