Writing and content decision

Weglot

A capable developer can build a useful, narrower replacement (automatic translation + serving under language URLs and a basic editor) in about a week; the paid product’s polished dashboard, enterprise features, and brand-backed services are the main reasons to keep the commercial option for larger teams.

Visit website
You pay

$17/mo

$204/yr

Read off the official pricing page.

You’d pay instead

$50one-off30 h to build

$50/mo5 h/mo upkeep

On cash alone, building overtakes the subscription at 4 seats.

No open-source build does this yet

Nothing published replaces this one, so a replacement starts from an empty file. Here is what it would have to cover.

What a replacement has to do

  • Detect site content → send strings to a translation API → store translated HTML/strings → serve language-specific URLs and inject language switcher → provide a simple dashboard for editing translations and exporting

What it still won’t have

  • Hosted dashboard with polished visual editor and UX
  • Custom-brand translation model and training tuned by provider
  • Managed uptime, SLA, and enterprise features (SSO, dedicated proxy)
  • Built-in multilingual SEO automation and integrations across many platforms
  • Professional support, managed translators, and premium workflows

What remains hard

  • Brand trust111,368+ global brands rely on Weglot to translate their websites
  • Brand trustWebsite translation leader on G2 for 4 years in a row
Read the build prompt

First-year cost

Keep paying

Paying is—cheaper in year one.

On cash alone, building overtakes the subscription at 4 seats.

Paid seatsseats

Money you would actually spend

Keep paying
—

Subscription price × seats × 12

Build it
—

AI build —APIs + hosting —

Time you would spend

—

—

What you would spend

What we assumed

The verdict above measures whether you could build it. This one is only about money.

Runnable build prompt

Not run yet
Build a minimal self-hosted website translation service using Node.js (Express), PostgreSQL, and a translation API (DeepL or Google). Include: (1) a crawler to extract translatable strings and metadata from a provided site root; (2) translation jobs that call a configurable MT provider and store results; (3) an HTTP layer that serves translated content under /{lang}/ paths and emits hreflang tags; (4) a lightweight dashboard to list and edit translations (text-only editor), and a language switcher JS snippet to inject into pages; (5) export/import of translations (CSV/JSON). Out of scope: WYSIWYG visual editor, custom model training, enterprise SSO, billing. Require robust error handling, retry/backoff for API calls, tests for crawler, translation job, serving, and DB migrations.
How we checked3 sources · 2/3 runs agreed · evidence score 84

How the score was reached

  • Build verdict base78
  • 3 cited sources+3
  • Price verified on pricing page+3
  • Evidence score84

The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.

How scoring works →

Cited sources · 3

Every page the run actually retrieved.

Integrity checks

What held up, and what did not.

✓ Price read off the page! 2 of 3 runs agreed; the verdict is the majority✓ Citations limited to fetched pages! 2 moats quoted from the page