SEO is an area in which everything is extremely fluid. It’s so dynamic that if you don’t deliberately follow trends, you can easily lose sight of something important.

Many changes can happen in just one month. Just last week, on March 23, Yandex launched an algorithm - Baden-Baden, aimed at combating over-optimization of texts. Such texts are no good - they have no benefit, let alone meaning: neither people nor smart search engines like them.

This news has brought horror to many website owners, because such spam can be found on almost every second resource. Is SEO a thing of the past now?

The last century is the use of a huge number of keywords as necessary and wherever necessary - thoughtlessly.

As we have already found out, there is nowhere in our field without changes. But SEO also has its so-called classics: something that will always be important and unchanged. Or at least 10 years. After all, 10 years for SEO is an eternity.

In connection with the launch of Baden-Baden, the topic of text ranking in search engines is becoming relevant today. This phenomenon is described in detail in an article for the Russian Seminar on Evaluation of Information Retrieval Methods (hereinafter referred to as ROMIP-2006). Despite the fact that the algorithms have become significantly more complex and now make much more demands, the base remains the base - still unchanged. As a result of the study, essentially terms were formed that can and should be adhered to to this day, and they are unlikely to change for another 10 years.

The essence of the study was as follows: based on the factors determining the textual relevance of the user's query page, an experimental ATR search was created from scratch to improve the ranking of documents in Yandex. The goal of ATR is to ensure that documents are perfectly matched to the request.

In the experiment, for each request, the document value was calculated - a numerical indicator of the document’s compliance with the request.

This is what the universal ranking formula looks like:

The formula includes the most important ranking factors - components, which we will discuss in more detail below.

List of indicators-factors of textual relevance taken into account in the formula of the ROMIP-2006 article

There are 5 of them in total:

1. Word frequency ( lemmas- the basic form of the word) according to the text.
2. Using pairs of words from the query.
3. Use of all query words (rare words give a plus to the overall indicator).
4. Using direct entry for long queries.
5. Using sentences that contain a significant number of query words.

Let's look at each of them in more detail.

1. Word frequency (W single)

The indicator that most influences search results. The concept of “lemmatization” is used here - bringing all word forms from the text to the original one.

For example:
Red, red, red → red
Bought, buy, bought → buy
Beds, bed, bed → bed

Accordingly, the search result without lemmatization will differ significantly from the search result taking into account this parameter.

Today, this is necessarily taken into account when checking text nausea with any of the services. You can immediately see which words were prioritized during promotion and how you can influence the optimization: add or remove keywords from the text.

Here is an example from the Advego.ru service. Go to the “Seo text analysis” section and paste your text into the empty window. Below we look at the “Words” table: all lemmas, their number and frequency are displayed there.

The logical conclusion: overspam, especially with direct entries, is evil. Already in 2006, search engines were able to understand the intentions of the authors of texts and distinguish between absurdly entered queries. What can we say about today, when Yandex and Google have moved to a new level using artificial intelligence - they see the meaning and benefits of the text.

For example, the new Baden-Baden Yandex algorithm severely punishes over-optimization: sites that abuse keywords face a drop of 10-40 positions, depending on the competition for the request. The algorithm has been tested for a long time, so it will not be possible to deceive artificial intelligence. The right way out is to be guided by the meaning when writing any text, expand your vocabulary with synonyms, do not forget about the USP, master the techniques of LSI copywriting and modify/rewrite texts according to its canons. As stated in the official Yandex notification: “Think about users, write for people, and Baden-Baden will not touch your site.”

Examples of texts when the frequency of occurrence of words in a document is exceeded:

It is immediately clear that the texts were written for the sake of keywords. There is simply nothing useful for a potential buyer in them.

It would be at least strange if in ordinary life we ​​heard a similar phrase from our interlocutor: “Looking at oil paintings of the sea, you can get moral peace” =). Therefore, before you write the key into the proposal, think about whether you would say this to your client in personal communication.

Good example:

There is no doubt that the keywords are included here naturally. " We provide assistance in organizing training in China“- no suspicions arise.

Write natural texts. 10 keywords for 500 characters is too much. Spam phrases and absurdly entered keys do not inspire confidence even among search engines with artificial intelligence, let alone a sane person.

Particular weight is given to the correct inclusion of keywords in the first paragraph, title, description, h1-h3

Not only the number of keys is important, but also their location. As you already understood from the subtitle, the words in the first paragraph of the text and the h1-h3 tags have significant weight. In tags and headings, it is better to place the keys at the beginning.

2. Occurrence of word pairs from the query (W pair)

This is a working technique, but we use it whenever possible. We combine and combine phrases.

Simply put, use keywords in phrases. Try to do several different keyword combinations. This will expand your audience reach and make the page more relevant to the queries that fall into your pairs.

3-4. The presence of all query words in the text and the presence of an exact occurrence of the phrase for multi-word queries (W phrase + W allwords)

Here you need to focus on the semantic core, take into account new search queries and their updating. You need words from potential queries that may be popular in the future. It is acceptable to focus on competitors from the top search results.

For the experimental search, all words from the search phrase were used because this gave +0.2 to the ranking. If this rule is not followed, the bonus will decrease to 0.03. But this indicator is probably not taken into account in the main search.

Here again diversity is important: there should be as few identical forms of words as possible.

Also, do not forget to use the phrase once in its exact occurrence, as we have already said, preferably somewhere at the beginning of the text.

5. The presence of sentences containing a significant number of query words (W halfphrase)

In addition, the number of sentences in the text containing a significant number of query words is taken into account - more than half of all words, taking into account part of the query. Do not forget that in one sentence they should be connected in meaning, and not used anyhow.

In order not to complicate your task when writing and not to think about each key, it is more convenient to first write the text itself, and then think about how to beautifully insert the key, then analyze all word forms and occurrences. Then check yourself against the list to see if all the conditions are met. But without fanaticism. After all, texts are written primarily for people, and then only for machines. Moreover, machines are becoming smarter every day - a lot has already been said about LSI copywriting, the vast majority actively use it.

