Post in 3 Languages: Claude + Make

A community clinic in a mixed-language neighborhood writes health-tip posts in English. Half their patients read Spanish or Vietnamese more comfortably, so the posts miss the people who need them most. Translating each one by hand never fits the week.

This build drafts all three at once. You type a topic into a sheet. Claude writes the post in your three languages, and each version lands as its own WordPress draft. A bilingual staffer or volunteer reviews each before it goes live, so a person still checks the nuance, the machine just removes the blank page in every language.

It is the same shape as our spreadsheet-to-WordPress build, with Claude producing three versions instead of one. We wire it in Make. A note up front: machine translation needs a human reviewer, which is exactly why these land as drafts.

What does this workflow actually do?

In one line: one topic becomes three drafts. Make watches your sheet. When you add a topic, Claude writes the post in each language you name, and Make creates a WordPress draft for each. Nothing publishes itself; a reviewer approves each language.

A few real cases, none of them the usual ones:

• The community clinic above, finally reaching patients in the languages they read.
• A folk-music festival posting artist announcements in the three languages of its regular crowd.
• A marine-supply shop on a border coast writing product notes for both sides of the water.
• A family-history researcher publishing findings in the languages their scattered relatives actually speak.

Writing the post once is the work. Re-writing it in two more languages is the wall most people hit. Claude clears the wall; a human still checks the result.

Why use Make if Claude can already write?

Worth asking, and the answer has an extra layer here. Claude is good at drafting and translating. It is not built to watch a sheet, loop through three languages, and create three WordPress drafts in the right place. That is plumbing, and Make handles plumbing.

Make does the no-judgment steps: catching the topic, passing it along, creating each draft. Claude does the language work. And you, the human, do the final judgment that neither tool should: checking that the translation reads right and respects the reader. Make is the line, Claude is the drafter, you are the editor. That third role is not optional with translation.

What do you need before you start?

• A Make account. The free 1,000 operations a month is plenty.
• A Claude API key from the console, not the chat app. See how to use Claude.
• A Google account with a sheet that has a Topic column.
• A WordPress site with an application password, and ideally a reviewer for each language.

Two Make words. A scenario is the whole sheet-to-WordPress automation. A module is one box. Three boxes, one scenario.

How does the workflow work, step by step?

Three modules, left to right:

Module	App	What it does
1. Trigger	Google Sheets	Fires when you add a topic
2. Brain	Claude	Writes the post in each language you name
3. Output	WordPress	Creates a draft for each language version

Step 1: Set up the topic sheet

Make a Google Sheet with a Topic column. In Make add Google Sheets, Watch New Rows, connect your account, and point it at the sheet. This is the trigger: a new topic starts the run.

Step 2: Let Claude write all three versions

Add a Claude module, Create a Prompt, paste your key, pick Sonnet. Name your languages in the prompt:

Write a ~500-word blog post on this topic, then translate it.
Topic: {{Topic from Step 1}}

Produce three versions: English, Spanish, and Vietnamese. Keep the
meaning and tone consistent. Label each version with its language
so they are easy to separate. Plain, friendly language throughout.

Swap in whichever languages you need. Labeling each version lets the next step split them into separate drafts.

Step 3: Create a draft for each language

Add WordPress, Create a post, connect with your application password, and map each labeled version to its own draft, with Status set to Draft. The Draft setting is non-negotiable here: translation is exactly the place you want a human to read before anything publishes.

How do you run it and check the result?

Click Run once. Make reads the topic, Claude writes the three versions, and WordPress fills with three drafts. When I tested this, the English and Spanish drafts were strong and the third still wanted a native speaker’s eye on a couple of phrases, which is the whole reason a reviewer stays in the loop. Tune the prompt, then turn the scenario on.

After that, one topic produces a draft in every language you serve. The blank page is gone in all three; the human judgment that translation needs stays exactly where it belongs, on the review.

What does this cost to run?

Piece	Free tier	If you outgrow it
Make	1,000 operations/month free	Core plan from about $9/month
Claude API	Pay per use	Three ~500-word drafts on Sonnet cost a couple of cents
Google Sheets	Free	Free
WordPress	Your existing site	No extra cost

Each topic uses a few Make operations and a couple of cents of Claude. The cost worth weighing is reviewer time, which is the point: a person should read each language. More on tiers in our Make guide.

What can go wrong, and how do you avoid it?

• It publishes unreviewed translations. Only if Status is not Draft. Keep it on Draft so a human checks every language.
• The versions run together. Keep the language labels in the prompt so Step 3 can split them.
• Idioms translate awkwardly. This is normal and why you have a reviewer. Ask Claude to favor clear, literal phrasing over clever wording.
• The API key is rejected. A chat login is not an API key. Generate one in the console with billing set up.

The same three-box pattern runs many jobs. See Make AI scenarios for more.

How do you build this in Zapier or n8n instead?

Same three jobs, different names.

Job	Make	Zapier	n8n
Catch a topic	Watch New Rows	New Spreadsheet Row trigger	Google Sheets Trigger node
Write the versions	Claude module	Claude (Anthropic) action	Anthropic node
Make the drafts	Create a post	WordPress Create Post	WordPress node

Make and Zapier are easiest to start with. n8n suits self-hosting. Compare them in Zapier vs Make vs n8n.

Frequently asked questions

Is machine translation good enough to publish?

Good enough to draft, not to publish blind. That is why each version lands as a draft for a human reviewer. For nuance-critical content, use a professional.

How many languages can it do?

As many as you name in the prompt, though more languages mean longer outputs and more review. Start with the two or three your readers actually use.

Can it set the right language tag in WordPress?

With a multilingual plugin, yes, by mapping the language to the plugin’s field. Start simple with separate drafts and add that later.

Which model should I use?

Claude Sonnet handles this well and cheaply. For sensitive content, a stronger model plus a human reviewer is the safer pairing.

Is the chat app the same as the API?

Same models, different door. Make needs an API key, not the chat login.

Primary sources

• Creating a scenario (Make Help)
• Claude API overview (Anthropic)
• WordPress REST API
• Google Sheets basics (Google)
• Machine translation — Grokipedia

Last reviewed: May 2026. Make, Claude, and WordPress change their interfaces; check the official pages for exact button names.

You may also like

• Draft WordPress posts from a spreadsheet with Claude + Make
• Turn a YouTube video into a WordPress draft with Claude
• Make.com Complete Beginner’s Guide
• Make AI Scenarios: automations that save hours
• How to use Claude AI
• Zapier vs Make vs n8n: which to start with
• The AI Tools Directory