Archive for the ‘ Search Engine Strategies ’ Category

Principles For SERP

Today’s publication of the start cycle of theoretical articles devoted to the calculation of the relevance of documents resource sites Internet search engines, today talk about the general principles for ranking search engines, as well as a classification ranking factors, give a general formula for calculating the value of relevance.

?????????????

FP ranking algorithms are not secret information. In addition, the network is periodically publication of certain features of TS algorithms. A typical example of this – the annual Russia Seminar on Information Retrieval Evaluation (ROMIP). This seminar – an initiative to establish a platform for independent evaluation of methods of information retrieval, focused on working with the Russian-language information.

Factors influencing the rankings are divided into static and dynamic.

Static factors do not depend on the request to the FS, such as the credibility of the page, which is also called PageRank. (not to be confused with toolbar PR Google).

Dynamic factors depend on the query text and are divided into internal (organization of the document) and external (reference ranking).

Factors ranking search engine

  • Static ranking factors – the credibility of page
    • weighted index
    • PageRank – credibility Pages Google (not to be confused with toolbar PageRank)
  • Dynamic ranking factors
    • External factors – reference ranging
    • Internal factors – the internal optimization
  • Proper ranking factors – your own catalogs of resources
    • Yahoo-catalog http://yahoo.com
    • Catalog Google http://www.google.com/dirhp

The number of ranking factors and their relevance in different alternatives and constantly changing FS over vremeni.Eto linked to the ongoing development of themselves and the development of SAR retrieval algorithms. Conditional formula for calculating the relevancy can be represented as follows.

, (1) , (1)

where Total value relevance of document ??????? request ; ;

The value relevance of the document package - Quality ; internal optimization;

The relevance of the text links from other documents in the document ; The figure for brevity may be called referential relevance.

Figure authoritativeness page ; ;

Some nondecreasing function; allow simplification that ; ;

- Some of the factors that allow approximate formula relevance under each FP.

However, the formula (1) does not take into account its own ranking factors of PS, which include their own resource directories of search engines. The conditional formula to calculate the relevance with their own ranking factors substation can be represented as follows.

, (2) , (2)

where , , , , Some approximate coefficients for each of their FS;

- Relevance own ranking factors FP.

In this article, I described the ranking factors FP, let their classification, let a general formula for calculating the value of relevance.

Today the guest post by Alexei, who has a blog on the promotion of a website, blogging, earning online. Alex will describe how to move your site from one domain to another if the need arises, and not lose our position in search engines, visitors and PR Google.

How and why people change domains, the question of a separate article. It sometimes happens that the old domain “morally” outdated and there is no sense to develop a project within that domain. But save the content, external links, and visitors, of course, want. The following FAQ how to do it correctly, so as not to lose positions and visitors.

The problem is simple – to organize a redirect from all pages of the old domain completely to the new domain.
For example: with all the pages of the old domain www.example-old-site.com make a redirect to the domain www.example-new-site.com, and its pages.

1. Create a sitemap.xml for the old domain.
2. Create content (contacts, description of the company, future plans) and a decent link to the new domain. (On the new domain, you must begin to create links to earlier).
3. Run the site but the new domain.
4. Register and check the old and new domains using Google Webmaster Tools & Google Analyticals
5. Create an arbitrary page 404 in the old domain, which will prompt the user to visit the new domain.
6. Test the redirects from the old to the new domain. if this is the so-called redirect 1:1. (www.example-old-site.com/category/book-mustaches.html will lead to www.example-new-site.com/category/book-mustaches.html)
7. Set 301 redirect from the old domain to new domain.
8. Get the old sitemap.xml to Google and Bing (Bing Webmaster Center) (This step will allow search engines to download the old URL, see that it is 301 redirects and update their mass index)
9. Tell Google about what you changed the address for your site with services Change of Address.
10. Create a new sitemap.xml and upload to search engines (it will tell them about the new URL is not present in the old domain).
11. Wait until Google Webmaster Tools  update its data, detect errors and inform about them.
12. To closely monitor the positions of the new domain to make sure that it is well indexed.

For all these actions with the domain you need full access rights to it, to set 301 redirect.

