International SEO: One Brand, Many Markets, One Entity
-
International SEO targets the right content to the right country and language without diluting authority. -
hreflang and a clear URL structure tell engines which version belongs to which market. -
Entity consistency across markets is what lets AI answer engines recognize you everywhere. -
Localization is more than translation: intent, currency, and norms change per market.
1. What is international SEO?
2. Choosing a targeting structure
| Structure | Pro | Con |
|---|---|---|
ccTLD (site.de) | Strongest geo signal, local trust | Splits authority, costly to maintain |
Subdirectory (site.com/de) | Consolidates authority, simple | Weaker standalone geo signal |
Subdomain (de.site.com) | Clean separation | Authority less consolidated than a subdirectory |
3. hreflang done right
4. Localization, not translation
5. Entity consistency for AEO
A [DTC apparel] brand was live in [6] markets, each run by its own local team who had never compared notes. I audited one thing: did the brand and its hero products resolve to the same entity, with the same attributes, in every market? They did not, not even close. Every market had drifted into its own dialect. Different category framing, different product facts, different names for the same thing.
In the markets where the entity resolved cleanly and consistently, the brand got named in local-language AI answers. In the fragmented markets it was a ghost, even where the classic rankings looked perfectly healthy. The engine could not confirm it was the same brand, so it refused to vouch for it. Entity consistency across markets is not a detail. It is the whole game.
Translation makes the words local. It does not make the entity coherent.
International SEO is still taught like it is a decade ago: tag the right language, dodge duplicate content, avoid cannibalization, collect your ranking. That framing has aged badly. The real job now is making sure the same product resolves to the same entity whether the query comes in English, German, or Japanese, and almost nobody is auditing for it.
Translate a page and you get local copy describing what looks, to a machine, like a slightly different product in every market. Localize the entity instead, with stable attributes and a disciplined cross-market identity, and the engine can finally merge your entire global footprint into one trusted node. One of those is busywork in six languages. The other decides whether you exist in AI answers abroad.
Hand localization to the cheapest vendor and you get six fluent translations that each quietly describe a subtly different entity. The copy reads flawlessly native. The attributes drift a few degrees in every market. The engine sees six nodes it cannot confidently merge into one, and a fragmented entity is a weak entity in every language at once.
Getting this right takes someone who treats every market as the same entity expressed locally, holding the attributes rock-stable while only the language moves. That is a senior, coordinated discipline, not a per-market paste job farmed to whoever was cheapest. The bulk-translation shops optimize for fluent words. We optimize for one coherent entity the engine trusts everywhere it appears.
Six fluent pages describing six different things is not international SEO. It is fragmentation.
- 1. Google Search Central. International and multilingual SEO, including hreflang guidelines.
- 2. Schema.org. Vocabulary for describing entities consistently across pages.
- 3. W3C. Language tag standards (BCP 47) used in hreflang.
- 4. Google Search Central. Managing multi-regional and multilingual sites.
- 5. Google Search Central. hreflang annotations and return-tag requirements.
- 6. Unicode CLDR. Locale data for languages, regions, and formats.