Hello everyone! When you keep a blog or content site, there is always a need for compiling semantic kernel For the site, cluster or articles. For convenience and system, it is better to work with a semantic kernel according to the well-established scheme.

In this article we'll consider:

• how is the collection of the semantic kernel for writing an article;
• what services can and need to be used;
• how to fit the keys to the article;
• my personal selection experience.

How to collect semantic kernel online

1. First of all, we need to use the service from Yandex -. Here we will make an initial sample of possible keys.

In this article, I will collect Xia on the topic, "how to lay a laminate." Similarly, you can use this instruction on the preparation of the semantic kernel for any topic.

1. If our article is on the topic "How to lay a laminate", then we will introduce this request to obtain information about the frequency in wordStat.yandex.ru..

As we can see, in addition to the target request, we fell a lot of similar requests containing phrase "Laminate"Here you can sow all unnecessary, all the keys that will not be considered in our article. For example, we will not write on similar topics, such as "How much is it worth putting laminate", "Sneorically laid laminate" etc.

To get rid of many knowingly non-suitable queries, I recommend using operator "-" (minus).

1. We substitute minus, and after all the words are not the topic.

1. Now, we allocate everything that remains and copy requests to notepad or in Word.

1. By inserting everything into the Word file, running in the eyes and delete everything that will not be disclosed in our article if false requests are found, then you will be able to check the combination of keys for the presence of them in the document. Ctrl + F., window opens ( side panel Left), where we enter the sought words.

The first part of the work is completed, now you can check our workpiece of the semantic kernel of Yandex to clean frequency, the quotation operator will help us.

If the words are not enough, it is easy to do right into the Wordstat, substituting the phrase in quotes and finding clean frequency (quotes show how many requests were with the content of this particular phrase, without additional words). And if, as in our example of the semantic core of the article or site, it turned out a lot of words, it is better to automate this work using the Mutagen service.

To get rid of numbers Use the following steps with Word Document.

1. Ctrl + A. - To highlight the entire contents of the document.
2. Ctrl + H. - Calls a window to replace characters.
3. Substitute in the first line ^ # and press "replace everything" This combination will delete all the numbers from our document.

Be careful with the keys that contain the numbers, the above actions can change the key.

Semantic kernel selection / articles online

So, I wrote in detail about the service. Here we will continue learning the preparation of the semantic kernel.

1. We go to the site and use this program to compile the semantic kernel, since i have not met better alternative.

1. First, Otrace Clean frequency for this pass "Parser Wordstat" → "Mass Parsing"

1. Insert all our selected words from the document in the parser window (Ctrl + C and Ctrl + V) and choose the "frequency of the Wordstat" in quotes.

This automation of processes is worth total 2 kopecks for the phrase! If you pick the semantic kernel for the online store, then this approach will save you time for the singering penny!

1. We click to check for check and, as a rule, 10-40 seconds (depending on the number of words) you can download words already having frequency in quotes.

1. The output file has an extension. CSV it opens to Excel. We start filtering the data to make the semantic kernel online.

• We finish the third column, it is needed to set competition (next step).
• We put the filter for all three columns.
• Filter "Descending" column "Frequency".
• Everything has frequency below 10 - removed.

We got a key listwhich can be used in the text of the article, but first need to check them for competition. After all, if this topic is made on the Internet along and across, does it make sense to write by this key Article? The probability that our article on it will go to the top, extremely small!

1. To check competition online semantic kernel go to "Competition".

1. We begin to check each request and the value of the competition is substituted into the appropriate column in our Excel file.

Cost Checks one key phrase is 30 kopecks.

After the first replenishment of the balance, you will be available 10 free checks Every day.

To determine the phrases that are worth writing an article take the best frequency ratio-competition.

Writing an article is:

• frequency of at least 300;
• competition is not higher than 12 (better less).

Drawing up the semantic kernel online applying low-competitive phrases Give you traffic. If you have a new site, then it will not appear immediately will have to wait from 4 to 8 months.

Almost any subject, you can find SC and RF with low competition from 1 to 5, in such keys really per day receive from 50 visitors.

To cluster semantic kernel requests, they will help you make the right structure of the site.

Where to insert the semantic kernel in the text

After collecting the semantic kernel for the site it's time to enter key phrases into the text and here are some recommendations for beginners, and those who "do not believe" into the existence of search traffic.

Signing rules in keyword text

• The key is enough to use 1 time;
• words can be inclined by cases, change places;
• you can dilute phrases in other words, well, when all key phrases are diluted and readable;
• you can remove / replace prepositions and question words (such as, and so on.);
• you can insert into the phrase signs "-" and ":"

eg:
There is a key: "How to put a laminate with your own hands" In the text, he may look like this: "... In order to lay the boards of laminate with your own hands we need ..." or so "Everyone who tried to put laminate with their own hands ...".

Some phrases already contain otherFor example phrase:
"How to lay a laminate with your own hands in the kitchen" already contains a key phrase "How to put a laminate with your own hands". In this case, it is allowed to omit the second, as it is already present in the text. But if there are few keys, it is better to use it in the text or in Title, or in Description.

• if it is impossible to shove the phrase into the text, then leave, do not do this (at least two phrases can be used in Title and Description and do not write them into the body of the article);
• Before, one phrase is the title of the article (the fattest frequency and competition)The webmasters language is H1, this phrase is enough to use once in the body of the text.

Contraindications to the Scriptures of Keys

• it is impossible to separate the key phrase of the comma (only in extreme cases) or point;
• it is impossible to enter the key to the text in direct form so that it will look no naturally (not readable);

Title and page description

Title and Description. - This is the title and the description of the page, they are not visible in the article, Yandex and Google shows them search results user.

Speech Rules:

• title and description should be "journalistic", that is, attractive to the transition;
• to contain the thematic (relevant query) text, for this fit into the title and in the description of the key phrases (diluted).

General symbol requirements at the plugin ALL IN ONE SEO PACKThe following:

• Title - 60 characters (including spaces).
• Description - 160 characters (including spaces).

Check for plagiarism your creation or obtained from you can with.

On this with the topic, what to do with the semantic core after compilation we figured out. In conclusion, I will share my own experience.

After compiling the semantic kernel according to the instructions - my experience

You might think that I'm sad to you, something is not plausible. So as not to be unfounded here is a screenshot of statistics for the last, (but not the only one) year of this site, how I managed to rebuild the blog and start receiving traffic.

This training is the preparation of the semantic kernel, though long but effective, because in the site building the main thing the right approach And patience!

If you have any questions or have criticism, then write in the comment, I will be interested, also share your experience!

In contact with

IN this moment For search promotion, such factors as the content and structure are played as important as an important role. However, how to understand what to write text what sections and the Create Pages on the site? In addition, you need to find out exactly what exactly the target visitor of your resource is interested. To answer all these questions, you need to assemble the semantic kernel.

Semantic kernel - List of words or phrases that fully reflect the subject of your site.

In the article, I will tell you how to pick it up, clean and split into the structure. The result will be a complete structure with queries clustered through pages.

Here is an example of a core of requests broken into the structure:

Under clustering I understand the breakdown of your search queries On separate pages. This method will be relevant both to promote Yandex and Google. In the article I will describe completely free way Creating a semantic kernel, however, I will show options with various paid services.

