Weglot and TranslatePress differ less in where they store translations than in who they expect to do the translating. TranslatePress hands you a visual editor, writes the result into your own WordPress database, and assumes your team will work through it. Weglot machine-translates the whole site first and hosts the output, so that your team mostly does not have to. Price, performance and what you keep on cancellation all follow from that one difference.
One correction before the comparison, because it is repeated widely enough to change decisions: several ranking articles state that TranslatePress requires a paid add-on to get proper multilingual SEO. Its own pricing page lists the SEO Pack add-on as included on every paid tier.
TranslatePress is built around a front-end visual editor. You see the page as a visitor does, click a string, and translate it in place, with the result stored in your WordPress database. AI translation is included on every tier as a first pass, metered in words, but the product's centre of gravity is a person working through the site with a clear view of context.
Weglot inverts that. It reads your rendered HTML, applies a machine-translation pass across the whole site immediately, hosts the output on its own servers, and offers a visual editor for refinement afterwards. Coverage arrives first and editing is optional; Weglot states that two-thirds of its customers never edit their translations at all, which is the company's own figure and a fair description of the intended workflow.
That difference explains why comparisons of the two so often talk past each other. TranslatePress is priced and designed for a team with translation capacity. Weglot is priced and designed for a team without one. Neither is a compromise version of the other.
No, not on its paid plans. TranslatePress lists the SEO Pack add-on as included across all three paid tiers, alongside AI-led translation, URL slug translation, a language switcher and Google Translate integration. Several comparison articles say otherwise, and the claim appears to have been copied forward without a date attached.
The distinction worth keeping straight is between the free plugin and the paid tiers. If you are running TranslatePress free, check the current feature list for what SEO handling is included before assuming. If you are pricing a paid licence, the SEO functionality is in the box, and any comparison telling you to budget separately for it is out of date.
TranslatePress is unusually direct about this, and the sentence is worth quoting because it is the crux of the comparison: "All the translations are stored locally in your server's database. You won't lose any of your translations if you decide to not renew your license."
Weglot's position is the mirror image. Translations are hosted on Weglot's servers, an export is available, and the translated pages stop serving when the subscription ends. That is a property of the hosted model rather than a flaw, and it is the same trade any rendered-HTML service makes. It buys you speed and no maintenance, and it costs you a dependency.
The practical question is not which is better but which position you want to occupy in three years. A site that expects to run multilingual indefinitely on a fixed budget is in a different place from one testing two markets for a season.
TranslatePress is a flat annual licence: Personal at €99 for one site with 50,000 AI words, Business at €199 for three sites with 200,000 AI words, and Developer at €349 for unlimited sites with 500,000 AI words. Additional AI word packs start at €24 per 100,000 words. Weglot is a subscription scaling with translated words multiplied by the number of languages, starting from a free tier of 2,000 words in one language and paid plans from about €15 per month billed annually.
The shapes cross over predictably. TranslatePress's cost is flat until the AI word allowance runs out, at which point it steps by a known top-up price. Weglot's cost tracks both catalogue size and language count together, so adding a fifth language moves the bill even if the content did not change. A large catalogue in several languages generally lands cheaper on the flat licence; a small site in one or two languages can be comparable on either. The breakdown of what Weglot actually costs to scale works the tiers through in detail.
One structural note: TranslatePress is WordPress-only, while Weglot also runs on Shopify, Webflow and other platforms. If replatforming is plausible, that is a real consideration rather than a spec-sheet line.
A store consumes translation allowances faster than its page count suggests, because product variations and attributes multiply strings and because the catalogue keeps turning over. On TranslatePress that shows up as AI words consumed; on Weglot it shows up as movement between tiers. Either way, the recurring cost of a store is driven by how often you publish, not by how big the site was on launch day.
The other WooCommerce-specific check is transactional email. Content that never renders to a public page, such as order and shipping notifications, sits outside what a rendered-HTML approach can reach. A database-backed plugin has a structural advantage there. It is worth confirming against current documentation for whichever you choose, because it is the gap a customer meets immediately after paying you.
Choose TranslatePress if you want to own your translations, pay a predictable annual fee, and have somebody who will do the editing. Choose Weglot if you want the site live in new languages quickly with minimal ongoing effort, and you are comfortable with the translated pages depending on an active subscription. Both are competent; the question is which resource you have more of, time or budget.
Both models assume someone is running it. Somebody reviews the machine output, keeps the glossary current, checks that this week's products came through in all languages, and spots the half-translated page before a customer does. At two languages that is minor. At six, with weekly drops, it is a job that never got a headcount.
The managed alternative moves that work rather than the tool: you add a product on your source store and it is live in every language without anyone touching it. For what that is worth when it runs properly, in live 50/50 A/B tests across more than 40 merchants and 50 site and country combinations, translated stores lifted conversion by an average of 40% at launch, with individual results from 15% to 100%, measured against an untranslated experience in the same market rather than as a permanent multiplier. The guide to ecommerce translation covers the four routes, and managed ecommerce localisation is the one where the upkeep is not yours.
There is a free plugin and three paid tiers, from €99 to €349 per year. The paid tiers include the SEO Pack add-on, AI translation word allowances, URL slug translation and multi-site licences. Check the current free-plugin feature list before assuming what it covers.
Yes. TranslatePress states that translations are stored locally in your server's database and that you will not lose them if you choose not to renew your licence. Weglot hosts translations on its own servers instead, where the pages stop serving when the subscription ends.
Both serve translated pages on dedicated URLs with hreflang, so both can rank. TranslatePress includes its SEO Pack add-on on all paid tiers, despite several comparisons claiming it costs extra. Weglot automates the same fields, with URL translation on its higher plans.
Yes, and being database-backed it has a structural advantage for content that never renders to a public page, such as order and shipping emails. Size it against your product variations rather than your page count, since variations multiply the string volume.
Weglot offers an export, but moving that content into another plugin's data structure is not automatic. If you think you may switch later, starting with translations in your own database avoids the migration.
It depends on catalogue size and language count. TranslatePress is flat until its AI word allowance runs out, then tops up at €24 per 100,000 words. Weglot scales with words multiplied by languages, so a fifth language raises the bill even with no new content.
See your own storefront in the languages you are considering, with the ongoing model stated up front rather than left to your team.
See your store localised