Multilingual in Webflow: what is native
Webflow Localization is built in rather than bolted on, which matters more than it sounds. Each locale gets its own URL, its own metadata and its own hreflang tags without a plugin arbitrating, so search engines are told cleanly which version belongs to which audience.
Content can be translated per locale, and layout can be adjusted per locale — a heading that fits in English and not in German can be sized differently there without forking the page.
It is priced as an add-on per locale rather than included in the site plan. That is the single most common surprise, and it is a surprise only because the decision usually arrives after the build — see what Webflow hosting includes.
What actually takes the time
**Length.** German runs roughly a third longer than English; Ukrainian and Polish are close behind. A hero that fits perfectly in one breaks in the other, and every fix is a design decision rather than a translation one.
**Things that are not text.** Images with words baked in, PDFs, form labels, error messages, email confirmations, the cookie banner. These are always found late and each is a small task.
**Tone, not vocabulary.** A machine translation is grammatically fine and sells nothing. The parts that persuade — the objection handling, the calls to action — need someone who sells in that language, not someone who speaks it.
**Keeping them level.** Every future edit is now two edits. Without an agreed owner, the second language drifts out of date within a year, which is worse than not having it.
The decision to make before the build
Which pages actually need the second language. Almost never all of them. The pages that sell, plus contact — the archive, the older posts and the careers page frequently do not, and pretending otherwise doubles the maintenance forever.
The Goat Residential is a case built with this constraint in view: student housing chosen by students and paid for by everyone around them, running multilingual, with a page that speaks to three audiences without splitting in two. The languages multiply that problem rather than replacing it.
Deciding this before the build is what keeps it affordable. Retrofitting a second language onto pages designed for one is where the cost lands, and it lands as design work rather than translation.
Getting the search side right
One URL per language, always. Switching language with a cookie while the address stays the same means search engines only ever see one version, and the other might as well not exist.
hreflang tags in both directions, so each version points at the others. Webflow handles this natively, which removes the most common technical failure in multilingual sites.
And translate the metadata, not only the page. A translated page with an English title is a page that will not be found by the audience it was translated for — the same discipline as Webflow's SEO settings, applied twice.
Questions we get asked
Does Webflow support multilingual websites?
Yes, natively through Localization: each locale gets its own URL, metadata and hreflang tags, and layout can be adjusted per locale for sentences that change length. It is priced per locale as an add-on rather than included in the site plan.
How much extra work is a second language?
More than translation and less than a second site. The real work is length changes breaking layouts, non-text items like images with baked-in words and email confirmations, and the ongoing fact that every future edit becomes two edits.
Should every page be translated?
Almost never. Translate the pages that sell, plus contact. Archives, older posts and careers pages usually do not earn their ongoing maintenance in a second language, and translating everything doubles the upkeep permanently.