If you need help with writing or want to find out whether the text on your website meets the requirements of search engines and copywriting rules, please contact us, we will give recommendations or write a new high-quality text.

A complete list of 200 ranking factors for the Yandex search engine. Use the checklist that will help bring your website to the TOP of search engine results.

Yandex ranking factors is a set of metrics and characteristics used by the search algorithm to determine the relevance of web resources to search queries for their appropriate sorting in search results.

Ranking factors in simple terms

To put it simply, these are the characteristics of your site, which determine its position in search results. Understanding ranking factors allows you to carry out effective SEO optimization of your website, as well as avoid sanctions from search engines. This topic has been hotly discussed by webmasters since the advent of search engines, since there is very little officially confirmed data on this topic.

It’s worth starting with the fact that there is no complete list of factors influencing the ranking of sites. The algorithms of search robots change along with changes in the number and weight of factors in the general ranking formula. As search engines have evolved, the number of ranking factors has steadily increased. If in the early 2000s sites could be brought to the TOP by influencing 3-4 metrics of your site, today there are hundreds of factors, each of which to a limited extent influences the position of a web page for each individual search query. In this regard, in Yandex there was a natural replacement of the linear TIC indicator with a more complex one (ICS).

Even officially confirmed factors are usually not explained in detail by the search company. Yandex only highlights the general principles of ranking, which will improve the quality of the web resource. The determination of most of the qualitative and quantitative characteristics that affect the issue occurs experimentally.

We have compiled a list of factors that can influence a website’s ranking to varying degrees. In order not to mislead you, we will separate officially confirmed and generally accepted information from hypotheses that have shown their effectiveness in practice.

- confirmed information;

- unconfirmed ranking factors.

You also need to understand that there are both upward and downward signals for ranking documents. We will analyze this in detail for each point.

List of Yandex ranking factors

To simplify working with factors, we have systematized them into spheres of influence, each of which we will consider separately.

Domain Ranking Factors

1. Domain name age. The older the domain, the more authoritative it is for search engines. This is logical, because the owner, at least throughout the entire existence of the site, paid for hosting and domain, confirming the importance of his resource for the network.
2. Domain history. Too frequent changes in ownership, which usually entails a change in DNS records, as well as site themes if the domain was sold in a domain store, can negatively affect rankings. You can check the history of a domain using this service.
3. Domain name. You can read more about this in our article. There is an opinion among optimizers that the exact occurrence of a keyword in a domain name increases the relevance of a web resource.
4. Domain zone. According to official Yandex data, the domain zone is not included in the general website ranking formula. However, it can indirectly influence the behavioral factors and capabilities of your business. This information is described in more detail in the article.
5. Domain renewal period. The time range for which a domain is purchased can serve as a signal of the seriousness or frivolity of its owner’s intentions to develop his resource. There are no compelling arguments to take this factor seriously. The maximum renewal period for domains in the and [RF] zones is 1 year, in the and zones - 10 years. The logic of this dubious factor automatically deprives domains in the and [RF] zones of the possibility of renewal for a longer period.
6. Domain Santskii. Pessimization in search engines and sanctions are assigned specifically to domains. If you are purchasing a new domain, or cannot find the reasons for the low ranking of your resource, try checking your domain for sanctions.
7. Hide WHOIS data. There is an opinion that companies with open data are more trusted by search engines than companies and individuals who hide their personal data in information about the domain owner. Regarding this factor, there are also no weighty arguments and facts that could indicate its validity.