Characteristic and striking example of moving from one domain to another, which I saw here: how to properly move domains

By 1970 the annual budget of the porn industry has reached 10 million. Today porn drawn to the billion mark. According to the company Alexa, from thousands of sites with the highest total traffic, four belong to the category of porn sites. Google serves 100 million monthly queries related to pornography, and this figure is constantly growing. All this confirms the old truth, it is applicable to the network: sex and everything associated with it, very well sold.

Among the countless porn stars and producers of porn you will find thousands of links to TechCrunch.com – quite a respectable news site technical guidance.

site: www.techcrunch.com intext: porn

Only in the last year at TechCrunch has added more than 550 posts that contain the word “porn” in his body. In some cases, the articles are devoted entirely topical and sensational items, such as for example talk about porn annex to the iPhone, but often they appear to be established only with great reserve. In general, TechCrunch regularly writes about porn, using any convenient excuse.

Although these articles rarely making it into the top, they are often caught a 1-3 page issue on a number of high-porn requests, including “youporn”, “you porn” and “iphone porn”.

How it affects the rate of traffic TechCrunch? With the monthly statistics for requests and knowing the average data on click-rank issue, we can easily calculate that “vzoslye” queries bring TechCrunch about a quarter million visitors a month.

TechCrunch optimizes your porn posts not only by search engines, but also by users. Often in the introductory part of the post is added quite candid image of the relevant topics. In addition to reducing your bounce rate (the number of new visitors who have come through the “adult” search request and immediately retired from the site), these images also increase clickthrough material when viewed through the news aggregators such as Techmeme. Finally, these images, like images in all other positions, do not contain a hyperlink that works to increase the number of clicks on a column with advertising.

Revenues from advertising – the dry residue from porn-blogging

1) New visitors. Doputim, everyone looks at average 3 pages. Consequently, the 286 625 hits mean almost 900 thousand extra hits every month. Let us go further: suppose, TechCrunch takes the $ 2-per-click (rough estimate of the value of advertising on the site). This means that in his search counter behalf of Ron Jeremy (a popular American porn actor – approx. Ed.) TechCrunch rescued tens of thousands of dollars every month.

Optimistic estimate

Conservative estimate

2) Repeat visitors (subscribers). If TechCrunch engaged in sales of any product, porn traffic it would be almost nothing. But he does not do so. He receives money from advertising. What he has more hits, the more he received impressions and clicks. Therefore, unlike many other websites and blogs, for which the porn traffic is useless because of the inability to deliver new customers, TechCrunc, serving an audience of technically savvy geeks, get a good idea to cash in on the secret aspirations of its readers, who came here as a “not for this .

In defense of TechCrunch can say that this trick does not use it alone. Wired also receives its revenue from porn-blogging, but this couple, apparently a pair of white crows among many other technically-oriented news sites. If you look at the statistics on percentage of pages that contain the word “porn”, all pages within the site, available in the index of Google, becomes immediately obvious: in comparison with its competitors, TechCrunch and Wired have a significantly higher proportion of content related to porn.

Do not miss your chance if you get rich? I will increase your conversion rate, if your posts will rank well on “adult” needs? Can not open it here in front of you truly exciting prospects? In the end, always makes sense to publish a post about another condom, posing as a pro web marketing, or about how you fucked with one of the services to collect data anal-Check Tool List

It turns out that Google made a new (less than a year :) ) called SBKTool (Keyword Based Search Tool) in addition to the existing Adwords Keyword Tool.

SBKTool

You can drive through the window of any site and see if he knows Google suited Keyword. Unclear whether all Keyword are “organic” to the desired site, but for some site is accurate. Sam set key also not quite complete, but given that this information about someone else’s site, even in such a tool becomes quite useful.

Besides, you can go to the section of Keyword and see what Google considers the most fat.

Google announced updates to Blog Search Google Blog Search. I myself often use this service when you need to find information on the same topic, especially reviews. Blog Search is important, because new interesting blogs there so much that it is impossible to keep track of all, and even more so to read them in the reader. Blog Search also allows you to filter out of a total flow of only what you need – and most importantly, that this filtered stream of information you can subscribe via RSS.

