We analyze what the semantics of the site consists of.
В прошлой статье под названием: «What is the semantic core of a site?» I have provided general information on this important tool for promoting a resource on the Internet. Today I will try to reveal its structure as fully as possible and give recommendations so that the semantics of the site is as complete as possible.
Before proceeding to the analysis, I want to give a small reference to Yandex Wordstat. When creating a semantic core, you will use this service all the time. Here you should go back to the previous article and say that the search engine does not like to share search queries with webmasters.
If you type the query “semantics” in the string Yandex Wordstat:
Then it shows you not the number of requests from users who are looking for content for this query, but the possible number of ad impressions for the query “site semantics”. However, it is generally assumed that this is an approximate number of requests. It is also important to keep in mind that this is information for the current month.
The semantic core of the site is usually divided into subgroups for ease of use.
It looks like this:
- High frequency queries – a single-word query that is entered in the search bar, such as”iphone”. There is not much use for this query, so it is not clear what exactly the user means when he enters it into the search bar. It is not clear what his intentis at the same time.
- Mid-frequency queries – more meaningful queries, for example, “red iphone xr” – the user is already looking for something more specific.
- Low frequency queries – deeper query: “red iphone xr problems”
- Ultra low frequency requests – “red iphone xr problems with sound from the speaker»
For these groups, we need to say the following. It is not necessary that the number of requests in all groups should be approximately the same. In the field dedicated to electronic technology, the number can be one, and in the field related to machine tool construction completely different.
These 4 groups have an additional division that looks like this:
- Information requests – the above examples of requests just fall under information requests. The user has a specific problem and is looking for information about it.
- Commercial requests – “fix red iphone xr price” or “buy red iphone xr” – are quite specific requests.
- Geo dependent requests – “buy a red iphone Moscow Butovo” a specific request showing that a person needs a phone in their area of residence.
- Geo independent queries are those that are not tied to a specific location.
- Seasonal requests – they are divided into event requests. For example, the next football championship or the release of the iPhone SE 2. And on regular requests that go in bursts throughout the year. Sometimes regular ones are tied to certain dates. For example, holidays, March 8 is the season of planting vegetables. That is, the number of requests increases dramatically, picks up a peak, holds for a while and “falls” back.
Recently, another term for queries has been gaining popularity, and I like it more, since the semantics of the site in this case takes on a broader meaning and the possibility of receiving traffic increases.
- Head – here is the most uncertain intent. The same high frequency queries as above. There is no point in optimizing the content for it.
- Body – this includes queries consisting of two or three words. By and large, this is not yet an explicit intent. For example, it may be the phrase: “buy an iphone xr in Moscow”, but nevertheless it already represents a certain commercial interest.
- Long tail – these are words and phrases that consist of 4-6 words in length. These queries are the most interesting, because they clearly state the request of a person when he is looking for something on the network. And, it is these requests that make up more than 50% of traffic to sites. It doesn’t matter if they are informational or commercial.
Where to get the queries for a semantic website?
There are keyword sources and aggregators. The latter are databases of keywords, they are compiled on the basis of sources. The data in them is often outdated or covered by competitors. In addition, there is a significant drawback – they are often paid. Whether you use their resources or not, you need to look at the situation.
The best source at the moment is Yandex Wordstat. It is better not to use Google for this, since the issue of keys is very littered and then you will have to spend a huge number of working hours to clean out the Augean stables of the site’s semantics.
Keywords that we are interested in are found in search engine databases and search suggestions.
In the next article, we will talk about what is the key idea of optimization and how to work with it as efficiently as possible.