Knowledge base v2.13.0 · Community edition · data through 2 October 2026
Day 2 · Thursday 1 October 2026 · 14:30
Alizée Baudez (International SEO Consultant) on language, region and authority, named from her self-introduction, which matches the official speaker list; then Zahra Laleh's talk on non-Latin scripts, presented on her behalf by a colleague who was not named; a second attendee recording covers the rest of that talk.
In a community speaker's audit case, a premium multifunctional-furniture maker about to be acquired had multilingual sites with clean translations by native speakers and hreflang tags that were in fairly good shape.
Speaker Alizée BaudezEvidence transcript
In the same furniture-maker audit, the Dutch site had only a fraction of the conversions of the brand's other language sites, although it showed the same catalogue, the same hero products and the same photography.
Speaker Alizée BaudezEvidence transcript
Checking a competitor during the furniture-maker audit, a community speaker found that a large furniture retailer's French and Dutch sites differ a little in the products they show and especially in their photography, with items apparently shown in much smaller rooms on the Dutch site.
Speaker Alizée BaudezEvidence transcript
A community speaker said her first guess, that Dutch homes are smaller than French ones, was wrong: Dutch and French homes have about the same floor area per person, and what differs is the type of house.
Speaker Alizée BaudezEvidence transcript
A community speaker said about 19% of homes in Europe are terraced houses on several floors, against 58% in the Netherlands; the matching Eurostat figures are 2019 shares of the EU and Dutch population living in semi-detached or terraced houses, not shares of homes.
Speaker Alizée BaudezEvidence transcript
In a community speaker's comparison, a typical Dutch home is narrow and spread over several levels, often with very steep, narrow staircases that make furniture harder to install upstairs, while a typical French home has a wide ground floor and usually wide stairs.
Speaker Alizée BaudezEvidence transcript
The lesson of the furniture-maker case was that the translation was correct but the target region had never been checked.
“the translation was right, but the region was never checked”
Speaker Alizée BaudezEvidence transcript
A community speaker framed international SEO around three signals to settle before working on hreflang: language (what people can read), region (who the content is for) and authority (why local people should trust the site).
Speaker Alizée BaudezEvidence transcript
A community speaker said language versions need adjusting for regional variants of one language: 90 is 'quatre-vingt-dix' in France but 'nonante' in Belgium, and Swiss German differs from German, for example in whether the letter ß (Eszett) is used.
Speaker Alizée BaudezEvidence transcript
Settling the region means deciding who the content is actually for, which shapes how the catalogue is organised, how prices look and which payment options each market gets.
Speaker Alizée BaudezEvidence transcript
Of language, region and authority, a community speaker called language easy to settle and region solvable with a little research, while authority (why local people should trust a brand) has to be earned and built over time.
Speaker Alizée BaudezEvidence transcript
A community speaker stressed that an hreflang annotation for a country (her example was German for Austria, de-AT) does not by itself make a site established or trusted in that country; local authority has to be worked for.
Speaker Alizée BaudezEvidence transcript
A community speaker cited a Semrush study that found at least one hreflang error on 75% of multilingual websites; the study, reported in February 2017, audited 20,000 sites with several language versions (the speaker recalled it as about five years old).
Speaker Alizée BaudezEvidence transcript
A community speaker said the SEO industry has treated hreflang as the hard part of international SEO; hreflang is complicated, but the real hard part is the homework of understanding consumers in each market.
“the hard part is actually doing the homework to understand your consumers”
Speaker Alizée BaudezEvidence transcript
A community speaker said the answers about what consumers in each market want expire, because regional differences change over time.
Speaker Alizée BaudezEvidence transcript
A community speaker showed a chart of search interest over time for ceiling fans in northern versus southern Europe, saying southern Europeans commonly have them at home while northern Europeans paid them no attention until the previous summer's heat waves (the talk was on 1 October 2026).