After reading the article, you will learn

• Select the right requests for your subject
• Collect the most complete nucleus of phrases
• Cleant from uninteresting requests
• Group and create structure

Collect the semantic core you can

• Create a meaningful structure on the site
• Create a multi-level menu
• Fill the pages with texts and write on them from metapping and title
• Collect the position of your site on requests from search engines

Collecting and clustering of the semantic kernel

The proper compilation for Google and Yandex begins with the definition of the main key phrases of your subject. For example, I will demonstrate its compilation on a fictional online clothing store. There are three ways to collect the semantic kernel:

1. Manual. Using the Yandex Wordstat service, you enter your keywords and your hands choose the phrases you need. This method is fast enough if you need to assemble the keys to one page, however, there are two minuses.
   • The accuracy of the "Chromas" method. You can always miss any important words if you use this method.
   • You will not be able to assemble the semantic kernel on a large online store, although you can use the Yandex Wordstat Assistant plugin to simplify - the problem does not solve it.
2. Semi-automatic. In this method, I assume the use of the program for collecting the nucleus and the further manual breakdown on the sections, subsections, pages, etc. This method The compilation and clustering of the semantic kernel in my opinion is the most effective because It has a number of advantages:
   • Maximum coverage of all themes.
   • Quality breakdown
3. Auto. Nowadays, there are several services that offer fully automatic kernel collection or clustering your requests. Fully automatic option - I do not recommend to use, because The quality of collection and clustering of the semantic kernel is at the moment quite low. Automatic queries clustering - gaining popularity and takes place, but you need to still merge some pages with your hands, because The system does not give an ideal ready solution. And in my opinion you just get confused and you will not be able to plunge into the project.

For compiling and clustering a full-fledged proper semantic nucleus on any project in 90% of cases, I use a semi-automatic method.

So, so that we need to perform the following steps:

1. Selection of requests for themes
2. Picking kernel on requests
3. Cleaning from non-target queries
4. Clustering (break the phrases on the structure)

An example of sewing semantic kernel and grouping on the structure I showed above. I remind you that we have an online clothing store, start paving 1 point.

1. Selection of phrases for your subject

At this stage, we need tool Yandex Wordstat, your competitors and logic. In this step, it is important to assemble a list of phrases that are thematic high-frequency requests.

How to select requests for collecting semantics with Yandex Wordstat

Come on the service, choose the city you need (a) / region (s), drive the most "fat" in your opinion requests and look at the right column. There you will find thematic words you need, both on other sections and frequency synonyms for the inscribed phrase.

How to select requests before making a semantic kernel using competitors

Enter the most popular requests in the search engine and select one of the most popular sites, many of which you are most likely, and you know.

Pay attention to the main sections and save yourself phrases you need.

At this stage, it is important to do correctly: to maximize all sorts of words from your subject matter and do not miss anything, then your semantic kernel will be the most complete as possible.

Applicable to our example, we need to make a list of the following phrases / keywords:

• clothing
• Footwear
• Boots
• Dresses
• T-shirts
• Underwear
• Shorts

What phrases enter meaningless: Women's clothing, buy shoes, dress on graduation, etc. Why? - These phrases are "tails" of "clothing" requests, "shoes", "dresses" and will be added to the semantic kernel automatically at the 2nd stage of collection. Those. You can add them, but it will be meaningless double work.

What keys do you need to fit? "Harfoots", "boots" are not the same as "boots". It is the word form that is important, and not that single words or not.

Someone has a list of key phrases long, and who he consists of one word - do not be afraid. For example, the online store of the doors for the preparation of the semantic kernel is quite possibly enough words "doors".

And so, at the end of this step, we should have a similar list.

2. Collection of requests for semantic kernel

For proper full-fledged collection, we need a program. I will show an example at the same time on two programs:

• On the paid - KeyCollector. For those who have, or who wants to buy.
• Free - Slovoeb. Free program for those who are not ready to spend money.

Open the program

Create new project And let's call it, for example, MySite

Now, for further collecting the semantic kernel, we need to do a few things:

Create new account On Yandex mail (old it is not recommended to use due to the fact that it can be banned for many requests). So, you created an account, for example [Email Protected] With password Super2018. Now you need to specify this account in the settings as Ivan.IVANOV: Super2018 and press the "Save Changes" button at the bottom. Read more - on screenshots.

We choose the region to compile the semantic kernel. You need to choose only the regions in which you are going to move and click Save. From this will depend on the frequency of requests and whether they will fall into the collection in principle.

All settings are completed, it remains to add our wicked phrase in the first step and click the "Start Collection" button of the semantic kernel.

The process is fully automatic and enough long. You can do coffee while making coffee, and if the theme is wide, for example, like the one that we collect - it is for several hours 😉

Once all phrases gather you to see something similar:

And on this stage is completed - proceed to the next step

3. Cleaning the semantic kernel

First, we need to delete the requests that we are not interested in (neurozzi):

• Associated with another brand, for example, "Gloria Jeans", "ECCO"
• Information requests, for example, "I wear boots", "Jeans Size"
• Similar on the subject, but not related to your business, for example, "B Clothing", "Women's wholesale"
• Requests, in no way related to theme, for example, "Sims Dresses", "Cat in Boots" (such requests after the selection of the semantic kernel there are quite a lot)
• Requests from other regions, metro, districts, streets (no matter what region you collected requests - another region still comes across)

Cleaning must be performed manually as follows:

We enter the word, press "ENTER", if in our created semantic kernel it is those phrases that we need, we allocate the found and click delete.

I recommend entering the word not entirely, but using the design without prepositions and endings, i.e. If we write the word "GLORY", it will find phrases "buy in Gloria Jeans" and "buy in Gloria Jeans". When writing "Gloria" - "Gloria" would not have been found.

Thus, you need to go through all the points and remove from the semantic kernel unnecessary requests to you. This may take a considerable time, and maybe it will turn out that you remove most of the collected queries, but the result will be a full-fledged clean and correct list of all kinds of promoted requests for your site.

Unload now all your requests to Excel

You can also massively remove non-target queries from semantics, provided that you have a list. You can do this with stop-words and it is easy to do for a typical group of words with cities, metro, streets. A list of such words that I use you can download at the bottom of the page.

4. Clustering of the semantic kernel

This is the most important and interesting part - it is necessary to divide our regulations on the pages and sections that in the aggregate will create the structure of your site. A bit of theory - than guided by requests:

• Competitors. You can pay attention to how clustering the semantic kernel from your competitors from the top and act in a similar way, at least with the main sections. And also to watch which pages are in the extradition of low-frequency requests. For example, if you are not sure to "do or not" a separate section for the "Red Leather Skirts" request, drive the phrase into the search engine and see the issuance. If the issuance contains resources where there are such sections, it means that it makes sense to make a separate page.
• Logics. Make the entire grouping of the semantic kernel using logic: the structure must be clear and represent the structured page of the pages with categories and subcategories in your head.

And a couple more advice:

• It is not recommended to install less than 3 requests to the page.
• Do not make too many nesting levels, try to do so that they were 3-4 (Site.ru / Category / Subcategory / Sub-subcategory)
• Do not make long URLs if you have many levels of nesting when clustering the semantic kernel, try to reduce the URLs of the highly on the categories hierarchy, i.e. Instead of "vash-site.ru/zhenskaya-odezhda/palto-dlya-zhenshin/krasnoe-palto", do "vash-site.ru/zhenshinam/palto / krasnoe"

Now to practice

Core clustering on the example

To begin with, we will share all requests for the main categories. Looking at the logic of competitors - the main categories for the clothing store will be: men's clothing, women's clothing, children's clothing, as well as a bunch of other categories that are not tied to the floor / age, such as simply "shoes", outerwear.

We group the semantic kernel is when excel help. Open our file and act:

1. We divide on the main sections
2. Take one section and divide it on the subsections

I will show on the example of one section - men's clothing and its subsection. In order to separate the keys from others, you need to highlight the entire sheet and click Conditional Formatting-\u003e Code Selection Rules-\u003e Text contains

Now in the window that opens, we write "Husband" and press Enter.

Now all our keys on men's clothing are highlighted. It is enough to use the filter to separate the selected keys from the rest of our assembled semantic kernel.

So turn on the filter: you need to highlight a column with requests and press the sorting and filter-\u003e filter

And now sorted

Create separate sheet. Cut the selected lines and insert them there. This way you will need to continue and break the kernel.

