Instructions for steps to compile a semantic core. A simple example of compiling a semantic core Query core

If you have a question “How to compose a semantic core,” then before deciding, you must first figure out what you are dealing with.

Semantic core of the site is a list of phrases that users enter into search engines. Accordingly, promoted pages must respond to user queries. Of course, you can’t shove a bunch of different types of key phrases onto the same page. One main search query = one page.

It is important that the keywords correspond to the theme of the site, do not have grammatical errors, have a reasonable frequency, and also correspond to a number of other characteristics.

The semantic core is usually stored in an Excel table. This table can be stored/created anywhere - on a flash drive, in Google Docs, on Yandex.Disk or somewhere else.

Here is a clear example of the simplest design:

Features of selecting the semantic core of a site

First, you need to understand (at least roughly) what phrases your audience uses when working with a search engine. This will be quite enough for working with tools for selecting key phrases.

What keywords does the audience use?

Keys- these are precisely the same phrases that users enter into search engines to obtain this or that information. For example, if a person wants to buy a refrigerator, he writes “buy a refrigerator,” or “buy an inexpensive refrigerator,” “buy a Samsung refrigerator,” etc., depending on his preferences.

Now let's look at the characteristics by which keys can be classified.

Sign 1 - popularity. Here the keys can be roughly divided into high-frequency, mid-frequency and low-frequency.

Low-frequency queries (sometimes referred to as LF) have a frequency of up to 100 impressions per month, mid-frequency (MF) - up to 1000, and high-frequency (HF) - from 1000.

However, these figures are purely conditional, because there are many exceptions to this rule. For example, the topic of cryptocurrency. Here it is much more correct to consider low-frequency queries with a frequency of up to 10,000 impressions per month, medium-frequency - from 10 to 100 thousand, and high-frequency - everything else. Today, the keyword “cryptocurrency” has a frequency of more than 1.5 million impressions per month, and “bitcoin” has exceeded 3 million.

And despite the fact that “cryptocurrency” and “bitcoin”, at first glance, are very tasty search queries, it is much more correct (at least in the initial stages) to focus on low-frequency queries. Firstly, because these are more precise queries, which means it will be easier to prepare relevant content. Secondly, there are ALWAYS tens to hundreds of times more low-frequency queries than high-frequency and mid-frequency queries (and in 99.5% of cases, also combined). Thirdly, the “low-frequency core” is much easier and faster to expand than the other two. BUT... This does not mean that the mids and highs should be ignored.

Sign 2 - user needs. Here we can roughly divide into 3 groups:

  • transactional - imply some kind of action (contain the words “buy”, “download”, “order”, “delivery”, etc.)
  • informational - simply searching for certain information (“what will happen if”, “what is better”, “how to do it correctly”, “how to do it”, “description”, “characteristics”, etc.)
  • others. This is a special category because... it is not clear what exactly the user wants. For example, let's take the request “cake”. "Cake" what? Buy? Order? Bake according to the recipe? View photos? Unclear.

Now about the application of the second sign.

Firstly, it is better not to “mix” these requests. For example, we have 3 search queries - “dell 5565 amd a10 8 gb hd laptop buy”, “dell 5565 amd a10 8 gb hd laptop review” and “dell 5565 amd a10 8 gb hd laptop”. The keys are almost completely identical. However, it is the differences that play a decisive role. In the first case, we have a “transactional” request, according to which we need to promote the product card. In the second - “information”, and in the third - “other”. And if a separate page is needed for the information key, then it is logical to ask the question - what to do with the third key? It’s very simple - view the TOP 10 of Yandex and Google for this query. If there are many trade offers, then the request is still commercial, and if not, then it is informational.

Secondly, transactional requests can also be divided into “commercial” and “non-commercial”. In commercial requests you will have to compete with “heavyweights”. For example, for the request “buy samsung galaxy” you will have to compete with Euroset, Svyaznoy, for the request “buy an ariston oven” - with M.Video and Eldorado. So what should I do? It’s very simple to “swing” at requests that have a much lower frequency. For example, today the request “buy samsung galaxy” has a frequency of about 200,000 impressions per month, while “buy samsung galaxy a8” (and this is a very specific model of the line) has a frequency of 3,600 impressions per month. The difference in frequency is enormous, but for the second request (precisely due to the fact that a very specific model is implied) you can get much more traffic than for the first.

Anatomy of search queries

The key phrase can be divided into 3 parts - body, qualifier, tail.

For clarity, let’s take the previously discussed “other” query - “cake”. What the user wants is unclear, because... it consists only of a body and has no specifier and tail. However, it is high-frequency, which means it has fierce competition in search results. However, 99.9% of people who visit a site will say “no, this is not what I was looking for” and simply leave, and this is a negative behavioral factor.

