Determine bystrobotovskuyu impurity in Yandex

In a recent article, "Cleaning the organic impurities from the issuance of Yandex" I mentioned this as an impurity as the recent results in the medium-SEO professionals bearing the name "bystrobotovskoy" impurities.

Identifying whether a particular document is relevant to the extradition Yandex bystrobotovskoy impurities can be very useful in solving analytical problems. The fact that the documents are indexed bystrobotom and falling in the issuance of a short time, are ranked differently from the main documents from the index, and therefore the issuance of these need to be cleaned in the analysis of the basic algorithm.

Then I recommended to identify bystrobotovskuyu impurity by the presence of specific markers of freshness document ( «N minutes ago", "of N hours ago" "yesterday," "the day before yesterday," or just a date no older than 3-4 days). But it seems like the mark is not a necessary feature of a document indexed bystrobotom.

I will explain the example. For example, as of this writing in the index Yandex present the following document without special mark of freshness in a snippet:

However, if we look at the cached copy, we can see that the document has been indexed in 3 minutes after his appearance on the site (according to the display time that has elapsed since the publication of the user of this material):

So soon after the appearance of documents will be more likely to get into the index through bystrobotovskuyu impurity. However, as mentioned above, the freshness of the label, which is characteristic for this impurity, in this document no snippet.

By studying the behavior of documents from bystrobotovskoy impurities, I noticed an interesting feature. In contrast to the documents of the main index, bystrobotovskie documents are displayed in the extradition request, consisting of a bundle of a text query and document-operator even in the case where the document is irrelevant text query from this bunch.

For example, take the form of a text query gibberish, issuing on which empty:

And add to this document-text query operator url: the address of the document as a value. Contrary to the logic of the document is displayed in the issue:

Moreover, the issue of extending the entire site using document-site: operator, we can see that in issue in the vast majority of documents are bystrobotovskoy label in a snippet.

And saved copies of those who have this label is not available, there are indications that the document is indexed bystrobotom.

Thus, it is possible with some confidence to say that in a similar way we can verify the document without bystrobotovskoy label in a snippet on the subject of whether or not it is indexed bystrobotom. Usually, it works on the documents of three-four days ago. We strongly recommend doing double-check with a few quite different from each other choices Abracadabra. The fact that some abracadabra Yandex interprets as typos and tries to pick replacement options, without notifying the user, and in this case it is possible to get a false positive response method.

Interestingly, the addition of a text query under consideration conjunction operators group "Morphology and search context" + ( "plus" - finds documents in which is necessarily present the selected word) or "" ( "quotes" - search for quotation), the wonderful effect bystrobotovskoy impurity disappears:

We use this fact to make sure that the cached copy of the document, which was discussed in the first example, is the same as what is in the index (it can in fact be otherwise, as I wrote in my article "Saving a copy of pages - this is not the that is in the index " ).

Would not hurt to check it out, because it is on the page, we saved up to make assumptions about the nature of its bystrobotnom indexing. This exact phrase their cached copy of the publication of posts is under consideration Page:

This means that in the index really is the copy that is shown as stored. By the way, for documents from impurities bystrobotnoy I have not seen examples skew stored and indexed copies.