Subscribe to RSS

blogsearch_google_com1

It can be both at the TOP news (those that are on the main page) and to any request / heading. I have already signed up for a few – look at how tightly to rain and will be updated if there is something interesting. Note that on some queries that I have signed quite a few years, so for those on-demand pulls positions, this feature is no longer news.

Top queries

blogsearch_google_com2

“Hot” queries that give the search. Also useful thing – you can see what people are interested in recently. . Logically, if you obscurely blog, this feature – an important assistant in selecting the theme the next post.

Recent posts

blogsearch_google_com3

Last updated popular blogs (at the same time, you can find out which of the popular blogs, Google believes that, too, may be useful in the work).

I like these changes!

Official Blog AdWords Blog announced a new feature that allows to separate traffic google.com traffic from partners Google – such as AOL and Ask.com. How? Split: Google search/search partners/content network.” As I have already explained, you just authorized and select the menu item “Split: Google search / search partners / content network.”

The results will be displayed separately. An important step forward. From now on, advertisers will have access to data, which they always miss, but it is – Google will be able to compare the impact of returns of its partners.

Relevant records are in the report center.

There were no couples pitfalls described Adwards:

  1. The format in which we store the data, leads to the fact that all the traffic recorded before January 1, 2007, falls into the category of “Google”.
  2. You may find that the parameters of CTR and Conversion Rate (CR) have slightly lower values.

Of course, advertisers want more – more options, based on these statistics. I think that sooner or later they will.

Ranking index and various documents found on the network, and in order to provide users the most relevant to the issuance, the search engine relies not only on the content of found pages – it also takes into account the quantity and quality of links leading to this page.

The search engine – for example, Google – can decide that your page is relevant user request based on its content and text links pointing to this page.

. It also may try to create some of the “relationship” between the pages, looking at the structure of their reference links. To this end, Google uses a system of Page rank, which calculates the measure of the importance of individual pages by analyzing the network links. This measure is important may be simply represented as a probability that you will be on this page, randomly clicking on links in the process of free Internet surfing.

So, the final ranking of the pages affected by a combination of three main factors: the relevance of content pages your query, the text of links pointing to a page, as well as the measure of the importance of this page, withdrawn from the structure of its referential links to other pages. Google can adjust the ranking of top documents, using the signals of very different nature, yet it is these factors have a decisive influence on what will see whether the page’s end users.

Systems such as Page rank ranking reference are far from ideal. Links can be manipulated in order to remove any page in the top.

The patent, Google received the other day, describing how to identify and neutralize the manipulation of links and thus filter out pages that have received high Page rank through Referral spam.

Link frame and criminal groups

The search engine can see, not whether the links that lead to this page, some specific properties, which are unlikely to have an honest references.

In the patent, Google will highlight two major types of reference of spam – Link Frame and criminal groups. Further detailing how they can be distinguished from completely innocent of links that point to completely innocent page.

Link Frame
Link frame consists of a large number of pages created in the first place to point to a single central page, artificially increasing the measure of its importance. A typical example is a web-shop with a lot of hidden from the user sites that link to its homepage. If the search engine deems appropriate to include these links, they can help online store to get to the top.

Pages, leading to a central resource Link Frame to have very low steppes importance (low Page Rank). At the same time really important resources are likely to have links not only with minor pages and sites with high Page Rank.

Criminal groups
The criminal group is a set of pages that are linked through the so-called ring per link – they refer to each other, mutually increasing their weight and misleading search engines. If a search engine does not summit weed out all those links, such behavior would facilitate fraudulent promotion in the top.

Pages, forming a criminal group, have no inclination to invoke somewhere outside, outside the group. This helps to distinguish them from normal pages, earned its credibility honest way.

Reactions to Artificial increase in importance
If a page or page group has been caught in the spam link, the search engine, in strategy with the patent, should try to calculate the amount of “artificially created by the importance” to adequately adjust the extradition.