Let’s add the “buy” specifier and get a transactional (and as a bonus, also a commercial) request “buy a cake.” The word “buy” reflects the user’s intent.

Let’s change the specifier to “photo” and get the request “cake photo”, which is no longer transactional, because the user is simply looking for photos of cakes and is not going to buy anything.

Those. It is with the help of the specifier that we determine what kind of request it is - transactional, informational or other.

We've sorted out the sale of cakes. Now, to the request “buy a cake” we will add the phrase “for a wedding”, which will be the “tail” of the request. It is the “tails” that make requests more specific, more detailed, but at the same time do not cancel the user’s intentions. In this case, since the cake is a wedding, then cakes with the inscription “happy birthday” are immediately discarded, because... they are not suitable by definition.

Those. if we take the queries:

  • buy a birthday cake
  • buy a wedding cake
  • buy an anniversary cake

then we will see that the user’s goal is the same - “buy a cake”, and “for the birth of a child”, “for a wedding” and “for an anniversary” reflect the need in more detail.

Now that you know the anatomy of search queries, you can derive a certain formula for selecting a semantic core. First, you define some basic terms that are directly related to your activity, and then collect the most suitable specifiers and tails (we’ll tell you a little later).

Clustering of the semantic core

Clustering refers to the distribution of previously collected requests across pages (even if the pages have not yet been created). This process is often called “grouping the semantic core.”

And here many people make the same mistake - they need to separate queries according to their meaning, and not according to the number of pages available on the site or in a section. Pages can always be created if necessary.

Now let's figure out which keys should be distributed where. Let's do this using the example of a structure that already has several sections and groups:

  1. Home page. For it, only the most important, competitive and high-frequency queries are selected, which are the basis for promoting the site as a whole. (“beauty salon in St. Petersburg”).
  2. Categories of services/products. It is quite logical to place queries here that do not contain any particular specifics. In the case of a “beauty salon in St. Petersburg”, it is quite logical to create several categories using the keys “make-up artist services”, “men’s room”, “women’s room”, etc.
  3. Services/products. More specific queries should already appear here - “wedding hairstyles”, “manicure”, “evening hairstyles”, “coloring”, etc. To some extent, these are “categories within a category.”
  4. Blog. Information requests are suitable here. There are many more of them than transactional ones, so there should be more pages that will be relevant to them.
  5. News. Keys that are most suitable for creating short news notes are highlighted here.

How query clustering is done

There are 2 main methods of clustering - manual and automatic.

Manual clustering has 2 main disadvantages: long, labor-intensive. However, the entire process is controlled personally by you, which means you can achieve very high quality. For manual clustering, Excel, Google Sheets or Yandex.Disk will be quite sufficient. The main thing is to be able to filter and sort data according to certain parameters.

Many people use the Keyword Assistant service for clustering. Essentially, this is manual clustering with elements of automation.

Now let’s look at the pros and cons of automatic grouping; fortunately, there are many services (both free and paid) and there is plenty to choose from.

For example, the free clustering service from the SEOintellect team is worthy of attention. It is suitable for working with small semantic cores.

For “serious” volumes (several thousand keys), it makes sense to use paid services (for example, Topvisor, SerpStat and Rush Analytics). They work as follows: You load key queries, and at the end you receive a ready-made Excel file. The 3 services mentioned above work approximately according to the same scheme - they group by meaning, analyzing the intersection of phrases, and also look at the TOP-30 search results for each request to find out how many URLs the requested phrase appears on. Based on the above, distribution into groups occurs. All this happens “in the background.”

Programs for creating a semantic core

To collect relevant search queries, there are many paid and free tools to choose from.

Let's start with the free ones.

Service wordstat.yandex.ru. This is a free service. For convenience, it is recommended to install the Wordstat Assistant plugin in your browser. That is why we will consider these 2 tools in pairs.

How it works?

Very simple.

For example, we will put together a small core of travel packages to Antalya. As “basic” we will have the 2nd request - “tours to Antalya” (in this case, the number of “basic” requests is not important).

Now go to https://wordstat.yandex.ru/, log in, insert the first “basic” request and get a list of keys. Then, using the plus signs, we add suitable keys to the list. Please note that if a key phrase is colored blue and marked with a plus on the left, it means it can be added to the list. If the phrase is “discolored” and marked with a minus, it means it has already been added to the list, and clicking on the “minus” will lead to its removal from the list. By the way, the list of keys on the left and the pros and cons are the very features of the Wordstat Assistant plugin, without which working in Yandex.Wordstat makes no sense at all.

It is also worth noting that the list will be saved exactly until it is corrected or cleared by you personally. Those. If you type “Samsung TVs” into the line, the list of Yandex.Wordstat keys will be updated, but previously collected keys will be saved in the plugin list.

