How does link equity transfer across ccTLDs and gTLD subdirectories? A definitive 2026 guide to international site architecture, authority inheritance, a decision framework and a per-market link-building plan.
| TL;DR Country-code domains (ccTLDs, like example.de) are the strongest possible geo-targeting signal, but each one is a separate website that must earn its own authority from scratch. A single generic domain (gTLD, like example.com) with country or language subdirectories (example.com/de/) consolidates every backlink onto one authority profile that all markets share. That is the core trade-off: ccTLDs buy local trust at the cost of fragmented link equity; gTLD subdirectories buy consolidated link equity at the cost of having to build geo-relevance manually. Settled 2026 default: most internationalising businesses should consolidate on one gTLD with subdirectories and use hreflang plus localised content for geo-targeting. Reserve ccTLDs for markets where local trust, regulation or branding genuinely demands a local domain. And note the change that dates most competitor guides: Google retired the Search Console country-targeting setting in 2022, so gTLD geo-targeting now runs on hreflang and localisation, not a toggle. |
Few decisions in technical SEO are as expensive to reverse, or as widely misunderstood, as international site architecture. Choose the wrong structure and you can spend years and a substantial budget building authority that never compounds, because it is scattered across half a dozen separate domains that, in Google’s eyes, barely know one another. Choose well and every link you earn anywhere lifts your visibility everywhere. The mechanism that decides which outcome you get is link inheritance: the question of whether, and how, the authority earned by one part of your international footprint passes to the others.
This guide settles the ccTLD-versus-gTLD question for 2026, with link inheritance as the organising principle rather than the usual vague talk of “local relevance”. We will define the structures precisely, explain exactly how authority does and does not flow through each, give you a decision framework you can apply this week, and work through documented migrations with real numbers. We will also correct the single most common error in competing guides, which still recommend a Google Search Console setting that has not existed since 2022. If you want the strategic layer that sits above architecture, our guide to international link building is the companion piece; this article is the structural foundation beneath it.
The decision at a glance
Before the detail, here is the settled position in one table. The right structure depends on how many markets you serve, how much local trust matters in them, and how much link-building budget you can sustain per market.
| Structure | Example | Link inheritance | Best for |
| gTLD + subdirectories | example.com/de/ | Full — all markets share one authority profile | Most multi-market businesses; consolidation |
| gTLD + subdomains | de.example.com | Partial — treated as semi-separate | Rarely optimal; technical separation needs |
| ccTLDs | example.de | None — each domain earns authority alone | Single-market focus; strong local trust/regulation |
| Separate gTLDs | examplede.com | None — fully separate sites | Almost never advisable; avoid |
If you remember nothing else: subdirectories inherit authority by default, ccTLDs inherit nothing, and the cost of ccTLDs is paid in duplicated link-building effort across every market you enter.
The architecture options, defined precisely
The terminology is loose in most discussions, so let us fix it. There are effectively four ways to structure an international presence, and the differences are not cosmetic.
Country-code top-level domains (ccTLDs)
A ccTLD is a domain whose extension is tied to a specific country: example.de for Germany, example.fr for France, example.co.uk for the United Kingdom. Google treats a ccTLD as a strong, explicit signal that the site is intended for that country, the clearest geo-targeting signal available. Each ccTLD is, however, an entirely separate website. example.de and example.fr are two different sites that happen to share a brand, and authority does not flow between them automatically.
One important caveat: Google treats certain “vanity” ccTLDs as generic because the wider world uses them generically. The official documentation names examples such as .tv and .me as being treated like gTLDs. The same generic treatment is widely understood to apply to popular tech extensions like .io, .ai and .co, which is precisely why startups adopt them without worrying about being geo-pinned to a small island nation. If you are relying on a ccTLD for a geo-signal, confirm it is not on Google’s generic list first.
gTLD with subdirectories
A generic top-level domain, most commonly .com, with country or language sections organised as subdirectories: example.com/de/, example.com/fr/, example.com/en-gb/. This is a single website. Every page lives on one domain, so backlinks earned by any section contribute to one shared authority profile. The geo-signal is weaker than a ccTLD’s and must be supplied through hreflang, localised content and on-page signals, but the link-inheritance advantage is total.
gTLD with subdomains
The same generic domain, but with country sections as subdomains: de.example.com, fr.example.com. This sits awkwardly in the middle. Google can associate the subdomains with the root, but treats them as semi-independent, so subdomains do not seamlessly share link equity with the root domain. For international architecture specifically, subdomains combine most of the management overhead of separation with few of the link-inheritance benefits of consolidation. The deeper subdomain-versus-subfolder analysis applies here in full: unless you have a concrete technical reason to separate, the subdirectory is the stronger choice.
