n8n automation

Automated WooCommerce product translation with AI

An n8n workflow takes untranslated products from a queue, and AI translates their titles, descriptions, attributes and SEO meta into the languages you choose. Each result is checked for completeness and length before it reaches your store.

WooCommerce · WordPress REST API · OpenAI · Google Sheets

In short

Automated WooCommerce product translation with AI is an n8n workflow that prepares language versions of your products: the title, short and long description, attributes, meta title and meta description. It takes products to translate from a queue in a sheet or straight from WooCommerce, translates them in batches and saves them through the REST API in the way your multilingual plugin stores them. It’s built for stores entering new markets, or with hundreds of products, that don’t want to translate by hand or paste text from a translation tool into the admin panel. Every translation is checked for completeness and length, and everything goes into a log.

Workflow

This is the workflow in n8n

A simplified, interactive version of our workflow. Drag the nodes and run it to watch the data flow.

Drag the nodes · click “Run”

Execution log

    WooCommerce translation: Product queue → needs translating? → AI translates fields and SEO meta → completeness and length checks → saved to the store and logged

    Screenshot from the build

    The full workflow in our n8n

    A real, unedited screenshot from the n8n editor. The interactive version above is simplified; here you can see every node, branch and error path.

    Full size
    n8n workflow: AI WooCommerce product translation
    Workflow in n8n. Built by Bartłomiej Chwajoł.

    Sound familiar?

    What this solves

    You copy descriptions into a translator and back again, one product at a time.

    The workflow picks up the next products from the queue, translates every field in one model request and saves the result to your store. You get a log of what changed.

    Machine translation mangles names, units and trade terms.

    The prompt carries your rules: brand and model names stay as they are, units stay consistent and trade terms have agreed equivalents. The model returns JSON with every field kept separate, so nothing gets mixed up.

    After translation, meta titles and descriptions come out too long or empty.

    The workflow checks that every field is filled and that the meta fits your agreed limits. If not, it sends the model a second, narrower request instead of saving bad text.

    How it works

    Step by step

    1. 01

      Schedule

      A Schedule Trigger runs translations at set times, for example overnight when the store is quieter.

    2. 02

      Product queue

      The workflow reads the queue from a sheet or a product list from the WooCommerce REST API. A Limit node sets the batch size and a loop handles products one by one.

    3. 03

      Does it need translating?

      An IF node skips products that already have a translation in that language, so you never pay for the same work twice.

    4. 04

      Prompt and translation

      Code builds the prompt from the product data and your rules, and an OpenAI model translates the title, descriptions, attributes and SEO meta in a single pass.

    5. 05

      Parsing and checks

      The response is parsed from JSON. One IF checks that all fields are present, another that the meta title and description fit the limits.

    6. 06

      A second attempt if needed

      If something is off, the model gets a second request with a specific note, such as “shorten the meta description”.

    7. 07

      Save and log

      The translation is saved to the store through the WooCommerce or WordPress REST API, and the sheet gets an entry: product, language, status and any error.

    Implementation details

    A language version of your store is more than translated descriptions. The attributes used in filters matter too, as does the meta data people see in search results. This workflow translates all of those fields at once, in one format, and saves them where your store expects to find them.

    In step with your multilingual plugin. WooCommerce on its own doesn’t keep content in several languages. Plugins such as WPML or Polylang do, and each stores translations a little differently. During setup we check your configuration and set up saving through the REST API so the translation is linked to the original rather than created as a separate, unrelated product.

    Checks without reading everything. The model returns JSON with separate fields. The workflow checks that none is empty and that the meta title and description fit the limits you set. If something fails, a second, narrower attempt follows. If that fails too, the product goes into the log with an error description and waits for a person, instead of landing in the store half-translated.

    Glossary and brand voice. The prompt lists the words we never translate (brand, product lines, technology names), preferred equivalents for trade terms, and your tone. At the start you’ll usually review a sample from each category with someone who knows the language, and the feedback goes into the prompt before the workflow takes on the whole catalogue.

    Safe for your store. The first runs are best done on a staging copy or a handful of products. The workflow only writes language versions and never touches the original, and its API access covers only what it needs.

