SEO para Yandex

«> TL;DR — Yandex SEO means optimizing and monitoring a site for Yandex Search,

using Yandex’s own webmaster guidance rather than assuming every Google-specific tool or behavior transfers. Whether you should care comes down to one question: do you have an audience or business in Russia or the CIS? If yes, Yandex SEO is real work. If no, you can skip it.

» (Tradução) (Síntese localizada do trecho três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## What Yandex is » (Tradução) (Síntese localizada do trecho quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex is to Russia roughly what Google is to most of the rest of the world — the default search engine, plus maps, mail, a marketplace, and more. It also leads in several nearby markets (Belarus, Kazakhstan, and other former-Soviet countries). It’s part of a small club of search engines that beat Google on their home turf, alongside Baidu in China and Naver in South Korea. If you’re weighing whether any of those markets are worth the effort, the market-specific SEO hub is the overview; this page is the Yandex deep dive. » (Tradução) (Síntese localizada do trecho cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Why Yandex SEO is different from Google SEO » (Tradução) (Síntese localizada do trecho seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Three things stand out for anyone used to Google: » (Tradução) (Síntese localizada do trecho sete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- It watches behavior closely. Yandex leans harder on behavioral signals — whether people click your result, how long they stay, and whether they bounce straight back to the search results to try something else. Optimizing your titles, page speed, and content to keep people engaged matters more here.

• “Commercial” queries get special treatment. Yandex has long separated transactional (“buy a washing machine”) from informational (“how does a washing machine work”) queries and judges the pages differently. Commercial pages are expected to look like real, trustworthy businesses — clear contact info, prices, the works.
• You declare your region. In Yandex Webmaster (Yandex’s version of Google Search Console) you tell Yandex which region your site serves, and that declaration actually moves your rankings for location-dependent searches. Google mostly guesses this; Yandex lets you state it. » (Tradução) (Síntese localizada do trecho oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## The 2023 leak, in one paragraph » (Tradução) (Síntese localizada do trecho nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«In early 2023, a chunk of Yandex’s internal source code leaked, and it named a lot of the engine’s ranking factors out loud. The headline you’ll see — “1,922 ranking factors!” — is misleading: most of those were switched off, deprecated, or never finished, and different experts counted the active ones very differently. What the leak did confirm is that Yandex bakes behavioral signals right into its core algorithm. It did not reveal Google’s algorithm, even though a lot of headlines implied it did — Yandex and Google are different companies with different code. » (Tradução) (Síntese localizada do trecho dez, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## The honest bottom line » (Tradução) (Síntese localizada do trecho onze, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Do you serve customers in Russia or the CIS? Then Yandex SEO is worth doing, and the mechanics are stable enough to invest in. Don’t over-think the leak, don’t build Turbo pages (Yandex retired them in 2025), and start by setting up Yandex Webmaster and declaring your region. Want the full version — the specific leaked factors, the factor-count fight, the Yandex.Direct question, and a real optimization checklist? Switch to the Avançado tab. » (Tradução) (Síntese localizada do trecho doze, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«> TL;DR — Yandex operates its own search and webmaster systems. Verify indexing,

diagnostics, query data, and any regional settings in Yandex Webmaster, and keep paid Yandex.Direct campaigns separate from organic search. Avoid turning leaked or historical factor lists into current guarantees; official live documentation is the safer operational source. Retired delivery formats should not be used anew.