In the first phase of a living person or a special algorithm must be studied in detail discovered pages to make sure that they are indeed spam. If the check has a positive result could be taken the following measures:

  1. Links posted on this page may not take into account in the Page rank.
  2. Weight of links posted on this page may be proportionally reduced (links to other sites are becoming less important).
  3. Links from this page may be predetermined penalty that reduces their importance.
  4. The importance of the page may be reduced way, not directly connected with the Page rank.
  5. The importance of the page may be reduced way, not directly connected with the system of Page rank, but its links to get your fine.

In the patent describes the mathematics underlying the proposed mechanism for the disposal of link frame and criminal groups. It is worth to explore it – especially if you’re really interested in how Google intends to fight link spam:

<!– /* Font Definitions */ @font-face {font-family:Verdana; panose-1:2 11 6 4 3 5 4 4 2 4; mso-font-charset:0; mso-generic-font-family:swiss; mso-font-pitch:variable; mso-font-signature:536871559 0 0 0 415 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-parent:”"; margin:0in; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:12.0pt; font-family:”Times New Roman”; mso-fareast-font-family:”Times New Roman”;} a:link, span.MsoHyperlink {color:blue; text-decoration:underline; text-underline:single;} a:visited, span.MsoHyperlinkFollowed {color:purple; text-decoration:underline; text-underline:single;} @page Section1 {size:8.5in 11.0in; margin:1.0in 1.25in 1.0in 1.25in; mso-header-margin:.5in; mso-footer-margin:.5in; mso-paper-source:0;} div.Section1 {page:Section1;} –>

Method for detecting link spam in hyperlinked databases

The official news from Google. Communicate with webmasters in the chat allows you to identify common views, which could correspond to reality in the past, but lost its relevance to today. For example, the my recent conversation with several friends on the optimal structure URL. One of them worried about the use of dynamic addresses, because (he said) “search engines have difficulty with them.” Another argued that the processing of dynamic URL no longer presents any difficulties for search engines. One of them also added that he never plainly did not understand the fuss, “about the dynamic and static addresses. At this point, I decided to write a topic dedicated to this issue. First, let’s be clear about what really is at stake.

What is a static URL?

Static URL – this is the address, which does not change, and typically does not contain a no parameters. The static address can look, for example, because: http://www.example.com/archive/january.htm. You can include the search for static URLs on Google, by typing filetype: htm in the search box. Updating these pages often takes quite a long time, especially if the amount of information is growing rapidly, because the code of each page have to have to write and edit on an individual basis. That is why the webmaster is having to deal with frequently updated site – for example, Internet shopping, forums, blogs or content management systems – often use dynamic addresses.

What is a dynamic address?

If a content site is stored in a database and display on demand, as a rule, use dynamic addresses. The site in this case plays the role of a template for content that exists independently of its visual presentation. Typically dynamic URL look like this: http://code.google.com/p/google-checkout-php-sample-code/issues/detail?id=31. Sign of dynamic addresses used symbols? = & Dynamic addresses have the drawback that several different URL may correspond to the same content. Thus, users That is why the webmaster’s often try to adjust the alignment of dynamic addresses to static form.

Do I need to mask the URL dynamic by static?

Are some key points that need to be remembered in connection with dynamic addresses?

  1. Create and maintain a mechanism to correctly converts the dynamic address to static, it is difficult.
  2. Where reliable and easier to use dynamic addresses, and allow ourselves to solve the problem of finding and ignoring unnecessary settings.
  3. If you intend to convert the same address, please remove them from all optional parameters, but leave them dynamic.
  4. If you intend to issue a static address instead of dynamic, you should duplicate your content static equivalent.

What are the addresses of Google is easier to perceive – static or dynamic?

I was dealing with a lot of webmasters that thought, as mentioned above, friends, that a static or external static addresses are an advantage in terms of indexing and ranking their sites. This belief is based on the assumption that the search engines have difficulties with crawling and analysis of addresses containing the IDs of sessions or units of content. Anyway, I was able to Google much progress in the two designated areas. Yes, the static address can be a slight advantage in terms of clickable, because the user is easier to read. Nevertheless, the use of web sites based on the database does not entail any significant loss in terms of indexing and ranking. Thus, the use of intact dynamic addresses preferably attempts to hide the parameters and make them look static.