Speaker Alizée BaudezEvidence transcript
The first of three questions to ask before the technical work of international SEO is which languages, and which language variants, the site actually serves, keeping languages and variants apart.
Speaker Alizée BaudezEvidence transcript
The second question before international SEO work is what genuinely differs between the regions of each language, such as photography, content, vocabulary, the way products are presented, pricing, catalogues and payment options.
Speaker Alizée BaudezEvidence transcript
The third question is whether you can name three local sources that already treat the brand as present in a country; if you cannot, a community speaker advised not to expand there but to fix the markets you already serve.
“don't bother expanding to another country; just fix the ones you already have”
Speaker Alizée BaudezEvidence transcript
A community talk, presented on its author's behalf by a colleague, was about multilingual SEO for languages written in non-Latin scripts, such as Persian and Arabic.
Speaker a second community speakerEvidence transcript
Arabic and Persian contain letters that look identical to users but are different characters to software, with different Unicode code points; the example given was the letter ye, which has an Arabic and a Persian form.
Speaker a second community speakerEvidence transcript
Because Arabic and Persian lookalike letters have different code points, a searcher may type one variant of a word while a website's text uses the other.
Speaker a second community speakerEvidence transcript
Lookalike Arabic and Persian character variants cause problems for data analysis and keyword research: data for one term can be split across the variants, which makes keyword research less accurate.
Speaker a second community speakerEvidence transcript
Persian and Arabic are written right to left and English left to right, so a title that mixes the two scripts can display in a confusing, unpredictable order even when its content is correct.
Speaker a second community speakerEvidence transcript
Used byrequirement DEV-INT-12glossary term Bidirectional text
The display problem of mixing right-to-left and left-to-right text also affects product titles and URLs, the presenter of the non-Latin-script talk said, showing an example from a large Iranian e-commerce site.
Speaker a second community speakerEvidence transcript
Used byrequirement DEV-INT-12glossary term Bidirectional text
Users of non-Latin-script languages do not always search in their own script: the same Persian query may be typed in Persian script or in Latin letters, with the same intent and the same expected results.
Speaker a second community speakerEvidence transcript
Used byrequirement DEV-INT-11glossary term Transliterated queries
Keyword research for non-Latin-script languages should ask how people actually type and spell in local search, not only what the right keyword is, because switching keyboards is inconvenient, especially when typing on mobile.
Speaker a second community speakerEvidence transcript
The presenter of the non-Latin-script talk said that in competitive Persian, Turkish and Arabic searches, bought backlinks and paid editorial content still visibly influence rankings and are widespread (an observation; no data was shown); the presenter stressed this described the situation and was not a recommendation.
Speaker a second community speakerEvidence transcript
The presenter of the non-Latin-script talk said Google announced in October 2023 that it had improved its spam-detection coverage for languages: the spam policy was global, but the coverage improvement was language-specific.
Speaker a second community speakerEvidence transcript
The presenter of the non-Latin-script talk said Persian offers far fewer natural link opportunities than English, with fewer publications, niche blogs and websites.
Speaker a second community speakerEvidence transcript
For SEO in languages such as Persian, the presenter of the non-Latin-script talk advised accepting different competitive dynamics and basing strategy on the actual maturity of that language's market, not only on the global policy.
Speaker a second community speakerEvidence transcript
Google search features often launch in some languages or countries first and expand later (the presenter's example was site names, launched in several languages and then extended to all languages in 2023), so comparisons of performance across languages and countries should not assume a feature is live everywhere at once.
Speaker a second community speakerEvidence transcript
In the presenter's observation, AI answers to Persian queries are written in Persian but cite some English sources; the presenter explained that Persian content on a topic is often thinner, of lower quality or less relevant, so the systems retrieve from languages with better content, which the presenter called cross-lingual retrieval.
Speaker a second community speakerEvidence transcript
The non-Latin-script talk concluded that multilingual SEO is not just translation: script, language behaviour and market maturity also matter, and basic SEO tactics, though the same everywhere, must be fitted to each market.
