SEO. Internal optimization. Part 1.

From the previous note we learned about the concept of SEO. We also know what a search index is and how it is updated using search robots.

In an ideal world, a search robot should load a page, analyze its content, and create a list of keywords that best correspond to the content. (By the way, just in case, keywords - these are, roughly speaking, search phrases, I hope there are no questions about this. =)) But we understand that it is not possible. In order to do this, a search robot must have human thinking, a semblance of intelligence. (Although, despite everything, robots are very smart, so don't underestimate them, but more on that later).

In practice, everything happens differently. Search robots can generate keywords themselves (at least, Google bots definitely can). But they do it quite slowly and not always correctly (indeed, how to extract from a set of words (which is how it looks to a robot) those that are actually important? The simplest algorithm is to consider the most frequently occurring phrases as search keywords, making adjustments for truly important, significant words, etc., but practice shows that search robots are not in a hurry to do this. Perhaps this is because keywords do not have to be the most commonly used ones.)) It is much easier for a robot to check the ready-made keywords and see how well they correspond to the content.

Thus, we have a goal:

Before each page, you need to provide a list of keywords.

Keywords are described in the so-called meta tags.

For "advanced" users, it looks like this:

<meta name="keywords" content="keyword1, keyword2, keyword3" />

By the way... Regarding the use of commas to separate keywords... Whether it is necessary or not is not entirely clear to me, recently, especially for Google, it is better not to separate them, why - it may become clear a little later.

But the keywords should also be composed correctly.

So, let's try to highlight the "rules" for composing search keywords, with explanations.

1. Search keywords should reflect the content of the article as much as possible.

   Search robots are really smart. For them, it is not difficult to check the correspondence of keywords to the content. They can analyze semantically. They understand synonyms. From this we see the following rule:

2. Search keywords should not contain widely used words, for example, those that appear in almost any text/conversation

   Indeed, how can such words reflect the content of the article if they are everywhere? Examples of such words are: I, here, forever, exactly, except, ever, help, etc. (This is just a small example). And these are not empty words. In fact, there are so-called stop words. Stop words are words ignored by search engines. You can see part of the list of these words, for example, here: http://allproall.com/stopwords/ This is also just a small part. It is desirable to be able to distinguish important things from garbage. Roughly speaking, search engines try to discard stop words when processing (from a user's search query), but they can partially rely on them. In the case of resources, they only bring harm, because in search queries they exist only in context. In short, with your keywords that are in the stop list, there will be no search.

3. Keywords should also be emphasized in the text using the <strong></strong> tags (Visually - bold)

   Such actions increase trust (and, by the way, practice has shown to me the exceptional importance of such highlight, especially for Google) in the keywords specified in the meta tags. It is as if you are justifying yourself, saying: look, Google, my keywords really correspond to the content of the text, of course, you can do semantic analysis, but it's a complex procedure, look first at how I argue this set of keywords =)) (I got carried away a bit, didn't I?)

4. There shouldn't be too many keywords

   In principle, this rule can be derived from the previous ones - with the right selection, there should not be too many of them

5. Avoid duplications.

   Keywords should not be duplicated - neither completely nor partially (in principle, if you don't separate them with commas, it should be clear). This applies, of course, to meta tags. On the contrary, in the text, all occurrences of keywords should be highlighted.

6. Add a page description in the corresponding attributes of the meta tag (Also automated in most CMS)

   <meta name="description" content="Description of the article" />

   For search engines, generating a description is even more difficult than generating keywords, and they almost never do it correctly (of course, there are exceptions) and meaningfully. Therefore, take care to add a description. The description should, like the keywords, correspond to the content. Ideally, it is a fragment of your text. In the worst case, it is a very close retelling of key points.

That's all for now. In the next notes, we will continue to develop the topic of internal site optimization. If anyone has any thoughts about what was written above, I would be happy to hear your opinion, because in this matter I am actually an amateur and my activity is completely different, but I would like to write something more or less useful. =)