Technical Ranking Factors

1. Server response code. The server response code and the accessibility of the site as a whole largely determine the level of quality of the site. If your server is periodically unavailable to users and your site does not return a 200 response code, this is a serious reason for pessimism about your site. The results should be relevant to user queries, and if users can’t even open your site, then there’s nothing for them to do in the search. You can check your server's response code in the Yandex Webmaster panel.
2. Server response speed. This indicator has nothing to do with the speed of the site itself or the user’s Internet speed. It shows how quickly the server begins to provide information after a client request. The faster the server's response speed, the faster the user begins to see content being loaded onto the page. A good server response rate is 300-400 ms. You can also check the server response speed for your resource in the Webmaster panel.
3. Availability of a main site mirror. Your site should have a main mirror through which it is indexed and promoted in search. You can determine the main mirror of a site, firstly, by specifying it in the Host directive in the robots.txt file. Secondly, by selecting it in the section Indexing → Site relocation in the Webmaster panel. In addition, from all non-main mirrors it is necessary to configure 301 redirects to the main mirror. The site must be accessible from one URL.
4. Image weight. Images are the most commonly used media files that seriously affect page loading speed, especially for users with weak Internet connections. Unoptimized images can seriously hinder your customers' experience on your site. Try to optimize images on the site not only by size, but also by weight.
5. CSS in the page code. Using styles in the HTML code of a page is bad form and should be avoided. Place all css code in a separate file.
6. JS in the page code. The situation is similar with JavaScript. Whenever possible, include the necessary external scripts in the head tag rather than using them directly in the HTML. Search robots analyze not only text and link information on the site, but also the code.
7. Page code compression. Try to minify the HTML code of the page so that it takes less time for search robots to process it.
8. Robots.txt. The most important txt file for search engines, which contains directives and instructions for allowing or prohibiting indexing of certain sections, pages, records. It must contain the Host directive, pointing to the main mirror of the site, as well as the path to the xml site map. The absence of this file can seriously slow down the indexing speed of your site, and its incorrect configuration can exclude useful pages and sections from the search results.
9. Sitemal.xml. The site map is the first place where search robots learn about new pages on the site. Not using sitemap.xml can seriously slow down the indexing speed of your pages. The importance of using a site map is confirmed by a separate section in the Yandex Webmaster panel. By the way, don’t forget to add your sitemap there.
10. Content display speed. After the browser has received response code 200 from the server, the page and its contents begin to load: media files, styles, scripts, etc. The speed at which content is displayed depends largely on the settings of your server and CMS. Try to optimize this indicator, as it directly affects the user factors of the site.
11. Availability of an SSL certificate. Officially, the presence of an SSL certificate and support for the https protocol is not an increasing ranking factor in the Yandex search engine. However, SSL can indirectly affect SEO; we discussed this issue in more detail in.
12. Site Security. The presence of viruses, spyware and other software on the site that poses a danger to users and their personal data will negatively affect the ranking of your site. If the problem is not resolved in a timely manner, your site may completely disappear from the search results.
13. Possibility of site indexing. You can artificially prevent site indexing using the robots.txt file or using a meta tag . Prohibiting the indexing of individual technical pages not intended for users is normal practice. However, it often happens that site owners are unaware of this prohibition written on their site. This tag completely prevents the page from being indexed by a search robot.
14. Unique IP address of the site. The coincidence of IP addresses of 2 resources with similar content tells the search engine that these sites are located on the same server and, accordingly, have the same owner. The results should be relevant and varied, so it is not in the interests of the search engine to display multiple sites of the same company in the results. If there are two thematically similar sites on the same IP address, the search engine may, at its discretion, leave in the results pages for overlapping keywords with the greatest relevance. Affiliation of resources by Yandex has not been officially confirmed. In addition, not a single IP address is allocated by the search engine to a limited quota of pages that may be present in the search results. This quota is called the site's crawling budget.
15. Correct operation of DNS. A distributed system for obtaining information about domains must receive correct and uniform information in all regions. We do not recommend changing domain NS records frequently. Otherwise, this may negatively affect the presentation of the site in the search results.
16. Displaying non-existent files and pages. Non-existent files and pages should be returned with a 404 response code. This response code indicates to the search robot that these pages do not exist and do not need to be indexed. If non-existent pages return a 200 (OK) response code, you can not only waste your crawl quota, but also disrupt your site’s ranking in search results by submitting many empty pages without content for indexing.
17. Redirecting users. Sites whose main purpose is to redirect users to other web resources are a type of search spam. Such sites are called Doorways and are excluded from search results.
18. Presence of hidden text. The presence of hidden text or links is regarded by Yandex as a type of search spam. As a rule, the appearance of hidden content is associated with the presence of vulnerabilities on the site that were exploited by attackers. This signal is a serious pessimistic factor, and sites with hidden content are excluded from search results until the violations are eliminated.
19. Availability of clickjacking. Another type of search spam, which is the collection of personal data from users without notifying them. Having detected this vulnerability on a site, the Yandex robot will exclude it from the search results.
20. Simulating user actions. This is a “black” method of website promotion, with the help of which some webmasters try to influence the site’s ranking in search results. Cheating user factors is a serious violation of Yandex optimization principles. This signal threatens your site with long-term exclusion from search results.
21. Favicon File. The presence of a Favicon on the site, optimized for various devices and operating systems, will not only improve user factors, but also increase the CTR of the snippet. Choose an original and memorable Favicon for your project.
22. Optimization for mobile devices. Responsive design has long become a standard in ranking websites in mobile search results. Sites optimized for mobile devices are more likely to be presented to users.
23. Micro markup. Schema.org micro-markup allows the search engine to obtain extended data about a specific page on the site. Micro markup helps to add information about products and their prices, recipes, films, the author of the publication, etc. to the snippet. An extended snippet is highly likely to receive a higher CTR, which is a positive sign in the relevance of the site.
24. Video optimization. Read the requirements for the player and the correct methods of data transfer in the Yandex directory. Correct markup of media content allows the page to participate in the Yandex Video search results and attract additional audiences.
25. Indexing technical pages. A large number of technical pages open for indexing that do not provide any value to users is a negative signal in the ranking of the site. Try to block all technical pages from indexing in the robots.txt file.
26. Server location. The location of your server affects not only its response speed, but also regional ranking factors. A site located in the Krasnodar north will participate with high priority in regional search results, including geo-dependent queries.

Content and text ranking factors

In the world of Internet marketing, there is an expression “Content is King”, which means “Content is the king”. This phrase reflects the power of content not only on users, but also on search engines, which rank sites largely depending on the quality of the content on the page.