Speaker a second community speakerEvidence transcript
The presenter of the non-Latin-script talk recalled that Google officially treats buying links for ranking as spam (the year of the announcement was not clear in the recording).
Speaker a second community speakerEvidence transcript
Google's documentation recommends hreflang annotations even when a site's versions differ only by small regional variations within one language, for example English-language content targeted to the US, GB and Ireland.
Publisher Google Search Central
Used byrequirement DEV-INT-08
Google's documentation lists the signals it uses to decide which locale a page targets: ccTLDs, hreflang statements, server location, and other signals such as local addresses and phone numbers, local language and currency, links from other local sites and Business Profile signals.
Publisher Google Search Central
Used byrequirement DEV-INT-08
Google's October 2023 spam update post (4 October 2023) says the update improved coverage in many languages and spam types, cleaning up spam reported in Turkish, Vietnamese, Indonesian, Hindi, Chinese and other languages, particularly cloaking, hacked, auto-generated and scraped spam.
Publisher Search Central blog (4 October 2023)
Google's site names were introduced on mobile results for select languages in October 2022, expanded to desktop in March 2023, and on 7 September 2023 became available in all languages where Google Search is available, on mobile and desktop.
Publisher Search Central blog (7 September 2023)
Google's post on multilingual searches (8 September 2023) says that, because of typing difficulty on some keyboards, a person in India might search in Hindi using Latin rather than Devanagari characters and want and receive Hindi results written either way.
Publisher Search Central blog (8 September 2023)
Used byrequirement DEV-INT-11glossary term Transliterated queries
When translations and hreflang are clean but one country version converts far worse, compare that market's product range, photography and delivery or installation constraints with local competitors before changing anything technical.
Author Ibrahim Anjro
Re-check regional demand for core product categories at least once a year, for example by comparing regions and seasons in Google Trends, because climate, housing or habits can shift what a market wants.
Author Ibrahim Anjro
Treat hreflang as a way to show the right language version, not as a market-entry signal: before launching a country version, make sure local publications, partners or directories already mention the brand, and localise pricing and payment options.
Author Ibrahim Anjro
List language variants such as fr-FR and fr-BE or de-DE and de-CH separately in a site's language inventory, and give each its own vocabulary and spelling review instead of reusing one translation for every country.
Author Ibrahim Anjro
For Persian and Arabic keyword research, look up each lookalike-character spelling of a term separately and add up the volumes, and check which variant the site's own content uses.
Author Ibrahim Anjro
The October 2023 spam update cited in the non-Latin-script talk names neither Persian nor Arabic (only 'other languages') and lists cloaking, hacked, auto-generated and scraped spam, not link spam, so it does not show that the bought links the presenter saw working in Persian or Arabic search were addressed.
Author Ibrahim Anjro
A community speaker said language versions need adjusting for regional variants of one language: 90 is 'quatre-vingt-dix' in France but 'nonante' in Belgium, and Swiss German differs from German, for example in whether the letter ß (Eszett) is used.
Language versions can also be language-plus-region variants: Google's example site had a generic English page (en), a UK English page (en-gb) and a Spanish page (es), each on its own URL.
Users of non-Latin-script languages do not always search in their own script: the same Persian query may be typed in Persian script or in Latin letters, with the same intent and the same expected results.
Google usually understands non-English words typed 'in English' (probably meaning romanised, Latin-letter spellings, which the speaker did not spell out); the speaker said this belongs to query interpretation, a Day 3 (serving) subject.
In the presenter's observation, AI answers to Persian queries are written in Persian but cite some English sources; the presenter explained that Persian content on a topic is often thinner, of lower quality or less relevant, so the systems retrieve from languages with better content, which the presenter called cross-lingual retrieval.
Users often assume AI is all-knowing and borderless, but AI is still language-dependent: if an AI answer is synthesized from the top results, a query in another language draws on a totally different set of data.