You launched into a new market and did everything by the book: Native speakers translators localised the content, the SEO agency set up the country folder or domain, adapted the imagery and implemented the hreflang tags. However, after six months, the pipeline from that market is still dry.
In theory, everything has been managed well and all the international SEO best practices have been followed. So why I don't see any meaningful results yet? The response is because there's another fifty percent of the work missing.. Hreflang is a routing instruction, not a credibility signal. It tells a search engine which version of a page belongs to which audience. It holds no opinion on whether your business deserves trust in that country, and a localised paragraph holds none either.
The half that decides commercial outcomes is authority, built locally and expressed in a form both search engines and language models can read. Authority separates an international SEO strategy that produces results from one that never gains traction, and most scopes I see leave it out completely.
Hreflang solves a matching problem and nothing else. It signals to a search engine that a page has equivalent versions for other languages or regions, so that a reader in Ireland receives the Irish page rather than the British one. It reduces the risk that near-identical regional pages compete with each other or get collapsed into a single version. The x-default value catches everyone your targeting does not name.
That is the full extent of it. Hreflang does not lift a page's ranking, does not make a business credible to buyers in a market and does not tell a language model why your expertise applies there. Google's own guidance on localised versions treats it as a targeting mechanism, not a performance lever, and describes it that way consistently.
Hreflang implementation is still very important though. Broken return tags, mismatched region codes and conflicting canonical signals will hold a site back, and fixing them is genuine work. The mistake is treating that work as the strategy rather than the prerequisite. A correctly configured international site is eligible to compete, however eligibility is not the same as competing.
Launching a /de/ country subfolder gives you a starting position in Germany, not a market position. The domain strength you built over ten years in your home market travels with the subfolder to a degree, which is why most businesses choose subfolders over separate country domains. What does not travel is the reason a German buyer, a German journalist or a German search engine would consider you a serious option in Germany.
The competitors that are already established in that market hold something you do not. They have links from sites that market actually reads, mentions in local trade media, reviews written in the local language and a trading history that local institutions can verify. None of that arrives when a subfolder goes live, in the same way that a strong homepage never removed the need to earn links to a new product page.
The structural choice changes the size of your starting position and nothing about the work ahead. A country code domain isolates you and forces you to build from zero. A subfolder shares existing strength but inherits the same problem covered below. A subdomain sits awkwardly between the two. Choose deliberately, then accept that the choice buys you a few months of head start rather than a market.
| What hreflang solves | What only local authority solves |
|---|---|
| Serving the right language version to the right user | Being considered credible by buyers in that market |
| Duplicate content conflicts between regional versions | Links and mentions from sites that market reads |
| Regional targeting signals for search engines | Recognition of qualifications and registrations |
| The correct version appearing in local results | Differentiation from your own other regional sites |
| Consistent indexing across country versions | Evidence a model can attribute to that market |
| Technical eligibility to compete | Commercial ability to compete |
| A one-off implementation | An ongoing programme |
A second problem now sits on top of the first, and Motoko Hunt of AJPR set it out well in Search Engine Journal. Her argument is that expertise obvious to a human reader can remain invisible to a machine, because a model recognises only the patterns it has seen repeated in training. Buyers increasingly reach you through systems that summarise before they link, so expertise a model cannot detect may as well not exist.
This is a different demand from the one search engines made ten years ago. Then the task was proving competence to people and letting the machine follow the links. Now the task includes making the proof itself machine readable.
Qualifications carry weight locally and almost none across a border. A chartered accountant in the United Kingdom, a registered building practitioner in Australia and a specialist qualification issued by a European regulator each signal seniority to a reader in that country. To a model trained mostly on English language sources weighted towards the United States, many of these carry no learned association at all.
Nothing about this failure is visible from the inside. Your regional page names a properly qualified expert, and a local reader recognises that qualification at a glance. The qualification convinces the people who read it but to an AI model that has never come across it before, it is just another job title.
Publishing the same positioning across multiple regional sites makes those sites easy to treat as one. Motoko Hunt describes this well in her article. The stronger the brand discipline, the less distinguishable any single market becomes, and local expertise that genuinely exists produces no separate signal.
Localisation and authority pull against each other here. Localisation asks you to say the same thing in another language. Authority asks you to say something in that market that you could not have said anywhere else.
Seven signals do most of the work, and each one is buildable within a quarter or two.
Apply one test to every regional page you own. Ask whether it contains anything a competitor in that market could not have written by reading your global site and running it through a translator. If the answer is no, the page carries no local authority, however well it is targeted.
Credibility becomes a usable signal only when you state the relationships behind it. A human reader fills in context automatically. A search engine or a language model needs those connections written down in a form it can parse, and they do it through structured data.
Structured data is a block of code on the page that describes your business and your people in a shared vocabulary rather than in prose. Search engines and models already understand that vocabulary, so it removes the guesswork about who wrote a page, what they are qualified in and which organisation stands behind them. Four moves cover most of what an international site needs.
Structured data creates no authority on its own. It makes the authority you already hold readable by the systems now deciding which businesses to put in front of a buyer, which is the whole problem in a market where your reputation is new.
This is the same discipline Google describes in its guidance on creating helpful, people-first content, applied market by market rather than site wide. It is also the substance of generative engine optimisation, which is largely this work aimed at models rather than at result pages. The businesses that win here will not always hold the deepest expertise. They will be the ones that made their expertise easiest to recognise.
Four questions separate a technical supplier from a strategic one. Put them to anyone quoting on your expansion:
A supplier scoped to hreflang alone will answer the first question with a technical plan and the third with silence. That is the gap you are trying to close, and closing it needs someone senior enough to work across structure, content and authority at the same time. My international SEO work runs on exactly that basis, and how I work sets out what an engagement involves.
If you are planning an expansion and want a second opinion on the scope you have been quoted, request a free top level SEO evaluation. There is no obligation to proceed and the findings are yours either way.