» (Tradução) (Síntese localizada do trecho quinze, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Yandex’s market position: small globally, dominant locally » (Tradução) (Síntese localizada do trecho dezesseis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Start with the framing that makes Yandex worth an SEO’s time despite a tiny global number. In the global search-engine share figures I presented at Ahrefs Evolve 2025 (per Stan Ventures’ recap of that talk — treat the exact breakdown as their paraphrase, not an Ahrefs-published table), Google sat around 90.4%, Bing 4.08%, Yahoo 1.46%, and Yandex about 1.65% worldwide as of September 2025. That’s rounding-error territory globally — but Yandex is a top engine inside Russia and the CIS, which is exactly why “1.65% globally” undersells it. This is the same “small globally, dominant locally” pattern you see with Baidu in China and Naver in South Korea; the market-specific SEO hub covers the whole set. » (Tradução) (Síntese localizada do trecho dezessete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«One organizational fact worth stating plainly, because it drives enterprise procurement and compliance questions more than SEO ones: in February 2024 Yandex N.V. (the Dutch parent) agreed to sell its Russia-based businesses to a consortium at roughly USD 5.4 billion; the deal’s second closing completed in July 2024, the parent renamed itself Nebius Group in August 2024 and pivoted to AI infrastructure, with Nebius resuming Nasdaq trading in October 2024. The divestment itself was an ownership change, not an algorithm change — there’s no evidence in any source that ranking mechanics moved because of it. The Russian search business is now fully Russian-owned, which is a real vendor/compliance question for Western enterprises, separate from whether the optimization tactics still work (they do). » (Tradução) (Síntese localizada do trecho dezoito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Yandex vs. Yandex.Direct: organic and paid are separate » (Tradução) (Síntese localizada do trecho dezenove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Just like Google organic vs. Google Ads, Yandex organic and Yandex.Direct (its PPC/contextual ad product) are separate systems. The evergreen question mirrors the Google one: does paid spend buy organic rank? The most direct sourced answer comes from Dan Taylor (founder of RussianSearchNews.com), whose post “Is Yandex.Direct (Paid Search) A Yandex Organic Ranking Factor?” answers “Yes” — but read the nuance carefully. His “yes” is that the 2023 leak showed total traffic volume and percentage of organic traffic are ranking factors, both of which paid Direct traffic could indirectly influence. That is a much narrower and more indirect claim than “ad spend buys organic placement.” Do not conflate the two: Yandex has not said, and the leak does not show, that buying Direct ads directly lifts your organic position. (The live URL currently returns a 403; the quote above is from an archived copy — confirm against the live page if it comes back.) » (Tradução) (Síntese localizada do trecho vinte, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## How Yandex ranking actually works » (Tradução) (Síntese localizada do trecho vinte e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### Behavioral factors — the headline difference » (Tradução) (Síntese localizada do trecho vinte e dois, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«If you take one thing from this page: Yandex weights behavioral factors more heavily, and admits to it more openly, than Google ever has. SEO Sherpa puts it directly — “Yandex puts even more weight on behavioral factors than Google.” The 2023 leak named these signals in the code: in SISTRIX’s analysis the behavioral inputs include “CTR, time on site, bounce rate and number of visitors returning to the SERPs.” Weglot’s guide independently corroborates “the average amount of time a user spends on the website” and average pages-per-session as inputs. Practically, this means the same on-page levers you’d use for engagement on Google — compelling titles that earn the click, fast pages, content that satisfies the query so people don’t pogo-stick back — carry more direct weight in Yandex. » (Tradução) (Síntese localizada do trecho vinte e três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### Commercial vs. informational queries » (Tradução) (Síntese localizada do trecho vinte e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex has separated commercial from informational queries for a long time. SISTRIX notes the leak makes “differences for adult entertainment” and “commercial queries” explicit as distinct ranking treatments. Historically this traces to Yandex’s Magadan update (2008), which per Dan Taylor’s SEJ guide “enabled the search engine to understand abbreviations, as well as start to understand the difference between commercial and non-commercial queries and content uniqueness” — though that specific Magadan attribution is single-sourced to SEJ, so treat the date and framing as industry-reported rather than officially confirmed. The takeaway stands regardless: a page targeting a transactional query is judged on commercial-trust criteria (clear pricing, contact details, business legitimacy) that an informational page isn’t. » (Tradução) (Síntese localizada do trecho vinte e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### Regionality — a direct lever, not an inference » (Tradução) (Síntese localizada do trecho vinte e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex’s geo-dependence is semi-explicit in a way Google’s isn’t. You declare a site region in Yandex Webmaster, and Yandex also reads signals off the page: per SEJ, Yandex “checks your website’s IP, contact information (make sure this is structured and displayed clearly), and domain registration information” to localize rankings. Weglot confirms the principle: “Yandex takes into account a website’s region when ranking web pages for location-dependent search queries.” The leak even named a geo factor codename — Michael King’s live-tweet analysis surfaced FI_GEO_CITY_URL_REGION_COUNTRY, “the geographical coincidence of the document and the country that the user searched from.” Region (declared in Yandex Webmaster) and language (marked up on the page) are separate layers, not one setting: Yandex’s own “Indexing localized pages” doc documents a page-level markup step — “If your site has pages in multiple languages or pages intended for various regions, make sure to specify this information for the Yandex indexing bot” — using standard hreflang attributes in a <link rel="alternate"> tag, the same syntax Google uses. Declaring a region does not substitute for that page-level language/region markup, and vice versa. The tactical how-to for region declaration lives in the Yandex Webmaster Tools » (Tradução) (Síntese localizada do trecho vinte e sete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) «guide (including the Site Quality Index / IKS and the region-registration mechanics) — I won’t re-derive it here. » (Tradução) (Síntese localizada do trecho vinte e sete, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