Change the name of this sheet to "Men's Clothing", a sheet where everything else is the semantic kernel, name "all requests". Then create another sheet, name it "Structure" and put it first. On the Structure page, create a tree. You should succeed like this:

Now we need to divide the large section of men's clothing on subsections and sub-subsections.

For the convenience of use and transitions according to your clustered semantic kernel, place references to the structure to the appropriate sheets. To do this, right-click on the desired item in the structure and do as in the screenshot.

And now we need to separate your hands to share requests, simultaneously removing what may not have been noticed and removed at the kernel cleaning phase. Ultimately, thanks to the clustering of the semantic kernel, you should have a structure similar to this:

So. What we learned to do:

• Choose requests you need to collect the semantic kernel
• Collect all possible phrases for these requests
• Clean
• Cluster and create structure

That thanks to creating such a clustered semantic kernel, you can do on:

• Create a structure on the site
• Create a menu
• Writing texts, metauring, Taitla
• Collect positions to track speakers on requests

Now a little about programs and services

Programs for collecting the semantic kernel

Here I will describe not only the programs, but also plugins and online services that I use

• Yandex Wordstat Assistant is a plugin, thanks to which, it is convenient to select requests from Vordstat. Great for quick compilation of the core on a small site or 1 page.
• Caicollector (Slobel - free version) - a full-fledged program for clustering and creating a semantic kernel. Evoids great popularity. A huge number of functional in addition to the main direction: selection of keys from a heap of other systems, the possibility of autoclasterization, collecting positions in Yandex and Google and much more.
• Just-Magic - Multifunctional online service for making core, authoring, testing text quality and other functions. Service conditionally free, for full-fledged work you need to pay a subscription fee.

Thank you for reading the article. Thanks to this step-by-step manual, you can make a semantic kernel of your site to promote in Yandex and Google. If you have any questions left - ask in the comments. Below are bonuses.

Frequently beginner webmasters, facing the need to create a semantic nucleus, do not know where to start. Although nothing complicated in this process. Simply put, you need to assemble a list of key phrases for which Internet users are looking for information on your site.

The more fully and more precisely, it will, the easier it is to write a good copy of the copy, and you get high positions in the search for the desired requests. How to make big and high-quality semantic kernels correctly and what to do with them next that the site goes to the top and collect a lot of traffic, and it will be speech in this material.

The semantic kernel is a set of key phrases entrusted by meaning, where each group reflects one need or user's desire (intent). That is, what a person thinks about, driven by his request to the search string.

The entire process of creating a nucleus can be submitted in 4 steps:

1. Come across a task or problem;
2. In the head we formulate how to find its solution through the search;
3. Drive a request to Yandex or Google. In addition to us, other people do the same;
4. The most frequent call options fall into analytics services and become key phrases that we collect and group for needs. As a result of all these manipulations, the semantic kernel is obtained.

Does it be necessary to select key phrases or can you do without it?

Previously, semantics were in order to find the most frequency keywords on the topic, enter them into the text and get good visibility on them in the search. Last 5 years search engines They seek to go to the model where the relevance of the document request will be assessed not by the number of words and a variety of their variations in the text, but according to the evaluation of the intent revelation.

Google has started in 2013 with the Hummingbird algorithm, Yandex in 2016 and 2017 with Palee and Queen technologies, respectively.

Texts written without Xia will not be able to fully disclose the topic, which means compete with the top in high-frequency and mid-frequency requests will not work. Bet on low frequency requests It makes sense - too little traffic on them.

If you want and in the future successfully promote yourself or your product on the Internet - you need to learn how to make the right semantics, which fully reveals the needs of users.

Classification of search queries

We will examine 3 types of parameters for which keywords are evaluated.

In the frequency:

• High-frequency (HF) - phrases defining the topic. Consist of 1-2 words. On average, the number of search queries begins from 1000-3000 per month and can reach hundreds of thousands of impressions depends on the subject matter. Most often under them are the main pages of sites.
• Medium-grade (sch) - separate directions in the subject. Preferably contain 2-3 words. With accurate frequency from 500 to 1000. Usually categories of a commercial site or theme for large information articles.
• Low-frequency (LF) - requests relating to the search for a specific answer to the question. As a rule, from 3-4 words. It can be a product card or article topic. On average, you are looking for from 50 to 500 people per month.
• When analyzing the metric or data of statistical meters, you can meet another kind of microch of the keys. These are phrases that often ask once on the search. It makes no sense to sharpen under them the page is not. It is enough to be in the top of the LF, which includes them.

For competitiveness:

• Highborne (VC);
• Mediterranent (SC);
• Longcover (NK);

For needs:

• Navigation. Express the user's desire to find a specific Internet resource or information on it;
• Information. Characterized by the availability of information in obtaining information as an answer to the request;
• Transaction. Directly associated with the desire to make a purchase;
• Fuzzy or common. Those on which it is difficult to accurately determine the intent.
• Geo-dependent and geon-dependent. Reflect the need to search for information or make a transaction in its city or without regional binding.

Depending on the type of site, the following recommendations can be given when selecting key phrases for the semantic kernel.

1. Information resource. The main emphasis should be done on the search for the articles in the form of sch and low-competition LF requests. It is recommended to disclose the topic wide and deep, dragging the page for a large number of LF keys.
2. Online store or commercial site. We collect HF, Sch and LF, which is most clearly segmented so that all phrases were transactional type and relate to one cluster. The emphasis is on the search for well-convertible NC NK keywords.

How to make a great semantic kernel - step-by-step instructions

We switched to the main part of the article, where I will consistently, disassemble the main stages that you need to go to build a kernel of the future site.
In order for the process to be clearer, all the steps are given with examples.

Search for basic phrases

Working with the Code begins with the choice of the primary list of basic words and phrases (HF), which are best characterized by themes and are used in a broad sense. They are also called markers.

It can be like the names of the directions and types of products, popular requests from the topic. As a rule, they consist of 1-2 words and there are dozens, and sometimes hundreds of thousands of shows per month. Better keys are better not to take not to drown in minus words at the expansion stage.

More convenient to select marker phrases using. Looking into a request to it, in the left column we see the phrases, which he contains in itself, in right - similar requests from which you can often find suitable for expanding the theme. Also, the service shows the basic frequency of the phrase, that is, how many times they were asked for a month in all wordforms and with the addition of any words to it.

By itself, such frequency is not very interesting, so it is necessary to use operators to get more accurate values. We will analyze what it is and for what.

Operators Yandex Wordstat:

1) "..." - quotes. Quotes request allows you to track how many times in Yandex was looking for a phrase with all its wordforms, but without adding other words (tails).

2)! - Exclamation point. Using it before each word in the request, we fix it with its shape and get the number of appeals in the search by keyword only in the specified word form, but with the tail.

3) "! ...! ...! ..." - quotes and an exclamation mark before each word. Most an important operator For an optimizer. It allows you to understand how many times the keyword is requested a month strictly according to a given phrase, as it is written, without adding any words.

4) +. Yandex Vordstat does not take into account the prepositions and pronouns when requesting. If you need it to show them - put a plus sign in front of them.

five) -. The second most important operator. With it, the words that are not suitable are quickly eliminated. To apply it after the analyzed phrase, we put minus and stop the word. If they are somewhat repeated the procedure.

6) (... | ...). If you need to get from Yandex, the data on several phrases simultaneously conclude them in the brackets and share the direct slave. In practice, the method is rarely used.

For the convenience of working with the service, I recommend putting a special extension for the Wordstat Assistant browser. It is placed on Mozil, Google Chrome, I.Baurazer and allows you to copy phrases and their frequency with one click "+" or "Add All" icon.

Suppose we decided to make your blog on CEO. Select 7 main phrases for it:

• semantic kernel;
• optimization;
• copywriting;
• promotion;
• monetization;
• Direct

Search for synonyms

When prompted by searching systems, users can use the words close in meaning, but different by writing.

For example, "car" and "car".