    If you add products across several channels, see automated product listing on WooCommerce, Allegro and Erli. We compare platforms in WooCommerce vs Shopify, and if the new language version needs to win traffic from Google, our SEO and GEO service can help.

    01

    Integrations

    • n8n
    • WooCommerce REST API
    • WordPress REST API
    • OpenAI
    • Google Sheets
    02

    What you need

    • A WooCommerce store with a multilingual plugin and languages set up
    • WooCommerce REST API keys or a WordPress application password with write access
    • An OpenAI API key on your own account
    • A sheet with the product queue (optional, the workflow can read from the store directly)
    • A brand glossary: what not to translate, preferred terms, tone
    • An n8n instance on your server or in n8n Cloud
    03

    How we tailor it

    • Any language pairs, with separate rules for each market
    • Translation of categories and global attributes too
    • Approval of translations before saving, for example in the sheet
    • Re-translation when the source-language description changes
    • A different model, such as Claude or Gemini, for selected languages

    FAQ

    Frequently asked questions

    Is AI translation enough, or do I need a translator?

    For most product descriptions AI gives you a solid base, especially with a refined prompt and a brand glossary. For legal or medical text, or your most important sales pages, we recommend a review by someone fluent in the language. The workflow can hold those products back for approval.

    Which multilingual plugins does it work with?

    Ones that let you save translations through the REST API, such as WPML (with its WooCommerce Multilingual add-on) or a suitable edition of Polylang. Before going live we test your setup on a staging copy of the store.

    Are SEO meta fields translated as well?

    Yes. The meta title and description are part of the same model request and get their own length check. How they’re saved depends on your SEO plugin, such as Yoast SEO or Rank Math, and we set that up during the project.

    What happens when I change the source-language description?

    In the basic version, the product goes back into the queue when you flag it in the sheet. We can also add change detection so the workflow knows by itself which translations are out of date.

    How long does translating the whole catalogue take?

    It depends on the number of products, description length, the number of languages and the model’s API limits. The workflow runs in batches in the background, so even a large catalogue goes through without blocking the store. We set the pace so you have time to review the first results.

    Get a quote

    Want this in your business? Tell us what it should do.

    Describe your process and the tools you use. We reply within one business day with an implementation plan.

    What do you need?

    Goal, deadline, links to inspiration — the more specifics, the more accurate the quote.

    Other builds

    More n8n automations

    1. 01 AI songs for YouTube OpenRouter · fal.ai · YouTube Data API · Telegram An n8n workflow takes an idea from the queue, writes the lyrics, generates a song with vocals, cover art and a video, then sends you a preview on Telegram. Nothing reaches YouTube until you approve it.
    2. 02 Allegro listing optimisation with AI Allegro REST API · OpenAI · Google Sheets · Gmail Every hour, an n8n workflow takes the next offers from a sheet, pulls their data and category parameters from the Allegro API, and AI writes a new title and description. Changes reach Allegro only after validation, and every offer is backed up first.
    3. 03 AI email assistant on Telegram IMAP · OpenAI · Telegram Bot API · SMTP An n8n workflow reads your inbox, filters out spam and sends the emails that matter to Telegram. You record a voice note with your answer, AI turns it into a finished email, and one button sends, revises or discards it.
    4. 04 AI-personalised email & SMS n8n · OpenAI / OpenRouter · Google Sheets · SMS gateway An n8n workflow pulls contacts from a Google Sheet, writes each person their own message, sends the email and, for mobile numbers, an SMS too. The sheet shows you straight away what went out and what was delivered.
    5. 05 Multichannel product listing WooCommerce REST API · Allegro API · Erli API · OpenAI You enter the product data once, and an n8n workflow picks the category on each platform, writes the copy, prepares the images and creates the product in WooCommerce plus offers on Allegro and Erli. At the end you get one record with the result from every channel.
    6. 06 AI SEO landing pages for WordPress Google Search Console · OpenRouter · WordPress REST API · Google Sheets An n8n workflow takes a keyword from a sheet, analyses the search results, plans the structure and writes a landing page with links to your existing pages. The finished page lands in WordPress as a draft for you to approve.