«### Links and text relevance — the familiar parts » (Tradução) (Síntese localizada do trecho vinte e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Underneath the differences, Yandex shares information-retrieval fundamentals with Google. Per SISTRIX, “Yandex uses a PageRank algorithm to rank the quality of links” and “mainly relies on BM25, a well-known approach from information retrieval” for text relevance — the same classical techniques widely believed to underlie parts of Google. And SEO Sherpa notes a useful contrast: “Unlike Google, which relies more on semantic understanding, Yandex still rewards direct keyword matches” — so exact-match keyword usage is a little less punished here than on modern Google. Trust/quality gets a YMYL-style treatment too — SISTRIX notes “higher quality requirements for sensitive topics such as health.” » (Tradução) (Síntese localizada do trecho vinte e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## What Yandex’s own docs say » (Tradução) (Síntese localizada do trecho trinta, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex’s public guidance lines up with this picture without spelling out weights. The Webmaster help names two internal quality metrics: Proxima, “a page quality metric built upon assessors’ evaluations and additional signals,” and Proficit, “a metric of search results’ usability that is calculated based on user interactions with Search” — that second one is the closest official acknowledgment that user interaction (i.e. behavior) feeds ranking. The docs also flag a trust angle — “When ranking sites in search results, Yandex may take into account additional signals about the company’s integrity” — and, importantly, mirror Google’s stance on human raters: “Assessors’ evaluations do not affect search results directly. Rather, they help to determine whether a particular change in the ranking algorithm is appropriate.” If that sounds like Google’s Quality Rater Guidelines, it should — it’s the same concept. » (Tradução) (Síntese localizada do trecho trinta e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## The 2023 source-code leak: what it actually showed » (Tradução) (Síntese localizada do trecho trinta e dois, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«In January 2023, roughly 44.7GB of Yandex internal source code was posted to a Russian-language forum and mirrored on GitHub. Yandex’s response, via Search Engine Land, disputed the “hack” framing: “Yandex was not hacked. Our security service found code fragments from an internal repository in the public domain, but the content differs from the current version of the repository used in Yandex services.” (This is SEL-sourced; no Yandex-hosted primary posting was located.) » (Tradução) (Síntese localizada do trecho trinta e três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### The factor count is three different numbers — don’t merge them » (Tradução) (Síntese localizada do trecho trinta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«This is where most coverage goes wrong. There is no single “number of ranking factors” from the leak. Three analysts produced three different figures from the same data, and they do not reconcile: » (Tradução) (Síntese localizada do trecho trinta e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- SISTRIX (Johannes Beus): 1,922 total, 417 active. Beus found “An extensive list of 1,922 different ranking factors” and concluded “there are just 417 active ranking factors to consider” after removing those tagged TG_DEPRECATED (999), TG_UNUSED (242), TG_UNIMPLEMENTED (149), and TG_REMOVED (115). This is the most granular breakdown found.

