Discussion
  • Enterprise SEO

One global domain with country teams: centralise SEO decisions, or let each market run its own? What broke in each model?

Published 1 reply

Discussion starter: an enterprise SEO question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.

Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.

1 reply

  • admin
    Admin
    Original poster

    Centralise the decisions whose effects cross market boundaries, and let markets own the decisions whose value comes from local knowledge. On one global domain that line falls in a fairly predictable place. The platform (templates, URL and hreflang architecture, robots.txt, redirects, canonical logic, rendering, release checks and measurement definitions) needs central ownership, because a mistake there can reach every market at once. Demand research, content priorities, local links and PR, and local conversion details belong with the country teams, because a central team cannot see them clearly from headquarters.

    Both pure models break, and they break in different ways. Full centralisation tends to produce a queue, plus content that reads like a translation of the headquarters market. Full devolution tends to produce technical drift that nobody owns until it surfaces as the wrong country’s page ranking, or as one market’s release breaking another’s. Public evidence here is practitioner reports rather than controlled comparisons, so read the failure tables below as a risk checklist, not measured rates. The workable answer is a hybrid with written decision rights, and the test for each decision is its blast radius, not where the org chart happens to put it.

    Why a shared domain changes the answer

    With separate country domains, a market team can make most of its mistakes inside its own property. On one domain with country folders, markets share templates, crawl, internal linking, robots.txt and, to a degree, how Google assesses the site. Google’s core update guidance asks owners facing a broad drop to check whether “your site overall (not just individual pages)” is helpful, and says “it could take several months for our systems to learn and confirm that the site as a whole is now producing helpful, reliable, people-first content in the long term”. That does not mean one weak market sinks the others. It does mean the quality of one market’s folder is not only that market’s business.

    Google’s guide to multi-regional and multilingual sites lists subdirectories as easy to set up and low maintenance, which is much of the appeal of one domain. The same guide lists the signals Google uses to work out a page’s intended audience: hreflang, country-code domains and server location, plus local addresses and phone numbers, local language and currency, links from other local sites, and Business Profile signals where available. Much of that second group is local knowledge and local work, which is why a fully central SEO team struggles to deliver it.

    Market fit is not a theoretical concern. Aleyda Solis’s analysis of the May 2026 core update found that intent, market fit and source type drove the biggest visibility shifts, and reported UK results moving from US.com retailers towards local-market domains. A global domain can still win locally, but on that evidence each market folder should look and work like a local business: prices, delivery, returns, support, reviews and examples that belong to that market.

    What broke under centralisation

    Failure How it happens How it shows up
    The queue Every market request competes for one team’s time, and the roadmap is ranked by the largest market’s revenue Months-long waits for smaller markets, then workarounds: landing-page builders, microsites or agency-run subdomains outside the standard templates
    Translated, not localised Headquarters keyword sets and briefs are translated, so local terminology, product naming and seasonality are missed Market pages that rank for few local queries, and Search Console queries that do not match the terms the page was written for
    Global switches with local side effects One decision applies everywhere: an IP-based redirect, a pruning rule, a template that drops local delivery or payment information Several markets dropping on the same deployment date
    Local requirements treated as exceptions Legal pages, currencies, local trust marks and payment methods wait for a global release Conversion gaps between markets with similar traffic

    The redirect case deserves a specific mention because it is a central decision that can quietly break every market. Google’s guide says to “Avoid automatically redirecting users from one language version of a site to a different language version”, and warns that if you serve different content by location, Google “might not find and crawl all your variations”. A global team that ships location-based redirects for conversion reasons can hide market versions from the crawler without any single market seeing the cause.