It is important to find how more synonyms as possible to the main words to increase the coverage of the future semantic nucleus. If this is not done, then when Parcing, we will miss a whole layer of key phrases that reveal the needs of users.

What we use:

• Brainstorm;
• Right column Yandex Wordstat;
• Requests scored on Cyrillic;
• Special terms, abbreviations, slang expressions from the subject;
• Blocks Yandex and Google - along with the "query name" are looking for;
• Snippets of competitors.

As a result of all actions for the chosen topic, we obtain such a list of phrases:

Expanding the main requests

We will break these keywords to identify the basic needs of people in this area.
More convenient to do this in the Key Collector program, but if it is a pity to pay 1800 rubles per license, use it free analogue - Sword.

According to the functionality, it is of course weaker, but for small projects is suitable.
If you do not want to delve into the work of programs, you can use the Just-Magilities service and Rush Analytics. But still it is better to spend a little time and deal with software.

I will show the principle of working in Kay collector, but if you work with a word, then everything will be understood too. The program interface is similar.

Procedure:

1) We will add a list of basic phrases to the program and remove the basic and accurate frequency on them. If we plan to promote in a particular region - we indicate regionality. For information sites, it is most often not necessary.

2) Sparks left column Yandex Wordstat according to the added words to get all requests from our topic.

3) At the exit, we turned 3374 phrases. We will remove the exact frequency on them, as in the 1st point.

4) Check if there are no keys in the list with zero basic frequency.

If we remove and go to the next step.

Minus-words

Many neglect the procedure for collecting minus words, replacing it with the removal of phrases that are not suitable. But later you will realize that it is convenient and really saves time.

Open the data tab in Kay collector -\u003e Analysis. Select the grouping type according to individual words and leafs list the keys. If you see a phrase that does not fit - press the blue icon and add a word instead with all its wordware into stop words.

In Slobel, work with stoppal words is implemented in a more simplified version, but you can also make your list of phrases that do not fit and apply them to the list.

Do not forget to use the sorting of basic frequency and the number of phrases. This option helps to quickly reduce the list of source phrases or cut off rarely encountered.

After we have compiled a list of stop-words, apply them to our project and proceed to the collection of search prompses.

Parsing Tips

When entering a request to Yandex or Google, search engines offer their options to continue from the most popular phrases that are driven by the Internet. These keywords are called search tips.

Many of them do not fall into the Wordstat, so when building semantic you need to assemble such requests.

Kay collector, by default parsitis them with a bust of the end, Cyrillic and Latin alphabet and with a space after each phrase. If you are willing ready to sacrifice the amount to significantly speed up the process, put a tick at the point "Collect only top tips without a busting and a space after the phrase".

Often, among the search engines, you can find phrases with good frequency and competition in tens of times lower than in the vordstat, so in narrow niches I recommend collecting a maximum of words.

The tip parsing time directly depends on the number of simultaneous appeals to the search engine servers. Maximum kei collector supports 50 traffic.
But in order to parse requests in this mode, you will need as much proxy and accounts in Yandex.

For our project, after collecting prompts, it turned out 29595 unique phrases. By time, the whole process took a little more than 2 hours on 10 threads. That is, if there are 50, we will put in 25 minutes.

Definition of basic and accurate frequency for all phrases

For further work, it is important to determine the basic and accurate frequency and cut off all zero. Requests with a small number of shows are left if they are target.
This will help to better understand the intrinsic and create a more complete structure of the article than in the top.

In order to remove the frequency, first sear all the unnecessary:

• thrings of words
• keys with other symbols;
• dutch phrases (through the "Analysis of implicit oak" tool)

For the remaining phrases, we will define accurate and basic frequency.

but) for phrases up to 7 words:

• We choose through the filter "The phrase consists no more than 7 words"
• Open the window "Collection with yandex.direct" by clicking on the "D" icon;
• If you specify the region;
• Select the guaranteed display mode;
• We put the collection period - 1 month and tick above the desired types of frequency;
• Click "Get Data".

b) for phrases from 8 words:

• We specify for the column "phrase" filter - "consists of at a minimum of 8 words";
• If you need to move in a particular city below, we specify the region;
• Click on the magnifier and choose "collect all types of frequencies".

Cleaning keywords from garbage

After we received information about the number of impressions for our keys, you can start using non-suitable.

Consider the procedure for steps:

1. Go to the "analysis of groups" kei collector and we sort the keys by the amount of word use. The task to find the inadequate and frequent and put them in the list of stop words.
We do everything as well as in the "minus words".

2. Apply to the list of our phrases all the stop-words found and run on it to definitely not lose target queries. After checking, click Delete "marked phrases".

3. Source the phrases of dumbers, which are rarely used in accurate entry, but have a high base frequency. To do this, in the settings of the Key collector program in the key & serp, insert the calculation formula: Key 1 \u003d (yandexwordstatbasefreq) / (yandexwordstatquotepointfreq) and save changes.

4. We make KEY 1 calculation and remove those phrases that this parameter turned out from 100 and more.

The remaining keys should be grouped on boarding pages.

Clustering

Distribution of groups in groups begins with clustering phrases along the top free program "Majento Clusterizer". I recommend paid analogue with a wider functionality and fast speed Works - Keyasort, but also free enough for a small nucleus. The only nuance that for work in any of them will need to buy XML limits. The average price is 5 p. For 1000 requests. That is, the processing of the middle core by 20-30 thousand keys will cost 100-150 p. The address of the service that uses, look in the screenshot below.

The essence of clustering the keys this method is to combine into groups of those phrases that have on top 10 Yandex:

• common URLs with each other (Hard)
• with the most frequency query in the group (Soft).

Depending on the number of such coincidences for different sites, the clustering thresholds are distinguished: 2, 3, 4 ... 10.

The advantage of the method is the grouping of phrases according to the needs of people, and not only on synonymous connections. This allows you to immediately understand which keywords can be used on one boarding page.

For informationons will suit:

• Soft with threshold 3-4 and then cleaning with hands;
• Hard on 3-ke, and then combining clusters in meaning.

Online stores and commercial sites are usually advanced by Hard-y with a three-time clustering threshold, so I will raise it in a separate article.

For our project after grouping by the Hard method on 3rd it turned out 317 groups.

Competition check

It makes no sense to advance through highly competitive requests. It's difficult to get into the top, and without it there will be no traffic on the article. To understand what topics it is advantageous to write using the following method:

Weiently focus on the exact frequency of the phrase group by which the article is written and competition on the Mutagen. For information sites, I recommend, take the topic to work, in which the total accurate frequency of 300 and higher, and the competitive coefficient from 1 to 12 inclusive.

In commercial themes, focus on the marginality of goods or services and how competitors do in the TOP 10. Even 5-10 target queries per month can be a reason to make a separate page under it.

How to check compete on request:

a) manually, driven by the appropriate phrase in the service itself or through mass tasks;

b) in batch mode through the Kay collector program.

The selection of topics and grouping

Consider each of the resulting groups for our project after clustering and choose the topic for the site.
Majento Unlike Key Assort does not make it possible to download data on the number of shows for each phrase, so you will have to additionally remove them through Kay collector.

Instruction:

1) unload all groups from Majento in CSV format;
2) clutch the phrases in Excel on the "Group: Key" mask;
3) Load the resulting list in Key Sollector. In the settings, it must be a check mark in the import mode "Group: Key" and not to monitor the presence of phrases in other groups;

4) We remove the basic and accurate frequency for keywords from newly created groups. (If you use Key Assort, it is not necessary to do. The program allows you to work with additional columns)
5) We are looking for clusters with a unique intente, containing at least 3 phrases and the number of hits on all requests in the amount of more than 300. Next, we check 3-4 the most frequency of them on the Mutagen competition. If among these phrases there are keys with competition less than 12 - take into work;

6) We view the rest of the groups. If the phrases close by the meaning are found and it is worth viewing within one page - we combine. For groups containing new meanings, we look at the prospects for the total frequency of phrases, if it is less than 150 per month, then lay it up to the moment you go through the entire core. It may turn out to be combined with another cluster and gain 300 accurate shows - this is the minimum of which it is worth taking an article to work. To accelerate the manual group, use the auxiliary tools: fast filter and frequency dictionary. They will help you quickly find suitable phrases from other clusters;

Attention!!! How to understand that clusters can be combined? Take 2 frequency keys from those picked up in paragraph 5 for site Page and 1 request from the new group.
We add them to the Arsenkin tool "Unloading Top 10", indicate the desired region if necessary. Further, we look at the number of intersections in color for the 3rd phrase with the rest. We combine groups if they are from 3 or more. If there are no coincidences or one, it is impossible to combine - different intensis, in the case of 2 intersections, see the issuance by your hands and use logic.