1. Page title Title. The page title has a limited number of characters in the snippet. Try to make titile no more than 50-60 characters. The exact occurrence of a keyword increases the relevance of the document in the search results.
2. Heading H1. is an important element that reflects the entire content of the web page. The length of H1 should not exceed 60 characters. Having the main keyword in the title will increase the relevance of the document.
3. Meta Description. is the most important element involved in the formation of the search snippet of a document. The basic snippet will include a description of 230 characters. Using an exact occurrence of a keyword increases the relevance of the document in the search results.
4. Meta Keywords. This tag can be taken into account when determining the relevance of a page to search queries. Yandex, unlike Google, continues to use the information contained in this tag, so it is better to fill it out. Add the main keywords relevant to your document into meta keywords.
5. Text relevance. Matching the main keyword to the rest of the page text increases the page's relevance. The word most often used on the page should be combined with the theme of the rest of the text. In an article about women's swimsuits, the word "swimsuit" will most likely appear most often. The text on the page must be relevant to the given key.
6. Keyword citation frequency. The most important keywords tend to appear most often in the text of the page. This is normal practice and allows the robot to determine the most relevant queries for a given page. The frequency of quoting keywords should be natural; text over-optimization threatens your site with sanctions from Yandex. Baden-Baden is a pessimizing algorithm that has been excluding pages from search results for text spam since 2017.
7. Phrase priority. The presence of a keyword in the text of the page that exactly matches the search query increases the relevance of the page.
8. Text priority. Having separate parts of the keyword in different places on the page that match the keyword increases the relevance of the page to a lesser extent compared to the previous factor.
9. The presence of a request in the first sentence of the page. The coincidence of a query and a keyword in the first sentence of an article increases the likelihood of it being included in the snippet, as well as increasing the relevance of the document as a whole. Works similarly to text and phrase priority.
10. Availability of key synonyms. The use of synonyms, topical words and terms in the text of the pages increases its relevance to the key queries of this topic.
11. Availability of thematic synonyms in Description. The presence of thematic synonyms in the meta description increases the relevance of snippets and pages in general.
12. Availability of thematic synonyms in Titile. The presence of thematic synonyms in the Title increases the relevance of snippets and pages in general.
13. Presence of thematic synonyms in H1. The presence of topic synonyms in H1 increases the relevance of the document to a lesser extent than the use of direct occurrence.
14. Text quality. Spelling and punctuation errors in the text reduce the relevance of documents to key queries. Punctuation errors can negatively affect behavioral factors and ease of reading, spelling errors can disrupt the semantics of keywords.
15. Text complexity. The complexity of the text, both structurally and semantically, reduces the relevance of the page. The text should be simple, understandable and useful. Try not to create complex compound sentences with a lot of phrases and complex terms. Feel sorry for your audience, they will reciprocate you through user factors.
16. Request Segments. The presence of diluted keywords in the text, which themselves have a frequency, increases the relevance of the document. These are so-called subqueries, for which your page can also be ranked in search.
17. Morphological proximity of the text. You should not try to fit the text into the specific technical framework of search engines. The text must be interconnected both structurally and morphologically. The most frequently occurring keywords on a page should not be concentrated in one piece of text. Also, there should not be a large accumulation of keywords that are not related to the topic of this page in other pieces of text.
18. Text structure. To semantically and logically structure the text, use headings H1...H6, tables, bulleted and numbered lists, quotes and other HTML elements. Convenience and clarity of text have a positive effect on user factors.
19. Amount of content on the page. Depending on the topic, the amount of content on a page can play an important role in determining the relevance of a document. This usually applies to information sites that specialize in content production. Complete presentation of information and answers to all hypothetical questions increases the relevance of your page.
20. Unique content. It's no secret that search engines prefer new and unique content, and have a negative attitude towards search spam, reprints and rewriting. Your website should be the primary source of useful information. Content that is truly valuable to users easily receives natural links, social signals and good behavioral factors.
21. Automatic content generation. Any content automation will negatively affect the ranking of your site. Synonymizers, translators, text generators are only application tools for solving small local problems, but not content engines. Automation is good, for example, in generating unique meta tags for a large number of thematically similar products, but not in producing content.
22. Alt and image titles. SEO optimization of images will allow you not only to attract additional traffic from Yandex.Images, but will also increase the relevance of the document, as well as the convenience for users if your page does not load completely.
23. Duplicate text. Duplicating the same text both within one page and within the entire site as a whole will negatively affect the ranking of your resource.
24. Content update frequency. A high frequency of content publication while maintaining high levels of user factors tells search engines about the authority and usefulness of the resource, and is also a good signal in determining the relevance of the site.
25. Text optimization. It’s not enough to publish new content; you need to keep previously published pages up to date. This applies to both information sites and e-commerce. The relevance of content with outdated information decreases, as does, for example, a category with products that are not on stock.
26. Multimedia and files. Correct markup and use of media files can be a positive signal in the ranking of a resource. Their absence, especially in those niches where this is the standard, is a negative signal. The main guidelines that you should rely on are convenience and the question of whether this content is necessary for users.
27. Nesting. The further the content is from the main page, the less important it is considered. Level 1 nesting is the /blog/ section, level 2 nesting is /yandex-ranking-factors/. We recommend that the nesting of important pages does not exceed level 3.
28. URL Length. Trying to use as many keywords as possible in a url is a bad practice. In general, search engines recommend creating convenient and simple URLs that reflect the essence of the page, but do not give preference to them when ranking. This factor may indirectly affect user factors and repeat visits to the site.
29. Presence of a request in the page URL. The presence of a key query in the URL is a positive ranking signal that increases the relevance of the document. Human URLs are the most common address format that users are accustomed to. Agree, the URL /yandex-ranking-factors/ will tell the user much more about the page even before visiting it than, for example, /85437234/.
30. Validity of HTML markup. Valid HTML code without syntax errors is a positive ranking signal. Serious markup violations can negatively impact your site's ranking. Even today, in the vastness of the RuNet, you can find pages whose entire content is wrapped in a tag

    . You can check the validity of your markup using this service.

31. Language Compliance. Check the lang attribute in the opening tag . The language of the content posted on your site must correspond to it. In addition, an article in English optimized for queries in Russian is unlikely to get a place in the search results. However, it is quite acceptable to use the Latin alphabet to denote terms, attributes, abbreviations, etc.
32. Availability of comments. The presence of comments helps to increase the relevance of the document for low-frequency queries, increase audience activity on the page, as well as user factors. As a rule, this factor seriously affects the relevance of information sites.
33. Duplicate pages. Duplicate pages are a bad signal for website ranking. The need to duplicate documents with very similar content often arises in e-commerce, so to correctly determine the preferred page for indexing, use the rel="canonical" attribute.
34. H1 duplication. Duplicate first-level headings can make it difficult for a search engine to determine the relevance of a page for certain search queries. H1s should be unique throughout the site, and only one page on your site can answer specific user requests.
35. Duplicate Titile. The same situation occurs with duplicating the page title. Copying the title will make it difficult for the search robot to determine the relevant page. Titiles must be unique within the entire site.
36. Duplication Description. The situation is similar with the meta description of the site. Complete duplication of meta description on different pages will make it difficult for the search robot to determine the relevant page.
37. Duplication of other content. Duplicate videos, images, audio files and other content are also negative ranking signals. Don't try to use as much irrelevant content as possible just to increase page size.
38. Document age. Page age can be either a positive or negative signal in document ranking. In this matter, everything depends on the topic. If this is news, then obviously, more recent documents will have priority in ranking, but if it is a service, for example, for generating passwords, which has managed to earn high user factors, then its age will only be a plus.
39. Brand mention. The citation of a site, company or brand is not necessarily a link. Text citations and brand mentions on authoritative resources will become a positive factor in ranking.
40. Cloaking. A type of search spam that unscrupulous webmasters use to influence rankings. This is the display of different content for robots and site visitors. Often the result of a site vulnerability and the presence of malicious code. The presence of different content for users and search robots negatively affects the ranking of the site.