Separate generic domains
Registering wholly different domains per market (examplede.com, examplefr.com) is the worst of all worlds for SEO: total fragmentation with none of the local-trust upside of a true ccTLD. It is occasionally done for legal or acquisition reasons, but it is never the recommended starting point. We mention it only so you can rule it out with confidence.
How link inheritance actually works across each structure
Link equity, the ranking power that backlinks transfer to a site, accumulates at the level of the site or host. This is the hinge on which the entire international-architecture decision turns, so it is worth being exact about what “inheritance” means in each case.
On a gTLD with subdirectories, there is only one site. A German newspaper that links to example.com/de/ is pointing at the same domain that your example.com/en-gb/ and example.com/fr/ pages live on. That link lifts the whole domain’s authority, and every market draws on it. You run one authority-building programme and all markets benefit. This is link inheritance in its strongest, most automatic form, the gains are inherited across the entire structure by default.
On ccTLDs, there is no inheritance at all. A link to example.de helps example.de and nothing else. example.fr starts life with zero inherited authority and must earn its own from scratch, market by market. If you operate in eight countries on eight ccTLDs, you are running eight separate authority-building programmes. The brand recognition may carry across borders, but the link equity does not.
On subdomains, inheritance is partial and unreliable. Some authority may flow between a subdomain and the root through internal links, but Google’s treatment of subdomains as semi-separate entities means you cannot count on it. You end up doing extra work to recreate, imperfectly, the consolidation a subdirectory would have given you for free.
The practical consequence is stark. The cost of acquiring a strong editorial link is the same whether it points at a subdirectory or a ccTLD, but its return is radically different. On a subdirectory it compounds across your whole footprint; on a ccTLD it is confined to one market. Understanding this is the difference between an international SEO budget that compounds and one that merely spreads thin.
A worked comparison: six markets, two structures
Picture two versions of the same company, each entering six European markets with an identical link-building budget that yields the same total number of quality editorial links per year. Company A consolidates on example.com with /de/, /fr/, /es/, /it/, /nl/ and /pl/ subdirectories. Company B launches six ccTLDs: example.de, example.fr and so on.
Company A’s links, wherever they land, all accrue to one domain. After two years it holds one authority profile built from the full year-on-year link flow, comfortably into competitive territory, and every market ranks on the strength of that shared profile plus its own local relevance. Company B’s identical link output is divided six ways. Each ccTLD receives roughly a sixth of the links and sits low on its own authority curve, never reaching the altitude where the expensive top end of the curve pays off. Same spend, same links, and Company A outranks Company B in most markets, because consolidation let the authority compound while fragmentation forced it to spread thin. To draw level, Company B would need roughly six times the budget. That is the entire argument in one comparison, and it is why the framework below weighs sustainable per-market link capacity so heavily.
The International Architecture Decision Framework
Use this framework before committing to, or migrating, an international structure. It is built around the three variables that actually decide the outcome: market concentration, local-trust dependence, and sustainable link-building capacity per market. Answer each question, then read the verdict.
| Question | If yes… | If no… |
| Do you serve, or will you serve, more than two or three markets? | Lean gTLD + subdirectories (consolidation scales) | A ccTLD per market may be manageable |
| Is local trust or regulation decisive in your sector (finance, health, legal, government)? | A ccTLD’s local signal may justify the fragmentation | Consolidation wins; you do not need the local domain |
| Can you sustain a full link-building programme in every market, indefinitely? | ccTLDs are viable; you can feed each one | Consolidate — you cannot afford N separate authority climbs |
| Is your strongest market already mature on a ccTLD with deep authority? | Keep it; consolidate the newer markets onto a gTLD | A clean gTLD build is simpler |
| Do you want one set of analytics, one technical stack and one team to manage it? | gTLD + subdirectories | Separate properties are acceptable but costly |
The framework usually points the same way, and that is the honest finding rather than a hedge: for the majority of businesses internationalising in 2026, a single gTLD with subdirectories is the correct default, because almost no one can sustain genuinely independent link-building programmes across many markets at once. The ccTLD case is real but narrow: a single dominant market, a sector where the local domain materially lifts trust or is effectively mandatory, or a legacy ccTLD already carrying years of authority you would be foolish to throw away.