7) After grouping keys, we receive a list of promising topics for articles and semantics under them.

Removing requests for another content type

In the preparation of the semantic nucleus it is important to understand that for blogs and information sites, commercial requests are not needed. Just as online stores do not need information.

We run through each group and clean everything is unnecessary if it is impossible to accurately determine the intent of the query - comparing the issuance or use the tools:

• Verification for commercialization from pixel Tuls (free, but with a day limit of checks);
• service Just-Magic, clustering with check mark Check commercially query (charge, cost depends on the tariff)

After that, go to the last step.

Optimization of phrases

We optimize the semantic kernel so that it is convenient to work with it in the future SEO specialist and copywriter. To do this, we will leave the key phrases in each group that mostly reflect the needs of people and contain as much synonyms as possible to the main phrases.

Algorithm of actions:

• Sort the keywords in Excel or Kay collector alphabetically from A to Z;
• Choose those that reveal the topic from different sides and different words. Other things being equal left the phrase with a higher accurate frequency or in which the Key 1 indicator (the ratio of the base frequency to the exact);
• We remove the keyword with the number of shows a month less than 7, which do not bear new meanings and do not contain unique synonyms.

An example of how a competently composed semantic kernel looks like -

I noted the phrases not suitable for intensity. If you neglect by my recommendations on manual grouping and not to check compatibility it will turn out that the page will be optimized for incompatible key phrases and high positions on progressive requests no longer see.

Final check list

1. We select the main high-frequency queries that set the subjects;
2. We are looking for synonyms for them using the left and right column Wordstat, the websites of competitors and their snippets;
3. Expand the received recovery requests to the left column Wordstat;
4. We prepare a list of stop-words and apply to the resulting phrases;
5. Parsim Tips Yandex and Google;
6. Remove the basic and accurate frequency;
7. Expand the list of minus words. We clean from garbage and rahes requests
8. We make clusterization through Majento or Keyassort. For information sites in Soft mode, threshold 3-4. For commercial Internet resources using the Hard method with threshold 3.
9. We import data in Kay collector and determine the competition 3-4 phrases for each cluster with a unique intensity;
10. We choose topics and determine with the landing pages for queries based on the assessment of the total number of accurate shows in all phrases from one cluster (from 300 for informationons) and competition for the most frequency of them on the Mutagen (up to 12).
11. For each suitable page, we are looking for other clusters with similar users' needs. If we can consider them on one page - we combine. When the need is not clear or there are suspicions that as an answer to it there should be another type of content or pages - check by issuing or through the tools pixel Tuls or Just-Magic. For content sites, the kernel should consist of information requests, Commercial from transactional. We delete too much.
12. We sort the keys in each group by alphabetically and leave those of them that describe the topic from different sides and different words. Other things being equal to the priority of those requests that are less than the ratio of the basic frequency to the exact and higher number of accurate shows per month.

What to do with SEO-core after its creation

As a list of keys, they gave them to the author and he wrote an excellent article completely, disclosing all the meanings. Eh, something I am stuck ... The sensible text will be possible only if the copywriter clearly understands that you want from him and how to check yourself.

We will analyze 4 components, well-work, which you are guaranteed to get a lot of target traffic on the article:

Good structure. We analyze requests selected for the landing page and reveal what needs people have in this topic. Next, write a plan for an article that fully responds to them. The task is to make people going to the site received a volumetric and comprehensive response on the semantics you made up. It will give good behavioral and high relevance to the intensity. After you have compiled a plan, see the competitors' sites by Having a basic promotable search request. You need to do in such a sequence. That is, first we do yourself, then we look at others and if we need to be refined.

Optimization under keys. The article itself is sharpening under 1-2 most frequency keys with competition on Mutagen up to 12. Another 2-3 mid-frequency phrases can be used as headlines, but in a diluted form, that is, inserting additional non-theme into them, using synonyms and word . The main focus we make on low-frequency phrases of which pulls out the unique part - the tail and evenly implement into the text. Search engines themselves will find and glue.

Synonyms for basic requests. We write them down separately from our semantic kernel and put the task of the copywriter to use them evenly in the text. This will help reduce the density of our main words and the text will be optimized enough to get into the top.

Theatic-specifying phrases.By themselves, LSI does not promote the page, but their presence indicates that most likely the written text belongs to Peru expert, and this is already a plus to the quality of content. To search for thematic phrases, use the Tools tool for a copywriter from Pixel Tuls.

Alternative method of selection of key phrases using competitor analysis services

There is a quick approach to creating a semantic kernel, which is applicable both for beginners and for experienced users. The essence of the method is that we initially choose the keys not for the entire site or category, but specifically under the article, the seating page.

It can be realized by 2 ways that differ in the way we choose the themes for the page and how deeply expand the key phrases:

• using the main key parsing;
• based on the analysis of competitors.

Each of them can be implemented at a simple and more complex level. We will analyze all the options.

Without the use of programs

Copywriter or webmaster often do not want to deal with the large number of programs, but needed good topics and key phrases under them.
This method is just for beginners and those who do not want to bother. All actions are made without the use of additional software, using simple and understandable services.

What will take:

• Keys.So service for competitors analysis - 1500 p. In the promotion "Altblog" - 15% discount;
• Mutagen. Check for query competition - 30 kopecks, collecting basic and accurate frequency - 2 kopecks for 1 check;
• Buvillers - a free version or business account - 995 p. (now with a discount 695 p)

Option 1. Select the topic through the main phrase parcel:

1. We select the main keys from the subject in the broad sense using the brainstorming and the left and right column of Yandex Wordstat;
2. Further, we are looking for synonyms for them, the methods of which were said early;
3. We score all the received marker requests to the bookvilk (need to pay for the paid rate) in the advanced mode "Search by list of keywords";
4. We specify in the filter: "! Accurate! Frequency" from 50, the number of words from 3;
5. Unload the entire list in Excel;
6. We allocate all the keywords and send to the grouping to the "Clastier fists" service. If the site is regional chosen needed city. Clustering threshold for information sites Leave on 2, for commercial 3-ku;
7. After grouping, choose themes for articles by browsing the resulting clusters. We take those where the amount of phrases from 3 and with a unique intente. It is better to understand the needs of people helping the analysis of URLs of sites from Top in the column "Competitors" (on the right in the Kulakov service plate). We also do not forget to check the competition for Mutagen. We pierce 2-3 requests from the cluster. If more than 12, then you should not take the topic;
8. The name of the future landing page was determined, it remains to choose key phrases for it;
9. From the field "Competitors" copy 3 ullas with a suitable type of pages (if the site is information - take references to articles, if commercial, then on stores);
10. Insert them sequentially in Keys.so and unload all the key phrases on them;
11. We combine them in Excel and remove duplicate;
12. Data only service is not enough, so it is necessary to expand them. We will use the buckvilical again;
13. The resulting list is sent to clustering in the "Customerizer" fists;
14. We select the groups of requests that are suitable for the landing page, focusing on the intent;
15. We remove the basic and accurate frequency through the mutagen in the "Mass tasks" mode;
16. Upload a list with updated data by the number of shows in Excel. Remove the zero for both types of frequencies;
17. Also in Excel, add the formula for the relationship of basic frequency to accurate and leave only those keys in which it is less than 100;
18. We delete the requests of another type of content;
19. We leave the phrases that the most fully and different words reveal the main intent;
20. We repeat all the same actions on paragraphs 8-19 for the rest.