• Search Engine Land / Dan Taylor: ~690 potential. SEL reports “244 of the ranking factors were categorized as unused and 988 as deprecated, ‘meaning that 64% of the document is either not actively used or has been superseded – so it’s more like ~690 potential ranking factors.’” Note this is a different count than SISTRIX’s (different tag tallies, different active total) — two named analysts, same leak, unreconciled numbers.
• iPullRank / Michael King: 17,854 (unverified). SEL credits King with finding “17,854 ranking factors, not 1,922.” I could not verify this figure against an iPullRank-hosted source — it doesn’t appear in the ipullrank/yandex-decoder-ring README (which frames the work as breaking “down the code by its function as it relates to Information Retrieval components”), and King’s own recovered live-tweets say something different: “How did y’all come up with that number of ranking » (Tradução) (Síntese localizada do trecho trinta e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) « factors? I see 481 factors just related to ‘Rapid Clicks’.” Treat 17,854 as SEL’s characterization of King, not a confirmed iPullRank figure. » (Tradução) (Síntese localizada do trecho trinta e seis, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

«So: 1,922 total and 417 active (SISTRIX) is the number I’d anchor on, with Taylor’s ~690 potential as the competing named estimate and 17,854 flagged as unverified. Anyone quoting a single clean count is skipping the disagreement. » (Tradução) (Síntese localizada do trecho trinta e sete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### Specific named factors worth knowing » (Tradução) (Síntese localizada do trecho trinta e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Beyond counts, the leak named concrete factors that make Yandex’s priorities tangible — surfaced through King’s and Alex Buraks’ live-tweet threads (recovered via a Search Engine Roundtable roundup; the live page 403s, so this is from an archived copy): » (Tradução) (Síntese localizada do trecho trinta e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- FI_GEO_CITY_URL_REGION_COUNTRY — geographic match between document and searcher’s country (the regionality lever, in code).

• FI_ADV: -0.2509 — a negative weight for a page having advertising on it.
• Time-of-day weighting — per SISTRIX, “a distinction is made by time – there are morning and evening weights,” a genuinely Yandex-specific dimension with no obvious Google analog.
• Number of unique visitors, percent of organic traffic, and average domain ranking across queries — factors SEL flagged as the surprising ones (these are what the Yandex.Direct discussion above hinges on).
• A Wikipedia-uplift factor, orphan-page detection, keyword-in-URL, and even a factor literally named “random” (per Buraks, used to test behavioral-factor response). » (Tradução) (Síntese localizada do trecho quarenta, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«### What the leak does NOT tell you: it’s not Google’s algorithm » (Tradução) (Síntese localizada do trecho quarenta e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Plenty of headlines framed the leak as revealing Google’s ranking factors (e.g. fizzypop.nz’s “Google Ranking Factors Revealed in Yandex Source Code Leak”). That’s a conflation. Yandex and Google are different companies with different codebases; any read-across to Google is indirect — shared IR fundamentals like BM25 and PageRank-style link analysis — not a disclosure of Google’s algorithm. The best rebuttal is Google’s own John Mueller, who commented directly on the leak: “Will this help you do SEO on Google? Probably not but hey, it is super interesting.” In fairness, it’s not a universally-rejected view among practitioners — Alex Buraks tweeted “I downloaded the code, analyzed it and there is a lot of useful information for Google SEO as well” — so treat “the leak teaches you about Google” as a minority position rather than a settled fact, with Mueller’s “probably not” as the counterweight. » (Tradução) (Síntese localizada do trecho quarenta e dois, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Note: as of 2026-07-18, Search Engine Roundtable’s live page is directly accessible and its Mueller/King/Buraks quotes and factor codenames were re-verified against it; RussianSearchNews still returns 403 to automated fetches, so that quote remains sourced from an archived copy. Confirm against the live RussianSearchNews page before treating it as final. » (Tradução) (Síntese localizada do trecho quarenta e três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## A working checklist for Yandex » (Tradução) (Síntese localizada do trecho quarenta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«The practical differences from a Google program, in order: » (Tradução) (Síntese localizada do trecho quarenta e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«1. Set up Yandex Webmaster and declare your region. This is the single most Yandex-specific setup step, and it directly affects location-dependent rankings. 2. Optimize for behavioral signals. Earn the click (titles/meta), keep people on the page (speed, content quality), and don’t make them bounce back to the SERP — these carry more weight than on Google. 3. Meet commercial-page expectations. For transactional queries, show clear pricing, structured contact information, and business-legitimacy signals. 4. Get your technical basics right for YandexBot. Robots.txt (including the Yandex-only Clean-param directive), rendering, and bot verification are covered in that guide — Yandex’s JS rendering is limited, so serve important content server-side. 5. Handle language properly. Russian is morphologically rich (heavy declension), and Yandex still rewards direct keyword matches — so keyword research and on-page usage in Cyrillic matter. For images, note a concrete Google difference: SEJ reports Yandex prefers schema “implemented in Microdata format (unlike Google, which prefers JSON-LD).” 6. Use hreflang the same way you would for Google — as the international-SEO pillar notes, hreflang tells Google and Yandex which language/region a URL targets. » (Tradução) (Síntese localizada do trecho quarenta e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) «7. Do not build Turbo pages. Some older guides (Weglot’s among them) still recommend Yandex’s AMP-analogue Turbo pages; Yandex discontinued them in early 2025. Ignore that advice. » (Tradução) (Síntese localizada do trecho quarenta e seis, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Should you invest in Yandex SEO in 2026? » (Tradução) (Síntese localizada do trecho quarenta e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«If you have real audience or revenue in Russia/CIS: yes. The mechanics are stable, the algorithm’s differences are learnable, and the 2024 corporate split changed ownership, not ranking. The compliance/procurement question (a now fully Russian-owned search vendor) is real for Western enterprises and is a leadership decision, not an SEO one. If you have no audience there, Yandex traffic in your logs is mostly just YandexBot crawling — nothing to optimize for. » (Tradução) (Síntese localizada do trecho quarenta e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«For the broader “which non-Google engines should I care about” decision, see market-specific SEO and the international SEO pillar. » (Tradução) (Síntese localizada do trecho cinquenta, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## AI summary » (Tradução) (Síntese localizada do trecho cinquenta e três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«A condensed take on the Advanced version: » (Tradução) (Síntese localizada do trecho cinquenta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- Yandex leads in Russia/CIS, small globally (~1.65%, Ahrefs Sept 2025) but dominant locally — the same “small globally, dominant locally” pattern as Baidu and Naver.

