How to turn a CSV into a Google Shopping feed

Start with a clean export

Export a CSV with one row per product (or per variant) and consistent column headers. It doesn't need Google's exact field names — you'll map those next — but it should have a unique ID, price, availability, an image URL, and a product page link per row.

Map columns to Google's fields

Match your columns to Google's required attributes: id, title, description, link, image_link, availability, price, plus brand and GTIN where present. This mapping is where a feed tool saves you from hand-editing the file every time.

Fix the data problems CSVs always have

Raw exports are messy: promo text in titles, blank GTINs, prices without currency, inconsistent availability values. Clean these with rules — rebuild titles, set identifier_exists where there's no GTIN, normalize availability — instead of fixing the spreadsheet by hand.

Automate the refresh

A one-time upload goes stale immediately. Connect the CSV so it re-imports on a schedule (or re-upload from the same location), and have the feed regenerate automatically so prices and stock stay current.

Can I upload a plain CSV straight to Google Merchant Center?

You can, if the CSV already uses Google's exact column names and clean values. In practice most exports need mapping and cleanup first, which is why a feed tool sits between your CSV and Google.

How do I keep a CSV-based feed up to date?

Don't rely on one-off uploads. Connect the CSV source so it re-imports on a schedule and regenerate the feed automatically, so price and availability never go stale.

My CSV has no GTINs — is that a problem?

Only if you leave the field blank. If the products genuinely have no barcode, set identifier_exists to false in the feed mapping and rely on brand, MPN and a strong title.