Link Ranking Factors

Link ranking factors today have a much lower priority in determining the relevance of documents than they did 10 years ago. In addition, the algorithm for determining natural and commercial links has also changed. Attempts to influence search results by purchasing external links threaten your site with the Minusinsk - pessimizing Yandex algorithm. Many webmasters note that the greatest effect of links is visible in the first months of resource development. Thus, the search engine receives signals about the demand for the site by users, which allows for faster indexing of pages. The principle “the more external links the better” has long since become obsolete. When ranking, only natural links are beneficial, which can be obtained by creating useful content, selling quality products or providing quality services. Getting links for your resource should ideally be a natural process that does not require your intervention. Below we will look at the factors that can be used in the search algorithm for ranking and determining natural links.

1. External links/site age. The ratio of the number of links to a site and its age can tell Yandex about the naturalness of the process of receiving them. Too large an increase in the link mass of a young resource may indicate “cheating”. The continuously decreasing dynamics of external links for an authoritative resource may indicate a decrease in its popularity.
2. Reference diversity. The variety of links to your resource shows their naturalness. If 100 sites link to your resource, all the links are text, it’s unlikely that you earned them naturally. The natural link mass is as diverse as possible and includes anchor and non-anchor links, links from social networks, links from images and media files, etc.
3. Exact link anchor. The exact keyword shows the relevance of the resource on the other side of the link.
4. Converted anchor. Keyword synonyms or topic words can show the relevance of a page. An accurate link anchor is of great value
5. Link relevance. The anchor or link image should be relevant to the site it links to. Conventionally, the anchor link “Yandex ranking factors” should not lead to a page about image optimization.
6. Title attribute of a link. It is an additional signal of the relevance of the link. The titile attribute displays the title of the link when the cursor hovers over it, and thereby provides additional information that gives the user an understanding of whether he should follow the link or not.
7. Donor age. If older domains that have long been involved in Yandex rankings link to your site, this is a positive signal in determining the relevance of the site.
8. Donor quality. It’s not just age that determines the quality of a website. In RuNet you can find many age-appropriate, but absolutely useless resources. A link from such a resource is a bad ranking signal.
9. Donor IP addresses. If all sites linking to your resource have the same IP address, this is a negative signal. Most likely, all external links belong to one owner, who artificially tries to influence the ranking of the site.
10. Regionality of donors. The regionality of donors corresponding to your site can be a positive signal in the ranking of the site. If your site, specializing in working with audiences in Moscow, is mainly linked to Ukrainian sites, this is an indicator of the unnaturalness of your link profile.
11. Links from sites under sanctions. If you are being linked to by domains that have been penalized for search spam or security violations, this will become a negative factor in your site's ranking.
12. Donor theme. Matching the theme of your site with the theme of the site linking to you is a positive signal. If sites about cryptocurrencies link to your dental website, this will show that the link mass is unnatural.
13. Dynamics of growth/decrease of links. The dynamics of growth or decline of links should also be natural. A sharp increase or decrease in the link mass will indicate its unnaturalness.
14. Incoming links from home pages. Separate positive ranking signal. If authoritative sites link to your resource from the main pages, this indicates a high level of trust or even partnership with you.
15. Links from significant sites. A positive ranking signal that encourages sites that are linked to by particularly significant and authoritative resources, such as Wikipedia.
16. Link age. The age of external links to your resource is an indicator of the authority and relevance of your site. This is a positive ranking signal.
17. Bookmarks in browsers. Search engines collect all available information on resources, including from browsers. Having your site bookmarked or "favorited" by your users can be a positive ranking signal.
18. Mentioning a brand or domain in a request. As a rule, text links with a brand or domain name in the anchor are more natural in nature.
19. Percentage of broken external URLs. A high proportion of inbound links that point to your 404 pages is a negative ranking factor. However, their complete absence calls into question the naturalness of your link mass.
20. Naturalness of links. To determine the naturalness of links, Yandex has a separate algorithm consisting of many factors, the details of the formula of which the company does not disclose. When determining the usefulness of a link, the factor of its naturalness takes precedence.
21. Outbound links to authoritative sites. Outgoing links from your resource to other authoritative sites can be a positive signal, provided that these links are used by your visitors.
22. Outgoing links to low-quality sites on the contrary, they can become a negative ranking factor.
23. Number of outgoing links. A large number of outgoing links from your resource is a negative ranking factor. The purpose of the resource should be to provide useful content, and not to redirect users to other sites. However, this factor does not apply to all niches, for example, product aggregators.
24. Number of internal links. So-called linking can significantly improve the behavioral factors of your site: time spent on the site and browsing depth.
25. Quality of internal links. The main rule for using internal links is that they are natural and useful for users. There is no need to put links for the sake of links.
26. Broken links. A large number of internal links leading to non-existent pages can negatively affect the ranking of your site.
27. Near-link content. The morphological and semantic difference between the anchor text and the adjacent text can serve as a negative ranking signal.
28. % of hyperlinks relative to text volume.
29. Selling links on exchanges. This usually applies to information sites. Open sale of links from your own resource on link exchanges can become a negative ranking factor.
30. Buying links on exchanges. Data on sites that host links for payment are available to any user registered on the exchange. Most likely, search engines use this information in determining the naturalness of links.

Search Ranking Factors

Search ranking factors are those factors that Yandex receives directly from search and search services.