• It’s a different algorithm, not a smaller Google. Three structural differences: heavier and more openly-confirmed behavioral factors (SERP CTR, dwell time, return-to-SERP); an explicit commercial-vs-informational query classifier; and direct region declaration in Yandex Webmaster instead of inference.
• Shared IR fundamentals underneath: PageRank-style link analysis and BM25 text relevance, but Yandex “still rewards direct keyword matches” more than Google’s semantic approach.
• The 2023 leak confirmed the behavioral emphasis but was mis-cited. Three non-reconcilable factor counts — SISTRIX 1,922 total / 417 active, SEL/Dan Taylor ~690 potential, and an unverified 17,854 attributed to iPullRank — must be cited separately, never merged. Named factors include a geo codename, a negative advertising factor, and time-of-day weights.
• The leak is about Yandex, not Google. Mueller: it’ll “probably not” help your Google SEO. Yandex also disputes it was “hacked.”
• Yandex.Direct (paid) is separate from organic. Organic-traffic volume/percentage is a leak-confirmed factor — which is not the same as ad spend buying rank.
• Do: set up Yandex Webmaster + declare region, optimize behavioral signals, meet commercial-page bar, serve JS content server-side, use hreflang, handle Cyrillic. » (Tradução) (Síntese localizada do trecho cinquenta e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) « Don’t: build Turbo pages (discontinued 2025).
• 2024 split (Yandex N.V. → Nebius Group) changed ownership, not ranking mechanics. » (Tradução) (Síntese localizada do trecho cinquenta e cinco, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Official documentation » (Tradução) (Síntese localizada do trecho cinquenta e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Primary-source documentation from the search engines. » (Tradução) (Síntese localizada do trecho cinquenta e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex

