My Next.js Blog: Global in a Week (21 Languages)

A little while ago, I decided to try a simple experiment: build a Next.js blog in 21 languages and see how search engines would handle it.

There was no big business plan behind it. I mostly wanted to test multilingual routing, localization, indexing, and SEO on a real project instead of another demo app.

After the first week, Google Search Console showed 439 impressions and 5 clicks.

Yandex had also started picking up some of the pages.

The numbers are obviously small, but for a brand-new personal blog, I was happy to see that Google was already discovering pages and showing them in search.

Why I used Next.js

The blog is built with Next.js, and every article has a separate version for each of the 21 supported languages.

I chose Next.js because it gave me a convenient setup for:

• internationalized routing
• server-side rendering
• static generation
• reusable layouts and metadata
• generating a large number of localized pages

For this kind of project, that made the implementation much easier.

Of course, using Next.js does not automatically mean better rankings or faster indexing. It just gives you the tools to build pages that are easy for search engines to crawl and understand.

Google still decides what gets indexed and what gets shown in search.

The first search results

After roughly one week, I had:

• 439 Google impressions
• 5 clicks
• about 1.1% CTR

Five clicks is not much traffic, but traffic was not really the interesting part yet.

What interested me was that the site had already started getting impressions across a large multilingual structure.

One article on my blog can exist as 21 separate URLs.

That creates a lot more possible search entry points than publishing only in English.

It does not mean 21 languages will give you 21 times more traffic. Each page still has to earn its own visibility.

But it does make the site much more interesting to monitor.

Indexing took longer

The next thing I started watching was indexing.

My sitemap.xml is generated automatically, and I submit it through Google Search Console.

A few weeks later, I had 471 pages in the group I was tracking.

Out of those:

• 436 pages were indexed
• 35 pages were still not indexed

So around 93% of those pages had made it into Google's index.

That was a useful reminder that discovery, indexing, impressions, and clicks are all different things.

Submitting a sitemap does not mean Google will immediately index every page.

An indexed page also does not mean it will rank or get traffic.

And if a new page has not been indexed yet, that does not automatically mean there is a serious SEO problem.

Multilingual sites scale very quickly

One thing that became obvious pretty fast is how quickly the number of pages grows.

If I publish one article, I am not really publishing one page.

I am publishing 21 localized pages.

Ten articles can become 210 URLs.

Twenty articles can become 420 URLs.

That means small technical mistakes can also get duplicated very quickly.

A problem with metadata, internal links, canonical tags, hreflang, translations, or URL structure can affect dozens or hundreds of pages instead of just one.

So multilingual publishing gives you more opportunities, but it also gives you more things to check.

What I learned so far

The biggest lesson for me is that multilingual SEO is not simply about creating as many translated pages as possible.

The technical side matters, but there are several separate questions:

1. Can Google discover the URLs?
2. Can Google crawl them correctly?
3. Which pages actually get indexed?
4. Which languages receive impressions?
5. Which pages receive clicks?
6. Which localized versions eventually start getting consistent search traffic?

At first, I mostly looked at the total number of impressions.

Now I think it is much more useful to look at each part separately.

Current result

So far, the experiment looks like this:

First week:

• 439 Google impressions
• 5 clicks
• Yandex started showing some pages

A few weeks later:

• 471 pages tracked
• 436 indexed
• 35 still not indexed

I would not call that proof that multilingual SEO automatically works, or that Next.js gives some special SEO advantage.

But it does show that a small multilingual project can get discovered across a surprisingly large number of pages fairly quickly.

For me, the more interesting question now is not whether Google can find the site.

It is which languages and articles will eventually turn those impressions into consistent organic traffic.

Blog: https://jsvar.com/en/blog