1. Availability of Yandex.Metrica. It is not a ranking factor as such, however, after replacing the TIC with the IKS, the data from Metrica will be taken into account by the algorithm to determine the site quality index. The formula for calculating the ICS according to Yandex data is similar to the formula for ranking sites in general. Therefore, there is every reason to believe that data from Metrica can both negatively and positively affect the ranking of a site if there is a counter on the site.
2. Availability of YAN advertising. This factor applies to sites that place advertising on their pages. The presence of advertising from Yandex, and not from competitors, can have a positive effect on the ranking of the site.
3. Snippet CTR. The clickability of a snippet in search results indicates not only the relevance of the title and description, but also the relevance of the page to search queries in general. The snippet can be generated dynamically based on the page content.
4. CTR of quick links in the snippet can indicate their relevance and be taken into account as a behavioral factor.
5. Share of direct visits. Even if there is no Metrica counter on the site, Yandex receives information from other services and search. Thus, a high proportion of direct visits to a site can tell a search engine about the high quality of the resource.
6. Share of transitions from other PSs. A high percentage of referrals from other search engines can be a reason to increase the site's ranking.
7. Regional attendance. Information about visitors from specific regions can serve as a positive signal for a site to rank higher in that region.
8. Repeated site visits. A high percentage of repeat visits to a site can serve as a positive ranking signal.
9. Last URL. The number of sessions in which your URL was the last one when searching for a relevant answer to a search query is taken into account.
10. URL impressions after which users went to look for an answer in other PSs. Impressions and clicks on your and other sites after which the user used another search engine are taken into account. This signal may indicate to Yandex that search results on this topic are of poor quality. Assessor checks are possible.
11. Return to search. Bouncing back to search after visiting your site can be a negative ranking signal. If a user doesn't find the answer they need on your site that was relevant to their query, it's likely your site doesn't provide that answer and therefore should be demoted in the search results.
12. The ratio of the number of clicks on a specific domain relative to the search results. The number of clicks on a specific domain in the search results is taken into account, which does not always correlate with its position. The relative number of site clicks on different search results may be greater than the same value for other sites.
13. User data. Data about users and their quality is taken into account: interests, browsing history, data from Yandex services. Most likely, data on users who have access to Yandex services (site owners, optimizers and employees) is excluded.
14. Yandex Bar. This is additional software for browsers, which, like other services, can transmit metrics present in the website ranking formula.
15. Chat with a company in search. In August 2018, Yandex implemented the ability to chat with a company directly in search results. Data on the speed of response to messages, as well as the quality of responses, has become a new factor in ranking sites.
16. Increase in traffic while maintaining positions. An increase in internal transitions and transitions from other sites can be a positive signal for the search robot in ranking your site.
17. Search errors and typos. Yandex takes into account data on search sessions with errors and typos. If the site that the user is looking for was not presented in the search results, further recording of his visits occurs with a search for a relevant resource.
18. Number of pages in search. The more quality pages your site has in the index, the more authoritative the site is in the eyes of the search engine.

Behavioral Ranking Factors

Behavioral factors are one of the most influential factors that determine the relevance of a document in search results.

1. Aggressive advertising. The presence of aggressive advertising on the site causes rejection among users. As a rule, this factor negatively affects the length of the session and the depth of viewing.
2. Availability of erotic advertising. The presence of advertising of this nature on the site negatively affects the ranking of the site.
3. Site structure and navigation. The site should have a simple and understandable structure. The user should ideally reach each page of your site in no more than 3 clicks. A convenient website structure has a positive effect on website ranking.
4. Emoji in the snippet. With HTML5 support, it has become possible to add special characters to titile and description, which Yandex displays in the snippet. On the one hand, special characters can disrupt the semantic significance of the title and description, on the other hand, they can seriously increase the site’s CTR in search results. You must decide for yourself whether to use this opportunity or not.
5. Bounce Rate. At the moment, the metric considers sessions in which the user simultaneously spent no more than 15 seconds on the site and opened no more than 1 page as failures. A high bounce rate on your resource may indicate that the information on it is irrelevant.
6. CTR of internal links. Click-through rate shows the behavioral significance of internal sections and pages.
7. bread crumbs. A navigational must have that has become a standard for SEO optimization. Participate in the formation of quick links and display URL in the snippet. User significance lies in the fact that people, being on any page of the site, understand their location in the site structure.
8. Average continuous visit. This indicator is taken into account in the site ranking formula. The optimal amount of time a user should spend on a website depends heavily on the topic. A short session time is not always bad.
9. View depth. The indicator reflects the user's interest in the content on the site. High browsing depth is a positive ranking signal.
10. Long text. As a rule, the better you answer user questions, the more likely you are to rank high in search results. This factor works provided that all the content on the page is of high quality, which the user will be willing to read from beginning to end.
11. Advertising on every click. Websites full of advertising are ranked worse by search engines. If every time a user clicks on your site, they see an ad, this is a negative ranking factor.
12. Popunder. The presence of aggressive pop-ups is a negative ranking factor.
13. Internal transitions. Yandex takes into account the number of sessions with a timeout of 30 minutes, these are internal transitions.
14. Ratio of advertising and content. If the amount of advertising on the pages of your site exceeds the content, this will be a negative ranking factor. Remember that your website visitors are primarily interested in useful content.
15. Availability of product filters. The presence of all kinds of navigation tools, filters and sortings that make it easier for the user to find the necessary content increases behavioral factors.

Social Ranking Factors

With the development of the Internet, social networks have become widespread: Vkontakte, Facebook, Instagram, Google Plus, Odnoklassniki, Twitter, etc. Your business's presence and activity on social media is an important ranking signal for search engines.