Myth: “Dynamic URL is not indexed.”
Fact:
I have no problem to index the dynamic address and interpret the different parameters. I may have problems with crawling and ranking dynamic URL, if you bring them to the static type and hides the parameters that are valuable information for the Google-bot. It follows a recommendation – not to convert the dynamic URL to static. In general, best to use static content with static URLs, but if you decide to use dynamic content, you must provide us with an opportunity to analyze the structure of your URL. In other words, need not have to mask the parameters, leading them to a static view.

Myth: “Dynamic addresses must contain a minimum of three parameters.”
Fact:
The restrictions on the number of parameters does not exist, however, the general rule should be the desire (Short URL (this is applies to any address, and static and dynamic). Perhaps you should remove the parameters are not important Google-Bot, and provide users with a nice dynamic URL. If you are not completely sure what the parameters are subject to removal, I suggests not to remove anything – our system will drop all unnecessary. Illiterate conversion significantly complicates the task of selection of important parameters, ie, prevents us from properly analyze URL, which eventually leads to information loss.

. Then highlight a few of the most likely probable.

Does this mean that I should avoid the transformation of dynamic URL?

These are our recommendations – until i am not talking about removing unnecessary and potentially problematic parameters. Reduction of dynamic addresses to static thinking can prevent us from correctly interpreting your information whenever possible. If you intend to provide users with a static equivalent of your site, you should think about the conversion not only addresses, but also the structure of the content – in a way that the information in the output and really looked static. For example, you can generate a file for each of the possible addresses, and to make these files available elsewhere on the site so that users and search engines treated exactly to these files. In any case, the conversion of addresses without creating static copies of the content risks more harm than good. It shows us the dynamic address – I will be able to identify them all superfluous.

Can you give an example?

If a standard format for your dynamic address looks like this: foo? Key1 = value & key2 = value2, I recommend you leave it unchanged. Google itself will determine what options you can remove it. You can, and can remove themselves from the URL are optional for the user settings, but be very careful that you do not accidentally away anything important. Here is another example of the dynamic address to multiple parameters:

www.example.com/article/bin/answer.foo?language=en&answer=3&sid=98971298178906&query=URL

  • language = en – indicates the language of Article
  • answer = 3 – art is the number 3
  • sid = 8971298906- the session ID is 8971298178906
  • query = URL – article was found at the request of the next types: [URL]

Not all the parameters of this request are the useful information, so bringing to mind the URL www.example.com/article/bin/answer.foo?language=en&answer=3 probably will not create any problems – of addresses have been removed only optional parameters.

This is followed by several addresses that look like static, but can hinder crawling much stronger than the standard dynamic address:

  • www.example.com/article/bin/answer.foo/en/3/98971298178906/URL
  • www.example.com/article/bin/answer.foo/language=en/answer=3/sid=98971298178906/query=URL
  • www.example.com/article/bin/answer.foo/language/en/answer/3/sid/98971298178906/query/URL
  • www.example.com/article/bin/answer.foo/en,3,98971298178906,URL

Transformation of dynamic addresses to something similar might lead to crawler have on several occasions to index the same content that is available for a variety of locations – with a variety of session and the types of query. This format addresses prevents Google understand that URL and 98971298178906 have no directly related to the content underlying the given address. Next – an example of the correct transformation, where all optional features have been removed:

  • www.example.com/article/bin/answer.foo/en/3

While Google handles these addresses quite correctly, and did we want to warn you of the use of such transformations, because the mechanism is difficult to support and needs to be updated whenever the dynamic to the original address is added a new option. If this does not happen again at the exit will be a static-looking address important parameter hidden from searches. Therefore, most often the best solution is to use the continuous dynamic address. If you decide to remove them from the optional settings, remember that ultimately they must be dynamic, as in the example already:

  • www.example.com/article/bin/answer.foo?language=en&answer=3

I hope that the paper will be useful, and that ultimately it will help dispel the various conjectures related to the problem of dynamic URL.

Source: http://googlewebmastercentral.blogspot.com/2008/09/dynamic-urls-vs-static-urls.html