Translate selected columns in a CSV file and keep everything else untouched. IDs, dates, prices, and URLs stay byte for byte, only the columns you mark get rewritten. Supports 100+ languages and large bulk uploads.
Touch only the cells you choose
Drop your CSV in, pick which column holds the text, choose your target language, and download a CSV that has the same rows and shape with just those cells translated.
Product catalogs, support macros, FAQ banks, email templates, and survey responses all live in CSV files. Translating them by hand means opening a spreadsheet, copying one cell, pasting it into a translator, then pasting the result back. For a file with five hundred rows and three text columns, that is fifteen hundred round trips, and human attention will slip at some point. A CSV translator runs the whole job in one upload with zero copy paste.
The key idea is column selection. Your CSV almost certainly mixes data that should never change, such as product IDs, SKUs, prices, dates, and image URLs, with data that should be localized, such as product names, descriptions, marketing taglines, and support responses. A column aware translator only rewrites the cells in the columns you mark. Everything else passes through unchanged so the downstream importer that ingests the file still works on the first try.
You get a new CSV with the same row count, the same column order, the same delimiter, and the same encoding. Headers stay intact unless you ask otherwise. Quoted cells with commas, embedded line breaks, and special characters are preserved. The result is a drop in replacement for the source file in your target language, ready for re upload to your store, CRM, ticketing system, or analytics pipeline.
Upload your CSV
Drag and drop or pick the file from disk. The first row is treated as the header by default, so you can address columns by name. UTF-8 files work best. If the file uses a non standard delimiter, the parser will detect it from the structure.
Pick the columns to translate
Tick the columns that hold human readable text, such as title, description, or body. Leave columns like id, sku, price, created_at, or image_url unchecked. Only checked columns are sent to the translation engine, everything else streams through untouched.
Choose target language and download
Select one of the supported target languages, start the job, and watch the progress bar. When it finishes, download the translated CSV. The row count and shape match the input exactly so it imports cleanly back into your spreadsheet, store admin, or database.
A Shopify or WooCommerce export lists every product with id, sku, price, image URL, title, and description. Translate only the title and description columns. The rest stays intact and the file imports back into the same store as a new locale without any manual cleanup or risk of breaking your inventory tracking.
Helpdesk tools like Zendesk and Intercom let you export macros as CSV. Each row has a name, a category, and the reply body. Translate the body column to roll out the same canned responses in a new language without rebuilding the macro library by hand or losing categorization metadata.
Open ended survey answers come back in many languages. Export the responses as CSV, translate the open response column into a single shared language, and you can run sentiment analysis or theme clustering across the whole dataset without losing the original respondent IDs or timestamps.
Marketing teams keep transactional and lifecycle emails in CSV form with columns for template key, subject, preheader, and body. Translate subject, preheader, and body in a single pass while the template key stays the same so your sending platform can match each row back to the right trigger automatically.
Upload your CSV file, pick which column or columns hold the text to translate, choose a target language, and run the job. Every row in the selected columns is translated as a separate string while every other column, every header, and the row order stay exactly as you uploaded them. The output is a new CSV with the same shape that you can drop straight back into your spreadsheet or database.
No. Non target columns are passed through byte for byte. Numeric IDs, dates, SKU codes, prices, URLs, and any other untouched fields keep their exact original value. Only the cells inside the columns you marked for translation get rewritten. The column order, header row, delimiter, and quoting style of the original file are preserved in the output.
Cell contents with commas, quotes, or line breaks are handled correctly. The parser respects RFC 4180 CSV quoting, so a product description that contains a comma will not split into two cells, and a cell with double quotes around a phrase will keep that quoting in the output. UTF-8 encoding is preserved so accented characters and non Latin scripts come through cleanly.
Bulk jobs handle thousands of rows in a single upload. Larger files are processed in batches in the background, and you can watch progress while it runs. For very large files, splitting into chunks of a few thousand rows each tends to give the most responsive experience, but the tool will still finish a single big file as long as your browser session stays open.
Yes. Run the same source file through the bulk translator once per target language and you get back one CSV per language, each with the same row order and the same untouched columns. Many teams keep one master English CSV and generate a Spanish, French, German, and Japanese copy from it for product catalogs, support macros, or marketing assets.
Yes. The first row is treated as a header by default, which is how almost every CSV in the wild is structured. Header labels themselves are left untranslated so any downstream importer that expects exact column names still works. If you want the headers translated too, you can select the header row as part of the input and it will be translated like any other row.