1. Website profiles on social networks. The number of audiences and the quality of profiles are taken into account. A wide presence on social networks provided by the target audience is a positive ranking signal.
2. Number of transitions from social networks. The number of transitions and their quality, expressed in behavioral factors, is one of the social ranking signals.
3. Audience activity in profiles. The activity of the audience in the groups themselves is taken into account: likes, reposts, comments on posts, discussions and other sections.
4. Number of page likes. Social widgets on a website with good like rates will be a positive ranking signal.
5. Number of page reposts. Shares on your social media pages are a positive ranking factor that shows your audience's interest in your content.
6. Number of links from social networks. External links from social networks, their volume and quality can signal to search engines that the audience is of great interest in your project.
7. Authority of Referring Profiles. A positive signal will be presence in your groups, or mention of your project by authoritative profiles confirmed by social networks.
8. Relevance of the referring profile. Thematic similarity of linking profiles to your site is taken into account. A high degree of similarity will be a positive ranking signal.
9. Number of tweets. Typically, tweets and retweets help speed up page indexing.
10. The authority of retweet authors. Retweets from authoritative profiles can serve as a positive ranking signal for a website.

Regional ranking factors

1. Regional affiliation. Assign a region to your site in the Webmasters panel to determine the relevant audience.
2. Geographical proximity. The geographic proximity of your site can serve as a positive signal when generating search results.
3. Geo-dependent words in H1. The presence of geo-related words in the first-level heading increases the relevance of the document for geo-dependent queries.
4. Geo-dependent words in Titile. The presence of geo-related words in the title increases the relevance of the document for geo-related queries.
5. Geo-dependent words in Description. The presence of geo-dependent words in the description increases the relevance of the document for geo-dependent queries.
6. Geo-dependent words in page text. The presence of geo-sensitive words in the page text increases the relevance of the document for geo-dependent queries.
7. Regional subdomains. Using regional subdomains especially for the .com domain zone increases the relevance of your site to the regions covered.

Commercial Ranking Factors

Commercial factors generally do not influence rankings directly, but do so by improving user factors and increasing business credibility.

1. Actual address. The presence of the actual address of the company in the contact information shows the real existence of the business.
2. Convenient product navigation. Convenient display of product grids and uniform presentation of products makes the online store more convenient and easier to understand.
3. Company details. The presence of company details shows legal transparency and increases user confidence.
4. Landline number or 800 number. The presence of not only a mobile number, but also a landline number is one of the commercial ranking factors. For all-Russian online stores, the standard is number 800.
5. Company work schedule. A company's operating hours are a positive commercial signal that helps users understand when their order can be processed.
6. Messengers. Every person has at least one messenger. And your business supporting all possible communication channels is a positive signal.
7. Email. Domain-based email is a mandatory attribute of any serious online store.
8. Feedback form. Give users the opportunity to leave their feedback or report a problem directly on the site.
9. Online chat. Online chat is now displayed in search. Fast speed and high quality of answers are a positive ranking factor. Don’t forget to do the same in the chat on the website.
10. Availability of links to social media. networks. Don't forget to add links to your company's social accounts to confirm your connection to them.
11. Availability of maps and directions in contacts. Yandex Map and detailed directions to your offline business will become a positive ranking signal.
12. Yandex directory. If in addition to your online business you have offline points, be sure to register your company in Yandex.Directory. This will allow you to attract additional audience to the site and expand the search snippet.

We've released a new book, Social Media Content Marketing: How to Get Inside Your Followers' Heads and Make Them Fall in Love with Your Brand.

The algorithms of search engines are similar to a dense forest, in which there is one, but well-lit path. And you can see the road, and you know where it leads, but you never know whether you have turned right, who is looking at you from the thicket, and whether some Minusinsk will grab you from the bushes. Therefore, today we are talking about Yandex: we figure out what is known about it and find out how it works. If you don’t know how the most popular search engine in Russia works, read on.

What is known about her?

Nobody knows the exact principle of Yandex’s commercial ranking, not even the company’s employees themselves. This is not because it is stored in a secret bunker with a guard regiment, but because people have not been developing it for a long time - the machine does it for them. But this does not mean that absolutely nothing is known about the algorithm. Let's start with the basics.

What is ranking and how does the Yandex formula work?

– this is the ordering of pages by the greatest correspondence to the search query. The search engine analyzes all pages with suitable content and ranks them in descending order of usefulness for the user. The coolest, most informative, high-quality and popular sites occupy the first lines, and those that are a little worse fall lower.

The Yandex ranking formula is an algorithm that determines the order of pages in search results. It works in semi-automatic mode. Special Yandex employees (assessors) evaluate the robot’s work: if the results for some random request are compiled correctly, the robot receives a well-deserved “A” and asks mom for pocket money. If not very good, he works on the mistakes and tries again.

The algorithm determines the order of issuance using factors - manually specified criteria for assessing the quality of the page. There are about 800 of them in the Yandex search engine ranking formula, and no one except the Yandex employees themselves have seen the complete list. SEO specialists know a lot, but not everything. Most of the minor factors are simply impossible to deduce, because no one can get into the heads of Yandexoid employees.

When the assessor gives a rating to the results generated by the robot, the latter tries to understand why the rating is the way it is. For example, the algorithm did not take into account that some site simply had a terribly inconvenient structure and an eye-catching design, but, all other things being equal, it ranked it above its competitors. After several repetitions, the robot will understand the mistake and associate the low score with this factor.

The Yandex ranking formula is constantly changing: it recalculates the “weight” of factors and the value of existing pages based on new information about user behavior. This is necessary to compensate for the constant changes occurring in the network. For example, in a year, people may fall in love with sites with built-in chats, and the search engine will begin to rank them higher. Not without the help of assessors.

The Yandex ranking formula is huge - it is a huge array of data and algorithms that a living person would not be able to write. It takes into account the user’s region, the current situation with the preferences of visitors, the quality of resources, their authority, usability and who knows what else. But no one despairs, because Yandex itself gives some clues about the principle of its operation. For example, with commercial sites everything is quite transparent. Let's talk about the principle by which they are ranked.