What Google actually says in 2026 (and the advice that is now wrong)
Google’s documented position has two parts that people routinely conflate. First, the TLD type is not a ranking factor, a .de does not rank higher than a .com because of the extension. Second, a ccTLD is used as a geo-targeting signal: it tells Google which country the site is for. Both are true simultaneously, which is why the debate confuses people. The extension does not make you rank better; it makes you rank for the right audience.
For a gTLD, Google’s guidance on multi-regional sites is to supply geo and language signals explicitly, because the crawler will not infer them reliably, it generally crawls from the United States and does not vary its location to discover regional variants. In 2026 those explicit signals are hreflang annotations, genuinely localised content, local backlinks and local business signals.
The advice that dates a competitor’s guide instantly
Here is the teardown. A large number of international-SEO articles, including ones published recently, still tell you to set your gTLD’s target country using the “International Targeting” setting in Google Search Console. That tool was deprecated and removed from Search Console in September 2022. It no longer exists. Google’s own statement at the time confirmed it would continue supporting hreflang while retiring the country-targeting report, on the basis that the setting offered little value. If a guide tells you to flip a country switch in Search Console for your .com, it has not been meaningfully updated in over three years, and you should treat the rest of its advice with the same suspicion.
The 2026 reality is therefore cleaner than the old advice suggested: for a gTLD, you cannot simply declare a target country to Google. You earn geo-relevance through hreflang, localisation and local links. That makes local link building, not a Search Console toggle, the real lever of international visibility on a consolidated domain.
The case for consolidation: gTLD plus subdirectories
The argument for consolidation is not theoretical. Documented migrations from fragmented structures onto a single gTLD with subdirectories show the link-inheritance benefit converting into traffic, often quickly.
NFON: ccTLDs to one gTLD
The cleanest enterprise example is NFON, a European B2B telephony provider, which consolidated multiple European ccTLDs into a single gTLD with country subdirectories. Organic traffic in Austria was reported to have risen by around 90% within five months, with organic leads doubling over the same period. The mechanism is exactly link inheritance: instead of each country domain climbing its own authority hill, every market began drawing on one consolidated profile. Treat the precise figures as a documented-but-secondary report rather than an audited number, but the direction is consistent with everything the architecture predicts.
Subdomains to directories
An international-SEO practitioner’s documented migration from subdomains to subdirectories on one domain reaches the same conclusion from a different starting point: subdomains on a gTLD waste consolidation potential, and merging international versions into directories on a single domain delivers measurable gains. The author’s recommendation is unambiguous, consolidate on one gTLD with logical subdirectories, implement hreflang and redirects carefully, and only fragment into more versions when a concrete need (different currencies, laws or strong local branding) demands it.
This is also why so many global brands run a single .com with country subdirectories rather than a constellation of ccTLDs, the structure lets one domain accumulate global authority while still serving localised content. For UK-focused operators expanding into Europe, the same logic applies, and our guide to link building for European markets covers how to feed the local-signal side of a consolidated structure.
When ccTLDs still win in 2026
Consolidation is the default, not a dogma. There are genuine cases where a ccTLD is the better call, and forcing a local-first business onto a generic domain is its own mistake. A ccTLD earns its keep when:
- Local trust is decisive. In some markets, users simply trust and click a local domain more readily. A .de in Germany or a .co.uk in Britain can lift click-through and conversion in-market in ways a generic .com does not, particularly for consumer-facing and transactional businesses.
- Regulation or data residency demands it. In sectors and regions where local consumer-protection and data-privacy regimes such as the GDPR shape buyer behaviour, a local ccTLD can signal that the business is bound by local law, which is a real conversion factor for finance, health, legal and public-sector services.
- You dominate, or intend to dominate, a single market. If 90% of your business is in one country, the fragmentation cost is largely hypothetical, you only have one authority profile to build anyway, so you may as well have the strongest possible geo-signal.
- You already hold a mature, authoritative ccTLD. Years of accumulated authority on an established ccTLD is an asset, not a liability. Do not throw it away to chase a textbook structure; build new markets on a gTLD and leave the strong legacy domain alone.
The throughline is that ccTLDs trade away link-equity consolidation for local strength. That trade is sound when local strength is the thing that makes or breaks the business, and unsound when you are spreading a finite link-building budget across markets that would have been better served compounding onto one domain. The Indian and South Asian markets, for instance, illustrate both sides: a local .in can build trust, yet many global brands serve the region perfectly well from a consolidated gTLD subdirectory backed by local links.