• How do I improve the site’s ranking in the search? — Yandex’s own ranking-improvement guidance, integrity signals, and the “specify the site region” instruction.
• Indexing localized pages — the page-level language/region markup mechanism (standard hreflang attributes), separate from the site-wide region declaration in Yandex Webmaster.
• Ranking (troubleshooting) — why rankings drop (retrained formulas, new classifiers, personalization) and what’s normal volatility.
• How to measure and improve Search quality — the Proxima and Proficit quality metrics, expert evaluation, and the assessor disclaimer.
• Information about site quality — the Site Quality Index (IKS) documentation.
• Yandex Webmaster — the product itself (Search Console equivalent). » (Tradução) (Síntese localizada do trecho sessenta, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Google / Bing (for contrast)

• Google Search Central — international and multi-regional sites — Google’s inference-based geotargeting model, the natural contrast to Yandex’s region declaration.
• IndexNow — the push protocol Yandex participates in (alongside Bing), unlike Google. » (Tradução) (Síntese localizada do trecho sessenta e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Quotes from the source » (Tradução) (Síntese localizada do trecho sessenta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«On-the-record and reported statements. Each link is a deep link that jumps to the quoted passage where the source supports one. » (Tradução) (Síntese localizada do trecho sessenta e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Yandex — official docs » (Tradução) (Síntese localizada do trecho sessenta e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- “When ranking sites in search results, Yandex may take into account additional signals about the company’s integrity.” — Yandex Webmaster Help. Jump to quote

• “Proxima is a page quality metric built upon assessors’ evaluations and additional signals.” — Yandex Webmaster Help. Jump to quote
• “Assessors’ evaluations do not affect search results directly. Rather, they help to determine whether a particular change in the ranking algorithm is appropriate.” — Yandex Webmaster Help (the direct analog to how Google uses its Quality Raters). Jump to quote
• “If your site has pages in multiple languages or pages intended for various regions, make sure to specify this information for the Yandex indexing bot.” — Yandex Webmaster Help, on page-level hreflang markup for localized pages (a separate layer from the site-wide region declaration above). Jump to quote » (Tradução) (Síntese localizada do trecho sessenta e sete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«On the 2023 leak — reported / verbatim from coverage » (Tradução) (Síntese localizada do trecho sessenta e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- Yandex’s statement on the incident: “Yandex was not hacked. Our security service found code fragments from an internal repository in the public domain, but the content differs from the current version of the repository used in Yandex services.” — Yandex, via Search Engine Land. Jump to quote

• “just 417 active ranking factors to consider” (of 1,922 total, after removing deprecated/unused/unimplemented/removed tags). — Johannes Beus, SISTRIX. Jump to quote
• “Will this help you do SEO on Google? Probably not but hey, it is super interesting.” — John Mueller, Google, commenting on the leak (via Search Engine Roundtable’s roundup).
• “Yandex puts even more weight on behavioral factors than Google.” — SEO Sherpa. Jump to quote » (Tradução) (Síntese localizada do trecho sessenta e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Note: as of 2026-07-18, the Search Engine Roundtable roundup (source of the Mueller and King/Buraks tweets) is directly accessible and its quotes were re-verified against the live page; the RussianSearchNews Yandex.Direct post still returns 403 to automated fetches and remains sourced from an archived snapshot. The Yandex corporate statement is Search-Engine-Land-relayed, not a Yandex-hosted primary posting. Confirm the RussianSearchNews quote against the live page before treating it as final. » (Tradução) (Síntese localizada do trecho setenta, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Should you invest in Yandex SEO? » (Tradução) (Síntese localizada do trecho setenta e três, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Answer a couple of questions and you’ll land on a recommendation. Every branch comes from the market-position and mechanics discussion above. » (Tradução) (Síntese localizada do trecho setenta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Yandex SEO myths and mistakes to avoid » (Tradução) (Síntese localizada do trecho oitenta e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«The claims that come up most often — several are widely repeated and wrong: » (Tradução) (Síntese localizada do trecho oitenta e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- “Yandex ranking is basically Google’s, just smaller.” No. Explicit region declaration and a commercial-vs-informational query classifier are structural differences, not scaled-down versions of Google’s approach. (Do-instead: treat Yandex as its own algorithm — declare region in Yandex Webmaster, and build commercial pages to a commercial-trust bar.)