Yandex commercial ranking formula

In 2013, Yandex introduced a new concept - commercial. It differs from the “usual” thematic one in that the usefulness of the resource for a potential buyer is assessed first. In general, the situation with ranking selling sites is standard - the most useful, convenient and relevant pages get to the top. The differences are in key factors. Attention is drawn to:

• User trust.
• Ease of use.
• Quality of the resource.
• Service.

It is clear that a machine does not always correctly assess the quality of a site, because it is not a person. From a technical point of view, everything may be logical and thought out, but in reality it will be difficult to use the site. To help the algorithm figure it out,we need a human perspective. The assessors mentioned above are responsible for it. They help the system learn to distinguish bad sites from good ones on its own.

The work of assessors is divided into two stages: assessing thematic and commercial relevance. The first stage is quite simple: a Yandex employee assigns a specific rating to the page. Resources are:

• Irrelevant.
• Relevant.
• Highly relevant.
• Useful.
• Vital.

Then commercial relevance is assessed. Vital and useful sites are excluded from the sample because they are already doing well. Assessors look at:

• The number and variety of goods and services for a specific search query. The range can be small, standard or large (this is the official rating scale). The criteria by which a site is assigned one of them is shrouded in mystery.
• Reliability and quality of service. There are four possible ratings: spam, standard, good and excellent. “Standard” resources maintain the average temperature in the hospital, “good” ones offer slightly better quality, and “excellent” ones stand out from both. Everything is clear with spam pages - these are fake pages on which you can’t buy anything.
• Usability and design. There are three possible ratings: bad, good and excellent.

Ultimately, all the assessors’ assessments are transformed into improvements to the Yandex commercial ranking formula, and it begins to see all the mistakes itself. Therefore, even if you are at the top now and think that everything is in order, it is not a fact that this will be the case in six months.

In general, the commercial relevance formula looks like this:

Let's make it clearer:

Commercial relevance= (Variety and quantity of products) x (2x(reliability) + (design) + (usability) + 2x(service quality).

According to Yandex, a good commercial website offers a lot of products, looks good, is reliable and easy to use. At the same time, the Yandex commercial ranking formula is very sensitive to the reliability of the site and the quality of service - they are multiplied by two. This is logical, because it is in the interests of the search engine to answer user queries as accurately as possible.

Commercial factors in the Yandex ranking formula

The overall relevance of a resource consists of commercial and. With the latter, everything is clear - if the content matches the request, then everything is fine. The first one is more difficult - the page needs to be designed according to the quality criteria of Yandex itself. This:

• Contact details. Everything should be as complete as possible: telephone numbers, e-mail, Skype, address, location map, “how to get there”, work schedule. If there are several telephone numbers, please indicate all of them. If you have several branches or departments, register their contacts separately.
• Links to pages on social networks.They must be alive and without cheating. There are also requirements for regular updates. A link to the VKontakte group, where you post once a year, will not give anything.

• No advertising. Everything is simple here - if you sell something yourself and at the same time advertise other commercial sites, this is a minus. Same thing with custom pop-up banners. Yandex especially doesn’t like popunder banners.

• Possibility and convenience of goods delivery (for stores).Delivery information should always be visible. It would be a plus if you duplicate it in product cards. Be sure to indicate all methods, terms, costs and regions of delivery, and addresses of pickup points.

• Payment Methods. The more and more varied, the better.

• Detailed product cards. A good card includes informative descriptions, comparison options, “similar” blocks, video reviews, reviews, ratings, availability information, related products, and so on.

• Working customer support service. These could be call center phone numbers, “Help” sections, FAQ, or a call back request form.

• Connected online consultant. It's simple - it must exist and it must work.

• Discounts.Advantageous offers are ranked higher, so if you make a discount, indicate its size and expiration date.

• A simple and memorable domain name. The easier it is to remember and the better it relates to your topic, the better.

• Short and logical URLs. The user must understand in which section of the site he finds himself.

• Optimal average title tag length. The snippet only displays 70 characters. You need to pack everything you need in them and not spam it with keys.

• Compliance of the page content with the stated title. Everything is simple here - if the title says that there are prices or videos on the page, then they should be on the page too.

• Average page nesting. There should be no unnecessary categories and sections.

The Yandex commercial ranking formula has already learned to distinguish technically good sites from technically bad ones. The algorithm is able to analyze the usability of a site almost from a human point of view. All that remains is to teach her to think just like a human, so that she can take over the whole world. Yandex has been striving for this goal for a long time and is making amazing progress. Maybe in a couple of years we will stop running into sites that look like 2002 never ended and offer us the same quality content.

This article was originally written for a competition from Seopult, but for some reason they never responded to my letters and completely ignored me. They have a very strange competition. But it’s okay, there will be more interesting posts on my blog.

Finding a Relevant Page on a Website

Relevance– this is the maximum correspondence of the search query to the search results. That is, our main task is to optimize the promoted page as close as possible to the request, so that Search Engines accept it as the most relevant.

For Yandex

The first step is to determine which pages on your website are more relevant to the desired query (keyword) in Search Engines. For Yandex you can define it like this:

We go to the Advanced Search and look for the desired query on our website, and the first page in the search results will be the most relevant for Yandex, this is the page that needs to be promoted and optimized.

For Google

In the search bar we enter:

Keyword site:your_site.ru

The first page in the results will be the most relevant page for Google on your site. Google also has an Advanced Search, but this is faster.

As you can see, the relevant page is different for both PS; it is better to choose for promotion the page that is more relevant for the PS you need. For example, I always choose Yandex, because over time, with optimization and promotion, Google quickly and more readily changes the landing page to the one we need.

Landing Page Definition Services

If you need to find many landing pages for a large number of queries, then it is better to use special services, for example, you can do this using SeoPult, just add the site to the system and all the keywords. It looks like this:

Once you have landing pages for promotion, you need to optimize them and make them relevant to the selected keywords.

Factors Affecting Page Relevance

Domestic

1) Keyword in tag . There are a lot of disputes about how much text you can write in a title. There is no definite answer, I stick to the rule - no more than 65 characters, and you are on your own
define.