The hidden cost of ccTLD fragmentation: link-building economics
The case against unnecessary ccTLDs is rarely made in budget terms, which is exactly why it is so often underestimated. Here is the economic reality, stated plainly.
Suppose you operate in six markets. On a consolidated gTLD, a single link-building programme, earning, say, a steady flow of quality editorial links each quarter, lifts one authority profile that all six markets share. Your cost per unit of authority is divided across six markets, and because authority compounds, that single profile can reach a competitive altitude that no individual market site would on its own.
On six ccTLDs, the same total link-building output is split six ways. Each domain receives a sixth of the links and climbs a fraction of the way up its own authority curve. Because the back half of any authority climb is the expensive half, none of the six reaches competitive altitude. You have spent the same money and bought six mediocre profiles instead of one strong one. To match the consolidated outcome, you would need to multiply your link-building budget by roughly the number of markets, which almost no business can sustain indefinitely.
This is the calculation competitors omit and the one that should drive the decision. Fragmenting into ccTLDs is, in effect, a decision to multiply your link-building costs. If you are going to make that decision, make it deliberately, with the local-trust upside clearly worth the multiplied spend, not by default because a domain registrar suggested a local extension. For the broader programme this feeds, our link building strategies hub sets out how to build the authority that architecture then distributes.
The hybrid structures most guides ignore
The debate is usually framed as a binary, ccTLDs or a gTLD, but the strongest real-world setups are often hybrids, and competitor guides rarely mention them. Three are worth knowing.
Strong-market ccTLD plus a gTLD for the rest
You do not have to choose one structure for the whole business. A common and defensible pattern is to keep a mature, authoritative ccTLD for your dominant home market, where its local trust and years of accumulated authority are real assets, while building all newer markets as subdirectories on a single gTLD. You forgo cross-inheritance between the home ccTLD and the gTLD, but you stop the bleeding everywhere else and you do not destroy a strong legacy asset for the sake of a tidy diagram. The decision is per-market, not all-or-nothing.
ccTLD with language subdirectories
A single-market business that nonetheless serves multiple languages, common in Switzerland, Belgium, Canada or the Gulf, can run a ccTLD with language subdirectories: example.ch/de/, example.ch/fr/, example.ch/it/. Here the ccTLD supplies the country signal and the subdirectories handle language, with hreflang routing users to the right version. All the language sections still inherit the ccTLD’s authority because they sit on one domain. This is the right structure when your geography is fixed but your audience is multilingual.
The gccTLD trap
The hybrid that catches people out is the accidental one: choosing a vanity extension for branding and assuming it geo-targets. Because Google treats certain vanity ccTLDs as generic, a .io, .ai, .co, .tv or .me behaves like a gTLD, it gives you no country signal, so you must geo-target through hreflang and localisation exactly as you would on a .com. That is usually fine, but only if you know it going in. Treating a generic vanity extension as if it were pinning you to a country is a quiet, common, and entirely avoidable mistake.
The per-architecture link-building playbook (your Monday-morning deliverable)
Two things you can do this week regardless of which structure you run: an International Authority Audit to quantify your fragmentation, and a per-market allocation worksheet to spend link-building effort where it actually compounds.
Step 1: The International Authority Audit
- List every domain, subdomain and ccTLD your brand operates. In your backlink tool, check each one separately and record its authority metric and referring-domain count. This is your fragmentation map.
- Sum the referring domains across all properties, then note how they are distributed. If 80% of your links sit on one property and the others are near-zero, you have a consolidation opportunity (or, if those others are ccTLDs you intend to keep, a clear signal of where your link-building deficit is).
- Identify which properties hold your commercial, revenue-generating pages, and whether the properties earning the links are the same ones. A mismatch, links on one domain, money pages on another, is trapped equity you can recover. The crawlers and indexes you need are covered in our best link building tools guide.
- Decide, per property, whether it should be kept (strong local case), consolidated (weak, fragmenting equity) or built up (kept but under-linked). Write the decision down. That document is your international architecture roadmap.
Step 2: The per-market allocation worksheet
- For a consolidated gTLD: run one central link-building programme aimed at root-domain authority, then layer local link building per market to supply the geo-relevance that the structure cannot generate on its own. Local citations, local press, local industry bodies, the geo-signal you used to set in Search Console now comes from these links.
