A client came to me in late 2023 with a Weglot bill that had jumped from £29/month to £199/month overnight. Not because her traffic exploded. Because she'd added a product FAQ section and crossed the 50,000-word threshold. The site hadn't grown meaningfully. The bill had nearly septupled.
That's the Weglot trap. It's a genuinely good product. Fast, JavaScript-based, easy to set up on any CMS. But the pricing model is word-count-based, and word count is one of the most unpredictable variables on a content-heavy site. When I'm scoping multilingual builds at Seahawk now, Weglot is rarely my first recommendation anymore. Not because it's bad. Because there are options that give agency owners and freelancers more predictability, more control, and in most cases, a lighter invoice.
Here's what I've actually used, what I'd reach for in different scenarios, and one or two tools I'd avoid despite the hype.
---
Why Weglot's Pricing Model Becomes a Problem at Scale
Weglot's free tier covers 2,000 words and one language. The Starter plan (around £17/month at current rates) takes you to 10,000 words. Sounds generous until you realise a single WooCommerce product with a description, meta fields, and a few attributes can run to 300 words. Add 50 products and you're already halfway to the cap.
The Weglot pricing page currently shows the Business plan at £79/month for 200,000 words across 5 languages. For a mid-size e-commerce build, that's not unreasonable. But the model punishes content growth, and no client wants to get a call from their developer saying "we need to upgrade your translation plan because you wrote more blog posts."
The other issue is portability. Weglot stores translations on its own servers. If you cancel, you lose everything unless you've exported manually. For agency white-label builds, that's a liability.
---
WPML: The Old Guard That Still Earns Its Keep
I've used WPML on probably 400+ sites over the years. It's not sexy. The interface hasn't changed dramatically since 2015. But it does something Weglot doesn't: it stores all translations in your own database.
That matters. A lot. When a client moves hosts, switches agencies, or decides to self-manage, nothing is lost. The translations live in WordPress like any other content.
WPML's Multilingual Blog plan starts at $39/year. The full CMS plan, which covers custom post types and taxonomies, is $99/year. For agency builds, you'd want the Agency plan at $199/year for unlimited sites. Do the maths on that against Weglot's per-site monthly pricing.
Where WPML struggles
The learning curve is real. Setting up string translation, configuring URL structures, and getting WooCommerce Multilingual to behave properly takes time if you're new to it. I've seen junior developers spend a full day just getting the language switcher to display correctly in a custom theme.
Performance is also a fair criticism. WPML adds database queries. On a poorly optimised server it shows. But on a properly cached WordPress install with a decent host, I've never had it cause meaningful page speed issues.
---
TranslatePress: The One I Recommend Most Often Now
Honestly, TranslatePress is where I send most clients these days. It's a WordPress-native plugin with a visual, front-end translation editor. You see the page as a visitor would, click on any text, and type your translation directly. No hunting through string tables.
The free version on WordPress.org covers one additional language with human or DeepL-powered automatic translation. The Personal plan at €9/month (billed annually) adds unlimited languages. Business at €16/month adds SEO translation, which matters if the client cares about ranking in the target language.
Seahawk used TranslatePress on a travel blog build last year. The client wanted English, German, and French. We had all three languages live and indexed in Google Search Console within a week. The client could handle ongoing translation herself using the visual editor without calling us every time she added a page. That's the real win.
DeepL integration
TranslatePress connects natively to DeepL, which at this point produces genuinely better automatic translations than Google Translate for European languages. You set it up, run automatic translation on existing content, then go back and polish the 15% that DeepL gets slightly wrong. Much faster than translating from scratch.
---
Polylang: Free, Flexible, and Underrated
Polylang gets dismissed because the free version is limited. But Polylang Pro at €99/year is a serious multilingual setup that competes directly with WPML at a lower price point.
Like WPML, it stores translations in the WordPress database. Like TranslatePress, it's relatively approachable. The string translation module, which you need for theme strings and plugin text, is a paid add-on, but it's included in the Pro bundle.
Where I'd reach for Polylang over WPML: simpler sites, budget-conscious clients, or when I want a lighter plugin footprint. WPML does more but also installs more. Polylang Pro is leaner.
Where I wouldn't use it: complex WooCommerce multilingual builds. WPML's dedicated WooCommerce Multilingual plugin is genuinely better for that use case.