Option 2. Select the topic through the analysis of competitors:

1. We are looking for top sites in our subject, driven by RF requests and viewing the issuance through the tool Arsenkin "Top 10". It is enough to find 1-2 suitable resources.
If we promote the site in a particular city, we indicate regionality;
2. Go to the KEYS.So service and enter the URLs of the sites that found and look at what kind of competitors' pages bring the most traffic.
3. 3-5 of the most accurate frequency requests from them check for competitiveness. If it is above 12 in all phrases, then it is better to look for another topic less competitive.
4. If you need to find more analysis sites, open the "Competitors" tab and set the parameters: similar to 3, collaboration - 10. We assort the data on the decrease of traffic.
5. After we have chosen the topic, drive its name to the issuance and copy 3 ullas from the top.
6. Next, repeat items 10-19 from the 1st option.

Using a collector kei or word

This method will differ from the previously used to use for some operations program Key collector and a deeper key expansion.

What will take:

• key collector program - 1800 rubles;
• all the same services as in the previous way.

"Advanced - 1"

1. Parsim left and right column of Yandex throughout the list of phrases;
2. Remove the exact and basic frequency through Kay collector;
3. Calculate the key 1;
4. We delete the zero requests and with Key 1\u003e 100;
5. Next, we do everything as well as in paragraphs 18-19 of option 1.

"Advanced - 2"

1. We make steps 1-5, as in Embodiment 2;
2. We collect on each URL-in keys in Keys.so;
3. We remove the duplicate in Kay collector;
4. We repeat the paragraphs 1-4, as in the "Advanced -1" method.

Now compare the number of keys obtained and their accurate total frequency when collecting with different methods:

As we see from the table, the best result showed an alternative method for creating a core under the page - "advanced 1.2". It was possible to get 34% more target keys and the total traffic on the cluster turned 51% more than in the case of a classic method.

Below on the screenshots, it can be seen how the finished core looks like, in each of the cases. I took phrases with a precise number of deposits from 7 per month so that the quality of the keyword could be estimated. For complete semantics, see the table on the "View" link.

BUT)


B)


IN)

Now you know that the most common way is not always done, as everyone is doing, the most faithful and correct, but not to refuse other methods. Much depends on the theme itself. For commercial sites, where the keys are not so much, quite enough and the classic option. On information sites, you can also get excellent results if you correctly compile a Copyracy TK correctly, make a good structure and SEO optimization. We will talk about all this in detail in the following articles.

3 Frequent errors when creating a semantic kernel

1. Collect phrases by top. It is not enough to resort the Wordstat to get a good result!
More than 70% of requests that are brought by rarely or periods, they do not get there. But among them often there are key phrases with good conversion and really low competition. How not to miss them? Be sure to collect search prompts and combine them with data from different sources (, counters on sites, statistical and database services).

2. Mixing information and commercial requests on one page.We have already disassembled that key phrases differ in the type of needs. If a visitor comes to your site, who wants to make a purchase, and sees as an answer to his request page with an article, how do you think it will be satisfied? Not! Searching systems also think when the page is ranked, which means about the top in the middle and RF phrases you can immediately forget. Therefore, if you doubt the definition of the type of request, see the issuance or use the Pixel Tuls tools, Just-Magilities to determine commercialism.

3. Selection to promote very competitive requests. Positions for RF VK phrases are 60-70% dependent on behavioral factors, and to get them to get into the top. The more applicants, the longest the queue from those who want and above the requirements for sites. Everything, as in life or sports. It is much more difficult to become a world champion than getting the title in your city.
Therefore, it is better to enter a quiet, and not overheated niche.

Previously, it was even more complicated to be on top. In the top stood on the principle who managed, he ate. The leaders fell to the first places, and they could only move them, accumulating behavioral factors. And how to get them if you are on the second or third page ... Yandex ripped this closed circle in the summer of 2015, introducing a multiple gangster algorithm. The essence of it is just that randomically increase and lower the positions of sites to understand whether more worthy candidates appeared to find in the top.

How much money is needed for start?

To answer this question, we consider the costs of the necessary arsenal of programs and services to prepare and unknow key phrases per 100 articles.

The minimum (suitable for the classic option):

1. Slobel - Free
2. Majento clusterizer - free
3. To recognize the Captch - 30 rubles.
4. XML limits - 70 rubles.
5. Check for Competition of the Mutagen request - 10 checks per day for free
6. If you are not treated anywhere and ready to spend on a parsing 20-30 hours you can do without a proxy.
—————————
The result is 100 rubles. If you enter captchas yourself, and XML limits get in exchange for transferred from your site, then really prepare the kernel in general for free. It is only necessary to spend the day another on setting up and mastering the programs and another 3-4 days to expect the results of the parsing.

Standard semantist set (for advanced and classic method):

1. Kay collector - 1900 rubles
2. Kay Assort - 1700 rubles
3. Bookwarix (Business Account) - 650 rubles.
4. KEYS.SO competitor analysis service - 1500 rubles.
5. 5 proxy - 350 rubles per month
6. Antikapitch - approximately 30 rubles.
7. XML limits - about 80 rubles.
8. Checking Competition by Mutagen (1 Check \u003d 30 kopecks) - Make 200 rubles.
———————-
The result is 6410 rubles. Of course, it is possible to do without Keyassort, replacing it with a Majento clusterizer and instead of Kay Collector to use a word. Then enough and 2810 rubles.

Should I trust the development of the "Profi" kernel or better to figure it out and do it yourself?

If a person is regularly engaged in his favorite thing, pumping out in it, then following logic, its results should be exactly better than the newcomer in this area. But with the selection of key, everything turns out exactly the opposite.

Why in 90% of cases a newcomer makes better professional?

It's all about the way. The task of the semantist does not collect the best kernel for you, but to fulfill your work for the minimum time and that its quality arranged to you.

If you do everything yourself for those algorithms about which they said early - the result will be an order of magnitude higher for two reasons:

• You understand the topic. So, you know the needs of your customers or users of the site and you can at the initial stage to expand the marker requests for the parsing, using a large number of synonyms and specific words.
• Interested to do everything qualitatively. Business owner or employee of the company in which it works, of course, comes to the question more responsibly and will try to do everything to the maximum. The more complete the kernel and more low-competitive requests in it, the more it will be possible to assemble the target traffic, and therefore profit with the same investments in the content will be higher.

How to find the remaining 10% that make the kernel better than you?

Look for companies in which the selection of key phrases is key competence. And immediately discuss what the result you want, like everyone else or maximum. In the second case, it will be 2-3 times more expensive, but in the long run it repeatedly pays off. For those who want to order a service with me, all the necessary information and conditions. Quality guarantee!

Why it is so important to fully work out semantics

Here, as in any sphere, the principle of "good and bad selection" works. What is his essence?
Every day we are faced with the fact that we choose:

• meaning with a person who seems to be nothing, but does not cling or disappointing in itself to build harmonious relationships with those who need;
• work that does not like or find work, then what the soul is lying and do it with your profession;
• rent a room for the store in a non-checkpoint or still wait until you free, the appropriate option;
• take into the team not the best sales manager, and one who bestly showed himself at today's interview.

It seems everything is clear. And if you look at it on the other hand, presenting every choice as an investment in the future. Here it begins the most interesting!