- For ccTLDs: budget a genuine, sustained programme per domain, and be honest about whether you can fund it. If you cannot fund market three, that is your evidence that market three should have been a subdirectory.
- For both: prioritise local, in-language links over generic global links for each target market. A link from a respected German publication does more for your German visibility than a higher-authority but geographically-irrelevant link, on a ccTLD because it is the only authority that domain gets, and on a subdirectory because it supplies the local relevance signal.
Migrating from ccTLDs to a consolidated gTLD
If your audit says consolidate, treat the migration with the same rigour as any high-stakes move. Done well, it converts fragmentation into compounding authority. Done badly, it loses the very equity you set out to consolidate. The non-negotiables:
- Build a complete 1:1 redirect map. Every URL on every old ccTLD must 301-redirect to its exact equivalent in the new subdirectory structure. Missing redirects become 404s and lost links.
- Keep the old ccTLDs live to serve those redirects, do not let them expire. The redirects are how the accumulated authority transfers; killing the domains kills the transfer.
- Implement hreflang across the new subdirectories before launch, so Google can immediately map each section to its market and language and avoid showing the wrong version.
- Migrate URLs first; change content, design or platform later as separate phases. Combining everything makes results impossible to diagnose, which is exactly how migrations acquire their bad reputation.
- Expect, and plan for, a temporary visibility dip while Google reprocesses the redirects and consolidates signals, then monitor recovery. Watch for redirect chains and 404s, which are the usual culprits when recovery stalls.
- Run reclamation outreach on your highest-value historic links, asking publishers to update them directly to the new URLs. A live link beats a redirected one, and your audit already told you which links justify the effort.
What to expect on the timeline
Consolidation migrations follow a recognisable curve. For the first few weeks, expect a dip as Google recrawls the old ccTLD URLs, processes the redirects and begins attributing the inherited authority to the new subdirectories. Through the following weeks, rankings typically recover toward baseline as signals consolidate, and the upside, the compounding benefit of one shared profile, tends to build over the months after that rather than arriving overnight. The single biggest predictor of a clean recovery is redirect integrity: a complete 1:1 map with no chains and no orphaned URLs. If visibility has not recovered after a couple of months, audit the redirect map and crawl logs before assuming the migration itself was a mistake, because the cause is almost always a mapping gap rather than the strategy.
Where hreflang fits (and where it does not)
hreflang is frequently misunderstood as a geo-targeting or authority tool. It is neither. hreflang tells Google which language and regional version of a page to show to which user; it does not pass link equity, and it does not boost rankings. On a consolidated gTLD it is essential plumbing, the mechanism that stops Google serving your German page to a British user, but it is a routing instruction, not an authority lever.
The practical implication is that hreflang and link inheritance are separate concerns that must both be handled. Subdirectories give you the inheritance automatically; hreflang gives you the correct-version routing; localised content and local links give you the geo-relevance. A consolidated international site needs all three working together, and the deeper mechanics of how international links and hreflang interact deserve their own detailed treatment, which is a topic in its own right rather than something to cram into an architecture decision.
Three hreflang pitfalls account for most of the problems in practice, and all three are worth pre-empting. The first is missing return tags: hreflang must be reciprocal, so if your German page points to your French page, the French page must point back, or Google ignores the cluster. The second is mismatched or invalid region codes, using a language code where a language-region code is needed, or inventing combinations that do not exist. The third is annotating pages that are not all indexed: hreflang assumes Google has crawled and indexed every version in the cluster, which is reliable on small sites but fragile on large ones with many language variants. None of these affect link inheritance, your authority still consolidates regardless, but they routinely cause the wrong version to surface in the wrong market, which quietly undoes the localisation work the architecture was meant to support.
Five myths competitors still repeat
- “Set your gTLD’s target country in Search Console.” That tool was removed in 2022. Geo-targeting a gTLD now runs on hreflang, localisation and local links.
- “ccTLDs rank higher.” The extension is not a ranking factor. A ccTLD targets an audience; it does not boost rank. Authority and relevance still do the ranking.
- “Subdomains and subdirectories are the same for international SEO.” They are not. Subdirectories inherit authority automatically; subdomains are treated as semi-separate, so you forfeit consolidation for little gain.
- “Duplicate-ish content across regional versions will be penalised.” Google’s guidance explicitly accepts that regional variations may share content, and does not require you to hide it. hreflang resolves which version to show.
- “A vanity ccTLD gives you a free geo-signal.” Extensions like .tv, .me and similar are treated as generic by Google, so they carry no country signal at all. Check before you rely on one.