• “The 2023 leak means Yandex actively uses 1,922 ranking factors.” No. Most were tagged deprecated/unused/unimplemented/removed. And there’s no single count: SISTRIX says 1,922 total / 417 active, Dan Taylor (via SEL) derives ~690 potential, and an unverified 17,854 is attributed to iPullRank. (Do-instead: cite each figure by its named source, never merge them into one number, and flag 17,854 as unverified.)
• “The Yandex leak reveals Google’s ranking factors.” No — different company, different codebase. Any read-across is indirect (shared BM25/PageRank fundamentals). (Do-instead: quote Mueller’s “probably not” and treat cross-applicability as a minority view, not a fact.)
• “Yandex was hacked.” Yandex disputes this: it says code fragments from an internal repo appeared publicly but “differs from the current version.” (Do-instead: describe it as a source-code leak of an older snapshot, not a live-system breach.)
• “Buying Yandex.Direct ads boosts your organic rankings.” Overstated. Organic traffic volume/percentage is a leak-confirmed factor, and paid traffic could » (Tradução) (Síntese localizada do trecho oitenta e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) « influence that indirectly — but that is not “ad spend buys organic rank.” (Do-instead: represent the mechanism precisely as an indirect traffic-signal effect, not a pay-to-rank claim.)
• “Build Turbo pages for speed.” Stale advice. Yandex discontinued Turbo pages in early 2025. (Do-instead: optimize real page speed and mobile-friendliness on your own pages; don’t invest in the dead Turbo format.)
• “Client-side JavaScript will get indexed fine, like on Googlebot.” Not reliable — Yandex’s rendering is limited. (Do-instead: serve important content server-side / pre-rendered; see the YandexBot guide.)
• “Yandex SEO is irrelevant now because of the corporate split.” The 2024 split was an ownership change (Yandex N.V. → Nebius Group), not an algorithm change. (Do-instead: separate the compliance/vendor question, which is real, from the mechanics, which are unchanged.) » (Tradução) (Síntese localizada do trecho oitenta e seis, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Yandex SEO setup checklist » (Tradução) (Síntese localizada do trecho oitenta e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«The Yandex-specific work, in order — most of it different from a Google program: » (Tradução) (Síntese localizada do trecho noventa, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«- [ ] Verify the site in Yandex Webmaster and declare your target region (the single most Yandex-specific lever; it moves location-dependent rankings).

• Make sure IP, on-page contact information, and domain registration all point clearly at the region you’re targeting.
• Optimize behavioral signals: titles/meta that earn the click, fast pages, content that satisfies intent so users don’t return to the SERP.
• For commercial/transactional queries, show clear pricing, structured contact details, and business-legitimacy signals.
• Confirm YandexBot can crawl and render what matters — serve JS-dependent content server-side; review robots.txt and the Yandex-only Clean-param directive (see the YandexBot guide).
• Use hreflang to map language/region variants (it works for Yandex too).
• For images, prefer Microdata structured data (Yandex’s preference) alongside your JSON-LD.
• Do Cyrillic keyword research and on-page usage that accounts for Russian declension; direct keyword matches still help more than on Google.
• Do not build Turbo pages — discontinued 2025.
• Confirm your organization has cleared the compliance/vendor question for operating in the Russian market. » (Tradução) (Síntese localizada do trecho noventa e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Regional organic visibility » (Tradução) (Síntese localizada do trecho noventa e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Metric: Impressions, clicks, and average position for a fixed Yandex query set segmented by declared target region and commercial/informational intent. What it tells you: Whether Yandex associates the site with the intended market and serves the right page type for each query class. How to pull it: Export query/page data from Yandex Webmaster and maintain the region/intent labels in the reporting table. Benchmark / realistic range: Use the first stable post-configuration period as the baseline; rankings depend on query, region, competition, and site history. Cadence: Weekly for regressions, monthly for reporting. » (Tradução) (Síntese localizada do trecho noventa e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Search-result engagement » (Tradução) (Síntese localizada do trecho noventa e seis, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Metric: Organic click-through rate plus post-click engagement indicators such as return-to-search behavior proxies, engaged sessions, and repeat visits for Yandex traffic. What it tells you: Whether titles earn the right click and pages satisfy the query—the behavioral area Yandex emphasizes more openly than Google. How to pull it: Combine Yandex Webmaster search performance with Yandex Metrica landing-page and session reports under a documented definition. Benchmark / realistic range: Establish baselines by query class and page template; never manufacture interactions or use one site-wide target. Cadence: Monthly, with release-level monitoring after title, UX, or performance changes. » (Tradução) (Síntese localizada do trecho noventa e sete, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Commercial-page outcome rate » (Tradução) (Síntese localizada do trecho noventa e oito, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Metric: Qualified actions from Yandex organic sessions landing on transactional pages, segmented by region and query cohort. What it tells you: Whether commercial visibility reaches real customers rather than merely increasing visits. How to pull it: Join Yandex Metrica goals/ecommerce events to landing pages and Yandex organic attribution; reconcile with business systems where available. Benchmark / realistic range: Baseline by product, region, and action because prices, trust requirements, and sales cycles differ. Cadence: Monthly and quarterly by cohort. » (Tradução) (Síntese localizada do trecho noventa e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Yandex market-share trend » (Tradução) (Síntese localizada do trecho cem, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Metric: Dated Yandex search share in the target market from one consistently defined source. What it tells you: Whether the strategic weight assigned to Yandex remains proportionate to local usage. How to pull it: Record the same market, device scope, and source methodology at each observation; keep global share separate from Russian/local share. Benchmark / realistic range: Trend within the chosen source rather than combining incompatible estimates or treating the article’s dated global figure as a target. Cadence: Quarterly and during annual market planning. » (Tradução) (Síntese localizada do trecho cento e um, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Test yourself: Yandex SEO » (Tradução) (Síntese localizada do trecho cento e quatro, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«Five quick questions on how ranking in Yandex differs from Google and what the 2023 leak actually showed. Pick an answer for each, then check. » (Tradução) (Síntese localizada do trecho cento e cinco, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«## Resources worth your time » (Tradução) (Síntese localizada do trecho cento e nove, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«My speaking