---
Loco Translate: The Developer's Utility Knife
This one's not a Weglot alternative in the traditional sense. Loco Translate is a translation management tool for WordPress themes and plugins, not a multilingual content plugin.
But I'm including it because a lot of people reach for Weglot when what they actually need is just to translate their theme strings and plugin UI labels into another language. For that narrow job, Loco Translate is free, fast, and requires no subscription whatsoever.
I used it last month on a client site where the theme's checkout page had hardcoded English strings that weren't going through the standard WordPress translation system. Loco let me create a custom .po file and override those strings in about 20 minutes. No plugin upgrade. No monthly fee.
---
Linguise: The Closest Like-for-Like Weglot Swap
If you want Weglot's deployment model (JavaScript-based, works on any CMS, no WordPress dependency) but at a lower price, Linguise is worth a serious look.
It uses a front-end script similar to Weglot's approach but prices by page views rather than word count. The Starter plan is around $14/month for 200,000 page views. That structure suits content-heavy sites much better than word-count billing.
I haven't used Linguise on as many projects as the others on this list. Seahawk tested it on a Shopify build in early 2024 and the automatic translation quality was solid. The exclusion rules (telling it to skip certain CSS classes or HTML elements from translation) worked cleanly. The main gap vs Weglot is a smaller ecosystem and less mature documentation.
---
What to Actually Choose: A Practical Decision Framework
Here's how I think about it when a new multilingual project lands:
- WordPress site, client manages content, budget under £200/year: TranslatePress Personal or Polylang Pro.
- WordPress site, WooCommerce, multiple currencies, complex structure: WPML CMS + WooCommerce Multilingual. Yes, it's more expensive and more complex. It handles the edge cases.
- Non-WordPress CMS (Shopify, Webflow, custom build) or need to be CMS-agnostic: Linguise over Weglot if budget is a concern. Weglot if the client has the budget and wants the most polished dashboard.
- Just need to translate theme and plugin strings, no multilingual routing needed: Loco Translate, free, done.
- Agency building white-label sites for resale: WPML Agency or TranslatePress Agency. Avoid Weglot's per-site monthly model unless you're billing clients directly for it.
A few things I always check before committing to any tool:
- Does it generate language-specific URLs (e.g.
/fr/subfolders orfr.domain.comsubdomains)? Google needs this for proper hreflang indexing. - Does it translate SEO meta fields (Yoast, RankMath, AIOSEO)? Shocking how many plugins miss this.
- Where are translations stored? Plugin's own server = risk. Your database = safe.
- What happens if you cancel? Test the export flow before recommending to a client.
---
FAQ
Is Weglot still worth using in 2026?
For non-WordPress sites or quick setups where ease of installation matters more than price, yes. Weglot still has the smoothest onboarding of any translation tool I've used. But for WordPress specifically, TranslatePress or WPML will almost always be cheaper over a 2-year horizon.
Does TranslatePress affect SEO?
Done correctly, no. TranslatePress generates proper translated URLs with hreflang tags and lets you translate Yoast and RankMath SEO fields directly. I've had sites rank well in German and French using it. The key is making sure the SEO pack add-on is active.
Can I migrate from Weglot to another plugin without losing translations?
Technically yes, but it's manual work. Weglot lets you export translations as a CSV. You'd then need to import those into your new tool's format, which varies. WPML and TranslatePress both have import options but neither is a one-click migration from Weglot. Budget a few hours for a medium-sized site.
Is DeepL accurate enough to use without human review?
For common European language pairs (English to French, German, Spanish, Italian), DeepL is genuinely impressive. I'd still recommend a native speaker reviewing anything on a commercial page, especially product descriptions or anything with brand voice. For blog posts and FAQ content, 80% of the automatic output is usually publishable as-is.
What about machine translation quality in WPML?
WPML integrates with DeepL, Google Translate, and Microsoft Translator. DeepL is the best of the three for most European languages, same conclusion as with TranslatePress. The quality difference between WPML using DeepL and Weglot using its own translation engine is negligible in my experience.
---
Multilingual builds are one of those areas where the "obvious" choice (Weglot, because you've seen it recommended everywhere) isn't always the right one. The right choice depends on your CMS, your client's content volume, whether translations need to live in the database, and frankly, how much you want to spend over a two or three year period. None of the tools above are perfect. But for most of the 12,000+ sites we've worked on at Seahawk, one of them fits better than Weglot's word-count clock ticking in the background.