Architecture, link inheritance and AI citations in 2026
A dimension the original ccTLD debate never had to consider: generative engines and entity understanding. As AI search assistants increasingly answer queries by citing sources, the question becomes whether a fragmented international footprint helps or hinders being recognised as a single authoritative entity.
The direction of travel reinforces consolidation. Generative systems lean on entity comprehension, understanding which sites belong to which organisation and which carry topical authority in which market. A single consolidated domain presents one clear, strong entity. A constellation of ccTLDs asks the system to connect the dots between separate sites, and the more fragmented the footprint, the easier it is for the connection to be missed and for your authority to look thinner than it is. None of this overturns the local-trust case for ccTLDs, but it adds weight to the consolidation default: the structure that is easiest for classical search to credit with accumulated authority is also the easiest for an AI engine to recognise, trust and cite.
There is a measurement consequence too. If your footprint is fragmented, you will struggle to see your true position in AI answers, because each ccTLD reads as a separate, smaller player rather than one authoritative brand. Consolidation does not just help you get cited; it makes your visibility legible, to the engines, to your analytics, and to anyone trying to assess your brand’s authority in a market. As AI share-of-voice becomes a tracked KPI alongside rankings and links, the brands that concentrated their authority will find it easier both to earn citations and to prove they earned them. The brands spread across a dozen country domains will keep wondering why their obvious market strength is not showing up where the answers are now being formed.
Frequently asked questions
Is a ccTLD or a gTLD better for SEO in 2026?
For most businesses serving multiple markets, a single gTLD with country or language subdirectories is better, because it consolidates all your link equity onto one authority profile that every market shares. A ccTLD is better when you focus on a single market, or when local trust or regulation makes a local domain decisive. The extension itself is not a ranking factor; the difference is link inheritance and geo-signalling.
Do ccTLDs share link equity with each other or with a .com?
No. Each ccTLD is a separate website. A link to example.de helps only example.de; it does not pass authority to example.fr or to example.com. This is the central reason ccTLDs multiply your link-building cost: every market must earn its own authority from scratch.
Do subdirectories pass authority from the main domain?
Yes. Subdirectories such as example.com/de/ are part of one domain, so they share the root domain’s authority automatically and contribute their own earned links back to it. This is the strongest form of link inheritance and the main reason subdirectories are the default recommendation for international architecture.
How do I geo-target a gTLD now that Search Console targeting is gone?
Google retired the International Targeting / country-targeting setting in 2022, so you can no longer declare a target country to Google directly. In 2026 you geo-target a gTLD through hreflang annotations, genuinely localised content, local backlinks and local business signals. Local link building has effectively replaced the old Search Console toggle.
Should I migrate my ccTLDs to a single gTLD?
Only if your audit shows fragmented authority you cannot afford to sustain across markets, and the local-trust case for the ccTLDs is weak. If a ccTLD is mature and authoritative, or local trust is decisive in that market, keep it. When you do migrate, use a complete 1:1 redirect map, keep the old domains live to serve the redirects, implement hreflang first, and expect a temporary dip before recovery.
Are vanity extensions like .io and .ai ccTLDs?
Technically yes, but Google treats popular vanity extensions as generic rather than country-targeted. Google’s documentation names examples such as .tv and .me, and the same generic treatment is widely understood to apply to .io, .ai and .co. They carry no country geo-signal, which is usually fine for global tech brands but means you cannot rely on them to target a specific country.
The settled verdict
International site architecture comes down to a single trade-off, and once you see it clearly the decision is rarely close. ccTLDs give you the strongest local signal and the strongest local trust, but they fragment your link equity, forcing you to build authority market by market at multiplied cost. A gTLD with subdirectories gives you full link inheritance, one compounding authority profile that every market shares, at the cost of having to manufacture geo-relevance through hreflang, localisation and local links rather than the extension.
For 2026 the default is settled: consolidate on one gTLD with subdirectories unless a single dominant market, decisive local trust, hard regulation, or a mature legacy ccTLD makes the local domain genuinely worth its fragmentation cost. Run the decision framework before you commit, run the authority audit before you migrate, and remember that the geo-signal you can no longer set in Search Console is now earned through local link building. Architecture decides whether your authority compounds or scatters; the strategy that builds that authority in the first place is the subject of our international link building guide, and the benchmarks to measure it against live in our link building statistics for 2026. Get the structure right, and every link you earn anywhere works for you everywhere.