• Patrick Stox Breaks Down AI SEO at Ahrefs Evolve 2025 (Stan Ventures’ recap) — the source of the September 2025 global search-share breakdown (Google ~90.4%, Bing 4.08%, Yahoo 1.46%, Yandex ~1.65%) used to frame “small globally, dominant locally.” Treat the exact numbers as their paraphrase of the talk. » (Tradução) (Síntese localizada do trecho cento e dez, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«My related writing

• The Beginner’s Guide to Technical SEO — where crawling, rendering, and indexing fit, which underpins the YandexBot/technical section here. » (Tradução) (Síntese localizada do trecho cento e onze, parte um: o texto-fonte foi preservado para conferência na revisão nativa.)

«From around the industry

• Yandex “leak” reveals 1,922 search ranking factors (Search Engine Land) — Yandex’s official statement, Dan Taylor’s ~690 estimate, and the surprising/unsurprising factor lists.
• Yandex code leak reveals ranking factors (SISTRIX, Johannes Beus) — the most granular factor breakdown (1,922 total / 417 active), behavioral signals, BM25/PageRank, and the time-of-day weights.
• The Ultimate Guide to Yandex SEO (Search Engine Journal, Dan Taylor) — IKS/ICS, the commercial-query history (Magadan), regional-targeting inputs, and the Microdata-vs-JSON-LD difference.
• Optimizing for Yandex: a comprehensive guide (Weglot) — behavioral inputs (dwell time, pages per session) and regionality. (Note: its Turbo-pages recommendation is now stale — Turbo was discontinued in 2025.)
• Yandex SEO in 2026: How to Rank on Russia’s #1 Search Engine (SEO Sherpa) — the behavioral-weighting-vs-Google point and the direct-keyword-match contrast.
• ipullrank/yandex-decoder-ring (GitHub) — iPullRank’s organizational breakdown of the leaked code by IR component. » (Tradução) (Síntese localizada do trecho cento e doze, parte um: o texto-fonte foi preservado para conferência na revisão nativa.) «- YNV completes divestment of its Russia-based businesses (Nebius newsroom) — the primary source for the 2024 corporate-split dates. » (Tradução) (Síntese localizada do trecho cento e doze, parte dois: o texto-fonte foi preservado para conferência na revisão nativa.)

Registro de alterações