Boosting International SEO with SiteLocaleAI

The Challenge

A SaaS startup built a static Gatsby site to showcase its product. While the site performed well in English, the marketing team wanted to capture traffic from Europe and Latin America. The main obstacles were:

• Zero‑code translation – hiring translators for each page was costly.
• Price localization – displaying prices in local currencies with psychological rounding.
• Search‑engine visibility – Google only indexed the English version, so international queries never surfaced the site.

The Solution: SiteLocaleAI

SiteLocaleAI offered a drop‑in, framework‑agnostic JavaScript library that could be self‑hosted. The team kept their own LLM API keys (Claude, GPT‑4o‑mini) and used the library’s CLI to generate pre‑rendered HTML for every target language.

Installation (one‑liner)

npm i @sitelocaleai/js

Integrating with Gatsby

Add the library to gatsby-browser.js:

import { initLocale } from "@sitelocaleai/js";

export const onClientEntry = () => {
  initLocale({
    apiKey: process.env.LLM_API_KEY,
    defaultLang: "en",
    supportedLangs: ["en", "es", "fr", "de", "pt"],
    priceConfig: {
      rounding: "psychological",
      currencyMap: {
        USD: "USD",
        EUR: "EUR",
        GBP: "GBP",
        BRL: "BRL",
      },
    },
  });
};

The initLocale call injects a tiny script that swaps text nodes at runtime and writes static translations during the build step when the CLI flag --pre‑render is used.

Pre‑rendering for SEO

npx sitelocaleai prerender --src ./public --langs es,fr,de,pt

The command crawls the built site, asks the LLM to translate each page, applies price rounding, and writes language‑specific HTML files (/es/index.html, /fr/index.html, …). Google’s crawler now sees fully translated pages, each with proper <link rel="alternate" hreflang="…"> tags automatically inserted.

Results

Metric	Before	After 3 months
Organic traffic (global)	12 k visits	38 k visits (+217 %)
Sessions from non‑English locales	1.2 k	9.5 k (+692 %)
Avg. session duration	1 min 12 s	2 min 04 s
Conversion rate (demo sign‑ups)	2.1 %	2.8 %
SEO index count (Google)	45 pages	212 pages (all languages)

The biggest win was the 3× lift in organic traffic from Spain, France, Germany, and Brazil, directly attributable to the pre‑rendered translations that Google could index. Price localization increased average order value by ~5 % because the psychological rounding (e.g., €9.99 → €9.95) matched local buying habits.

Why Self‑Hosted Matters

Because the LLM API key stays on the client’s server, the startup avoided:

• Data leakage – user‑generated content never left their own environment.
• Cost surprises – they only pay for the tokens they request.
• Vendor lock‑in – switching from Claude to GPT‑4o‑mini required a single config change.

Integration with Other Platforms

The same library works in WordPress, Shopify, and any static site generator. For WordPress, a plugin runs the CLI behind the scenes, so no Node.js is required on the host.

Lessons Learned

1. Start with a core language set – translating 5 languages covered 80 % of target traffic.
2. Use the CLI in CI/CD – the prerender step fits naturally into Netlify or Vercel pipelines.
3. Validate price rounding – run a quick audit to ensure the “psychological” values meet local regulations.

Next Steps

The team plans to add Japanese and Korean support, and to experiment with dynamic content translation using the runtime API for blog comments.

---

Ready to boost your global reach? Try SiteLocaleAI today and see how a few lines of code can turn a static site into an international traffic engine.

For more technical details, see the Installation guide and the SEO pre‑rendering documentation.