    What broke under devolution

    Failure How it happens How it shows up
    Technical drift Each market runs its own URL patterns, canonical logic and hreflang handling Wrong-country pages ranking (a UK page earning impressions in Australia when you filter the Performance report by country); hreflang annotations that are not returned, which Google ignores; “Duplicate, Google chose different canonical than user” on near-identical same-language market pages
    Shared-file collisions A market edits robots.txt, a redirect rule or a shared component for its own reasons Cross-market incidents traced back to one market’s change
    Risk spillover One market buys links, publishes unreviewed machine translation at scale, or hosts a third-party section Exposure under Google’s spam policies (link spam, scaled content abuse) on a domain every market shares. Hosted third-party sections raise site reputation abuse questions, and Google’s August 2026 update handles those differently inside and outside the EEA
    Incomparable measurement and duplicated spend Each market picks its own tools, agencies, KPIs and channel definitions No global view of what works, so a win in one market cannot be copied to another

    A split of decision rights that holds up

    Decision Owner Guardrail
    URL and folder conventions, hreflang generation, canonical logic Central Generated from one source of truth in the CMS, not hand-maintained per market
    robots.txt, redirects, rendering, template changes Central, with a pre-release SEO check Markets can request changes, with an agreed turnaround
    Language and country selection Central Suggest a better version with a banner; do not force a redirect between language versions
    Structured data and template modules Central builds, markets populate Validation in the release pipeline
    Keyword research, content priorities, local SERP analysis Market Shared page-type rules and topic taxonomy
    Translation versus local writing Market decides, central sets the quality bar No partially translated pages; machine translation reviewed by a fluent editor before publishing
    Local links, PR and partnerships Market One global link policy consistent with Google’s spam policies
    KPIs, channel definitions and dashboards Central defines, markets own their targets One definition of organic, brand and non-brand across all markets
    Market launches, closures and large content removals Joint Redirect maps and hreflang updates signed off centrally

    A practical detail that makes the split workable: give each market a URL-prefix Search Console property for its own folder, alongside the domain property the central team uses. Markets then see their own data without needing access to everything, and the central team keeps the whole picture.

    Signs the balance is wrong

    Signal Points towards Where to look
    Median days from market request to release keeps rising Over-centralised Ticketing system
    Workaround pages outside the standard templates Over-centralised A full crawl of the domain; Search Console properties nobody on the central team created
    Cross-market incidents caused by one market’s change Under-governed Incident log and release notes
    Hreflang errors or wrong-country impressions Under-governed Crawler checks; Performance report filtered by country and folder
    Smaller markets’ non-brand clicks falling while the largest market grows Priority bias at the centre Performance report per market folder

    Hypothetical example:

    A retailer runs 14 country folders on one.com, with a four-person central SEO team and a marketer in each market. The English-language folders for the UK, Australia and New Zealand are near-identical, and the Performance report shows UK pages taking impressions in Australia. Meanwhile the German team, tired of waiting, has an agency building campaign pages on a subdomain. The fix is not to pick a side. Hreflang generation and the subdomain question move firmly to the centre; product naming, delivery information and campaign content in Australia, New Zealand and Germany move firmly to the markets, built on the standard templates, with a two-week turnaround promise from the centre for template requests.

    Pilot the hybrid before reorganising everyone

    A reorganisation is expensive to reverse, so pilot it. Pick two or three markets, write the decision-rights table for them, and run it for two quarters, so the pilot covers several release cycles and at least one seasonal peak. Track request lead time, cross-market incidents, hreflang error rate, and non-brand clicks and conversions in the pilot markets against comparable markets outside the pilot. If lead time falls and incidents do not rise, extend it. If incidents rise, tighten the guardrails before scaling. If nothing moves, the constraint is probably resourcing rather than the model.

    Start with your own incident and request history

    Before redrawing the org chart, take the last year of SEO incidents and stalled market requests and ask two questions of each: how many markets did it hit, and who owned the decision? An incident that spread from one market’s change marks a decision to centralise. A single-market change that sat in the central queue marks one to hand back. That list is the first draft of your decision-rights table, and it gives the pilot a baseline drawn from failures your organisation has actually had.

    Need help with this on your own site?

    Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.

    See SEO services

Add a reply

Replies are open to approved members. Log in or apply to join.