According to this scheme, we run all the pre-prepared “basic” keys through Wordstat, collect everything we need, and then by clicking on one of these two buttons we copy the previously collected list to the clipboard. Please note that the button with two leaves copies the list without frequencies, and with two leaves and the number 42 - with frequencies.

The list copied to the clipboard can then be pasted into an Excel spreadsheet.

Also during the collection process, you can view impression statistics by region. For this purpose, Yandex.Wordstat has the following switch.

Well, as a bonus, you can look at the request history - find out when the frequency increased and when it decreased.

This feature can be useful in determining the seasonality of a request, as well as for identifying a decline/growth in popularity.

Another interesting feature is the statistics of impressions for the specified phrase and its forms. To do this, you must enclose the query in quotation marks.

Well, if you add an exclamation mark before each word, then the statistics will display the number of impressions by key without taking into account word forms.

No less useful is the minus operator. It removes key phrases that contain the word (or several words) you specify.

There is another tricky operator - the vertical separator. It is necessary in order to combine several lists of keys into one (we are talking about keys of the same type). For example, let’s take two keys: “tours to Antalya” and “trip to Antalya”. We write them in the Yandex.Wordstat line as follows and get 2 lists for these keys, combined into one:

As you can see, we have a lot of keys where there are “tours”, but no “vouchers” and vice versa.

Another important feature is the frequency binding to the region. You can select your region here.

Using Wordstat to collect a semantic core is suitable if you are collecting mini-cores for some individual pages, or do not plan large cores (up to 1000 keys).

SlovoEB and Key Collector

We're not kidding, that's exactly what the program is called. In a nutshell, the program allows you to do exactly the same thing, but in automatic mode.

This program was developed by the LegatoSoft team - the same team that developed Key Collector, we will also talk about it. In essence, Slovoeb is a heavily trimmed (but free) version of Key Collector, but it is quite capable of working with the collection of small semantic cores.

Especially for Slovoeb (or Key Collector) it makes sense to create a separate account on Yandex (if they ban you, it’s not a pity).

It is necessary to make small adjustments one-time.

The login-password pair must be entered separated by a colon and without spaces. Those. if your login [email protected] and the password is 15101510ioioio, then the pair will look like this: my_login:15101510ioioio

Please note that there is no need to enter @yandex.ru in your login.

This setup is a one-time event.

Let's make a couple of points clear:

  • How many projects to create for each site is up to you to decide
  • Without creating a project, the program will not work.

Now let's look at the functionality.

To collect keys from Yandex.Wordstat, on the “Data Collection” tab, click on the “Batch collection of words from the left column of Yandex.Wordstat” button, insert a list of previously prepared key phrases, click “Start collection” and wait for it to finish. There is only one drawback to this collection method - after parsing is completed, you have to manually delete unnecessary keys.

As a result, we get a table with keywords collected from Wordstat and the base frequency of impressions.

But we remember that you can also use quotation marks and an exclamation point, right? This is what we will do. Moreover, this functionality is implemented in Sloyoba.

We start collecting frequencies in quotes and watch how the data gradually appears in the table.

The only negative is that the data is collected through the Yandex.Wordstat service, which means that even collecting frequencies for 100 keys will take quite a lot of time. However, this problem is solved in Key Collector.

And one more function that I would like to talk about is the collection of search tips. To do this, copy the list of previously parsed keys to the clipboard, click the button for collecting search tips, paste the list, select the search engines from which search tips will be collected, click “Start collection” and wait for it to finish.

As a result, we get an expanded list of key phrases.

Now let’s move on to Slovoeb’s “big brother” - Key Collector.

Key Collector is paid, but has much wider functionality. So if you are professionally involved in website promotion or marketing, Key Collector is simply a must-have, because Wordfucker will no longer be enough. In short, Kay Collector can do:

  • Parse keys from Wordstat*.
  • Parse search suggestions*.
  • Cutting off search phrases using stop words*.
  • Sorting requests by frequency*.
  • Identification of duplicate requests.
  • Determination of seasonal requests.
  • Collection of statistics from Liveinternet.ru, Metrica, Google Analytics, Google AdWords, Direct, Vkontakte and others.
  • Determination of relevant pages for a particular request.

(the * sign indicates the functionality available in Slovoyobe)

The process of collecting keywords from Wordstat and collecting search tips is absolutely identical to that implemented in Slovoyobe. However, frequency collection is implemented in two ways - through Wordstat (as in Slovoyobe) and through Direct. Through Direct, the collection of frequencies is accelerated several times.