Saved on this. kernel, 3-5 thousand. Satisfied like elephants! But what it leads to:

a) for information sites:

• Losses on traffic at least 1.5 times with the same investments in the content. Comparing miscellaneous methods obtaining key phrases, we have already found out the experienced way that an alternative method allows you to collect 51% more;
• The project quickly sends in extradition. Competitors are easy to get around us, giving a more complete answer on the intensity.

b) for commercial projects:

• Less leads or increase their cost. If we have semantics, like everyone else, we are moving on the same requests as competitors. A large number of suggestions with constant demand reduces the share of each of them in the market;
• Low conversion. Specific requests are better converted into sales. Save to seven. kernel, we lose the most conversion keys;
• Heavier to advance. Many wishing to be in the top - above the requirements for each of the candidates.

I wish you always do a good choice And invest only in plus!

P.S. Bonus "how to write good article With bad semantics, "as well as other life-to-job life and earnings on the Internet, read in my group

Greetings to you dear my readers!

I am sure that many of you will not be heard, but not even guess that there is such a thing as a semantic kernel! And what do you ask this? - I will try to explain to you simple words. The semantic core is a set of keywords, phrases, simple suggestions, phrases, etc., which gives a search engine (hereinafter referred to as PS) when you enter your request in the browser string.

What is the semantic core? The semantic core of the site is the basis of promotion and promotion, it is necessary for internal optimization. Without semantic kernel, the promotion of your project (site) will not be effective. The more competently composed the semantic core of the site, the less money you need for its successful promotion. Nothing is not clear yet, so? Do not be afraid, then I will try to decompose everything in detail all around the shelves. Read carefully and understand everything!

How to make a semantic kernel!

The first thing you need to do after you have decided on the topics of the blog, make a semantic kernel. To do this, take notebook and pen and write down all words, phrases, suggestions that characterize the subject of your blog. Each word, phrase or offer is, in fact, there will be future headlines for your posts, and the more words you think, the future you will have more choice when writing articles.

And to make a sufficiently solid list (200-300 words) you will leave a lot of time. Therefore, for convenience, we will use special services, such as Yandex Wordstat., Google AdWords, Rambler Adstat they will significantly simplify the task when. Of course, it would be possible to do only Yandex and Google, because These are giants in finding key queries compared to Rambler, but statistics says that 5-6% of people are still used as a search engine Rambler, so we will not neglect it.

What would you have much easier to learn the material, I will show everything on specific examples. Agree, the theory is good, but when it comes to practice, many problems begin. Therefore, we will be together a semantic core that in the future you can easily move the knowledge gained and experience on the subject of your blog. Suppose the topic of your blog "Photoshop » And all that is connected with it. Therefore, as it was written above, you must come up with and write down as many words, phrases, phrases as possible, phrases as possible, as you like to call them. These are the words that I have characterized with the topics of the blog about Photoshop. Of course, I will not list the entire list of words, but only a part that you can understand the very meaning of the preparation of the semantic kernel:

brushes for photoshop
Brushes for photoshop.
Photoshop brushes
Photoshop brushes
Effects Photoshop
Photographyhop effects
Photoeffect
Pictures Photoshop
Pictures photoshop.
collage
Photocollazh
photomontage
Frames for photos
Photo design

The list is compiled. Well, let's start. Immediately make a reservation, your list can be much different from mine and should be much more. I compiled this list of words for clarity, whatever you caught the essence of how to make a semantic core.

Yandex Wordstat keyword statistics

After your list is formed, it is necessary to cut off all the words you do not need, for which we definitely do not promote our blog. For example, I will not move according to such words as (brushes for photoshop torrent, photoshop brushes Makeup), these phrases are not clear for me at all, just sewing similar phrases such as (photoshop brushes for free and free photoshop brushes). I think the meaning of the selection of keywords is understood.

Then you see that Yandex Wordstat has two columns. The column on the left shows you that they were looking for people to drive, in the search string, in our case the phrase "Brushes for Photoshop". The right column shows that they still searched for people who were looking for a phrase "Brushes for Photoshop", advise you not to ignore the right column, and choose all words from it that are suitable for your subject.

Well, they also figured out, go further. Another very much important momentAs you see by the search results for "Photoshop brushes", we see a huge number of 61134 requests! But this does not mean that the phrase "Brushes for Photoshop" was hampered as many times in a month in the search string of Yandex. Yandex Wordstat is designed in such a way that if you admire the phrase "Brushes for Photoshop", he will give you the number of requests that will mean how many times people looking for any word form (Photoshop brushes, photoshop brushes but, cyst b Photoshop I.T.D.), phrases (offers) (free brushes for photoshop, download photoshop brushes for free, etc.), in which there is a phrase "Brush for Photoshop". I think it is also clear.

In order for Yandex Wordstat to give us an accurate number of requests, there are special operators, such as ("", "!"). If you enter the phrase "Brushes for Photoshop" in quotes, you will see a completely different digit that shows you how many times people looking for the phrase "brushes for photoshop" in different flows (Photoshop brushes but etc.).

When entering the phrase "! Brushes! For! Photoshop" in quotes and with an exclamation mark We will get the exact number of requests "! Brushes! For! Photoshop" in this form in which it is, i.e. Without all sorts of decline, word forms and phrases. I think you understand the meaning, chewmed as I could.

So, after you have formed an impressive list in Excel, you need to apply to each word (phrase) Operator "!". When you do everything, you will have a list with an accurate number of requests / a month to be corrected again.

But this is a little later, after we consider the two other selection systems for the selection of keywords (Google AdWords and Rambler Adstat). Since after their consideration your list of keywords will be significantly replenished.

Selection of Google AdWords Keywords

To select keywords, you also use Google AdWords, this is a similar service with Yandex Wordstat. Let's just go here. We will open the key selection window google words AdWords. We also drive the first phrase from our list "Brush for Photoshop" in the search bar. Please note that there are no operators in Google AdWords, but it is enough to put a tick opposite the words [accurate] in the "Compliance Types" column. As we see, the number of requests / a month in Google Adwords is significantly different from Yandex Wordstat. This suggests that any more people use the search system Yandex. But if you see the entire list, you can find those keywords that Yandex Wordstat does not show at all.

Also in Google AdWords you can find out a lot of interesting things (for example, approximate price per click), which should also be considered when selecting keywords. The more price per click is the higher the competitive request. I will not stop in detail here, the principle of selection of keywords is similar to Yandex Wordstat and speaking a little, you yourself can disassemble with everything. Go ahead.

Statistics for search queries Rambler Adstat

As I have already mentioned above that the Rambler Adstat is much inferior to two previous services, but still it is also possible to learn some information. Let's just go here and just introduce the first phrase in the search bar from our list "Brushes for Photoshop". Here I think it should not stop in detail. I repeat once again the principle of selection of keywords in all three systems similar.

We have become familiar with the three keyword seats. As a result, you have a huge list, formed from all three services, in which you have already selected for those requests that you do not plan to advance and duplicate duplicate. I have already written about it above. But it is only halfway in the preparation of the semantic kernel. You probably have already boiled brains, but in fact, if you get into it and figure out, there is nothing complicated here. Believe me, it is better to correctly make a semantic kernel once than in the future it will have to correct everything. And it is much more complicated to correct than to do everything from scratch. Therefore, be patient and go further.

RF, SC and LF Requests or VFS, CX and NVT

In the preparation of the semantic nucleus, there are still such concepts as high-frequency, mid-frequency and low-frequency queries, or they are also called RF, SC and LF requests, and the VFS, CID and NMCs may also be found. These are the requests that people are injected into search engines. The more people enter the same request in the search string, the high frequency will be a query (RF request), the same with the SC and LF requests. I hope it is also clear.

Now remember one very important point. At the initial stage of blog development, it should be promoted only on the LF requests, sometimes uses querys, it will depend on the query competitiveness. On RF requests, you are unlikely to be able, you just do not have enough money for it. Do not be afraid of the LF requests to go to the top possible and without investing. You most likely there was a question, and what requests to consider RF requests, SC requests and LF requests?

Here the exact answer I think can not give anyone! Blogs have different topics it will be different. There are very popular topics, in which the exact number of requests ("!") Reaches 20 thousand shows / per month or more (for example, "! Lessons! Photoshop"), and there are less popular in which the exact number of requests does not reach and 2000 Views / per month (for example "! Lessons! English").

In this case, I adhere to a simple formula that calculated for myself, we will demonstrate on the example "! Lessons! Photoshop":

VK, SC and NK Requests for EKS, SCZ and NKZ

Following the RF, SC and LF requests, there is another category. These are highly competitive (VC), medium competitive (SC) and low competitive (NK) requests, can also meet ACS, SCZ and NKZ. In this case, we will need to determine the competitiveness of those requests by which we are planning to move into the top, but this will be a separate post, on the topic ". . In the meantime, let's assume that the RF requests are VC inquiries, SC - SC and NC - NK. In most cases, this formula works, but there are exceptions when, for example, LF requests are highly competitive (VC) and, on the contrary, RF requests are NK. It all depends on the topics of the blog.

Scheme of the preparation of the semantic kernel

For clarity, let's look at a schematic example of the semantic kernel. Approximately it should look like standard scheme Semantic kernel.

But it is not necessary to get very tied to this scheme, because As blog is maintained, it can change. At the initial stage, you may have, let's say only four rubrics containing three LC requests, and over time it can change everything.

But nothing will understand most of you, especially who is first faced with the semantic core. Nothing terrible, I also did not understand many things at first until I studied the topic very well. I do not want to say that I am a pro in this topic, but a lot has learned much. And, as promised, let's disassemble everything on a specific example and according to our topic.

Immediately I want to say that I am not a specialist in "Photoshop", just this topic came to my mind when writing a post. Therefore, I chose requests in meaning. Well, this is the scheme of the semantic kernel I turned on the topic "Photoshop". You should get something on likeness:

Types of requests

All requests (our keywords) can be divided into three categories:

• Primary requests - These are the requests that one or two words can give a general definition of your resource or part of it. Primary requests, the most covering the overall theme of your blog better leave on the main page. In our case, this is: Photoshop lessons, the effects of photoshop, how to make photo collage.
  Primary requests, less covering the overall topics of your blog, but most accurately characterizing it some part, it is recommended to use your blog headquarters as separate headings. In our case, this is: Photoshop brushes, photoshop frame, photoshop patterns, photo design.

• Main requests - These are the requests that definitely determine the theme of your project and are able to give useful information Reader, teach him what he wants, or answer the frequently asked question as ??? Those., In our case, this is: how to add brishes in Photoshop, how to make a pattern in Photoshop, how to make a photo collage in photoshop, etc. The main requests, in fact, should be the headlines of our future articles.

• Additional (auxiliary) requests or them are also called associative - These are the requests that people have just entered into the browser search bar when searching for a basic query. Those. These are key phrases that are part of the main request. They seem to complement the main request, and be keywords when moving it into the top. For example: Photoshop for beginners online, Photoshop remove the effect of red eyes, collage of several photos. I think it is understandable.

Strategy for the preparation of the semantic nucleus

Now we need to split the entire list on the pages. Those. need from all your keywords to choose the primary requests that will be the headings of your blog and make separate tabs in Excel. Further, choose these basic and auxiliary requests and accommodate them different pages in the you created excel document (i.e. by headings). That's what I did:

As I wrote above: at the initial stage it is worth promoting your blog on LF or NC requests. And what to do with SC (SC) and HF (VC) requests, you ask? I explain.

It is hardly possible to move to RF (VC) requests, so you can delete them, but it is recommended to leave one or two RF (VC) request for the main page. Immediately make a reservation: it is not necessary to rush to the most popular request, such as "Photoshop", which has an exact number of hits / per month is - 163384. For example, you want using your blog to teach people to work in Photoshop. So take as the basis of the RF request - "Photoshop lessons", which has an exact number of hits / per month is - 7110. This request more characterizes your topic and it will be easier for it to advance.

But the SC (SC) requests can be placed on a separate page in Excel. As your blog will rise in the eyes of the PS (SC (SC) requests) will be gradually becoming in demand.

I know that newcomers now do not understand at all, what I say about, I advise you to read the article about being studied by which everything will be understood.

Conclusion

Here in principle, probably, that's all. Of course, there are programs that will help you in compiling semantic kernel As paid (key kollektor) and free (Swords, Lodel), but I will not write about them in this post. Perhaps somehow I will write about them a separate article. But they will only pick up the keywords, and you will have to dissemble them along the headings and you will have to do it yourself.

And how do you make a semantic kernel? Or maybe you do not make it at all? What are your programs and services when drawing up? I will be happy to hear your answers in the comments!

And finally look at an interesting video.

03.08.2018 Time reading: 5 minutes

What is a semantic core?

Semantic kernel is a set search phrases And the words in which the site promotion is carried out. These search words and phrases help robots to determine the topics of the page or the entire service, that is, to find out what the company does.

In Russian, the semantic is called the section of science on language engaged in the study of the semantic filling of lexical units of the language. Applied to search engine optimization This means that the semantic kernel is a semantic content of the resource. It helps to determine what information to convey to users and in what vehicle. Therefore, semantics is the foundation, the basis of all SEO.

What is the semantic site kernel and how to use it?

• The correct semantic kernel is necessary to accurately calculate the cost of promotion.
• Semantics is a vector of building internal SEO-optimization: the most relevant requests are selected for each service or product so that users and search robots are better found.
• It is based on the structure of the site and texts for thematic pages.
• The keys from semantics are used to write snippets ( brief descriptions pages).

Here is a semantic kernel - an example of its compilation in the company website for the site of the construction company:

The optimizer collects semantics, disassembles it in logical blocs, finds out the number of their shows and based on the cost of requests in the Yandex and Google top expects the total cost of promotion.

Of course, during the selection of the semantic kernel, the specifics of the company's work is taken into account: for example, if the company did not design and did not build houses from the glued bar, then we would be deleted and did not use the corresponding requests. Therefore, the mandatory stage of working with semantics is the coordination of it with the customer: no one knows the features of the company's work.

Types of keywords

There are several parameters for which are classified. key requests.

1. In frequency
   • high-frequency - words and phrases with a frequency from 1000 shows per month;
   • mid-frequency - up to 1000 shows per month;
   • low-frequency - up to 100 shows.

Collection of frequency in keywords Helps find out what users most often are asked. But a high-frequency request is an optional request with high competitiveness, and the preparation of semantics with high frequency and low competitiveness is one of the main aspects in working with the semantic core.

2. Type:
   • geospendent and non-independent - attached to the region promotion and unacted;
   • information - the user receives some information on them. The keys of this type are commonly used in articles - for example, reviews or useful advice;
   • brand - contain the name of the progressable brand;
   • transactional - imply action from the user (buy, download, order) and so on.
3. Other species are those who are difficult to attribute to any type: let's say the "profiled timber" key. Driving such a request in the search engine, the user may imply anything: buying a bar, properties, comparisons with other materials and so on.

   From the experience of our company, we can say that in such requests to promote any site is very difficult - as a rule, it is high-frequency and highly competitive, and this is not only difficult in optimization, but also expensive for the client.

How to collect a semantic kernel for the site?

• By analyzing competitors' sites (in Semrush, Serpstat, you can see the semantic core of competitors):

The process of compiling the semantic kernel

The collected queries are not yet a semantic kernel, then it is necessary to separate the grain from the spurred, so that all requests are relevant to the client's services.

To make a semantic kernel, queries need to be clustered (split into blocks on the logic of the provision of services). You can do this using programs (for example, Keyassort or Topsite) - especially if the semantics is volumetric. Or manually evaluate and sort out the entire list, delete unsuitable requests.

Then send the client and clarify whether there are errors.

The finished semantic kernel is a yellow brick track to the content plan, to the articles in the blog, texts for the goods cards, the company's news, and so on. This is a table with the needs of the audience that you can satisfy using your site.

• Distribute keys via pages.
• Use key requests in metales