This is done as follows: click on the D button (short for “Direct”), check the box to fill out the Wordstat statistics columns, check the boxes (if necessary) about what frequency we want to get (base, in quotes, or in quotes and with exclamation marks”, click “Get data” and wait for the collection to complete.

Collecting data through Yandex.Direct takes much less time than through Wordstat. However, there is one drawback - statistics may not be collected for all keys (for example, if the key phrase is too long). However, this minus is compensated by collecting data from Wordstat.

Google Keyword Planner

This tool is extremely useful for collecting a core based on the needs of Google search engine users.

Using Google Keyword Planner, you can find new queries by query (no matter how strange it may sound), and even by site/topic. Well, as a bonus, you can even predict traffic and combine new search queries.

For existing requests, statistics can be obtained by selecting the appropriate option on the main page of the service. If necessary, you can select a region and negative keywords. The result will be output in CSV format.

How to find out the semantic core of a competitor’s website

Competitors can also be our friends, because... You can borrow ideas for choosing keywords from them. For almost every page you can get a list of keywords for which it is optimized, manually.

The first way is to study the page content, Title, Description, H1 and KeyWords meta tags. You can do everything manually.

The second way is to use Advego or Istio services. This is quite enough to analyze specific pages.

If you need to perform a comprehensive analysis of the semantic core of the site, then it makes sense to use more powerful tools:

  • SEMrush
  • Searchmetrics
  • SpyWords
  • Google Trends
  • Wordtracker
  • WordStream
  • Ubersuggest
  • Topvisor

However, the above tools are more suitable for those who are engaged in the professional promotion of several sites at the same time. “For yourself” even the manual method will be quite enough (as a last resort - Advego).

Errors when compiling a semantic core

The most common mistake is a very small semantic core

Of course, if this is some highly specialized niche (for example, the hand-made production of elite musical instruments), then in any case there will be few keys (one hundred, one and a half, two hundred).

The larger the semantic core (but without “garbage”), the better. In some niches, the semantic core can consist of several... MILLIONS of keys.

The second mistake is synonymizing. More precisely, its absence

Remember the example of Antalya. Indeed, in this context, “tours” and “vouchers” are the same thing, but these 2 lists of keys can be radically different. “Stripper” may well be searched for “wire stripper” or “insulation removal tool.”

At the bottom of the search results, Google and Yandex have this block:

It is there that you can often spot synonyms.

Compiling a semantic core exclusively from high-frequency queries

Remember what we said at the beginning of the post about low-frequency queries, and the question “why is this an error?” You won't have any more problems. Low-frequency queries will bring the bulk of traffic.

"Garbage", i.e. non-targeted requests

It is necessary to remove from the assembled kernel all requests that do not suit you. If you have a cell phone store, then for you the request “cell phone sales” will be targeted, and “cell phone repair” will be garbage. In the case of a service center for repairing cell phones, everything is exactly the opposite: “repair of cell phones” is targeted, and “sale of cell phones” is garbage. The third option is if you have a cell phone store with a service center “attached” to it, then both requests will be targeted.

Once again, there should be no garbage in the kernel.

No grouping of requests

It is strictly necessary to split the core into groups.

Firstly, this will allow you to create a competent site structure.

Secondly, there will be no “key conflicts”. For example, let’s take a page that is promoted by the queries “buy self-leveling floor” and “buy an acer laptop.” The search engine may be confused. As a result, it will fail for both keys. But for the queries “hp 15-006 laptop buy” and “hp 15-006 laptop price” it already makes sense to promote one page. Moreover, it doesn’t just “make sense”, but will be the only correct solution.

Thirdly, clustering will allow you to estimate how many pages still need to be created so that the core is completely covered (and most importantly, is it necessary?).

Errors in separating commercial and information requests

The main mistake is that requests that do not contain the words “buy”, “order”, “delivery”, etc. can also turn out to be commercial.

For example, the request "". How to determine whether a request is commercial or informational? It’s very simple - look at the search results.

Google tells us that this is a commercial request, because... in our search results, the first 3 positions are occupied by documents with the word “buy”, and although the fourth position is occupied by “reviews”, look at the address - this is a fairly well-known online store.

But with Yandex everything turned out to be not so simple, because... in the TOP 5 we have 3 pages with reviews/feedback and 2 pages with trade offers.

However, this request still applies to commercial ones, because There are commercial offers both here and there.

However, there is also a tool for mass verification of keys for “commerce” - Semparser.

We picked up “empty” queries

Both base and quoted frequencies must be collected. If the frequency in quotes is zero, it is better to delete the request, because it's a dummy. It often happens that the base frequency exceeds several thousand impressions per month, and the frequency in quotes is zero. And immediately a concrete example - the key “inexpensive skin cream”. Base frequency 1032 impressions. Looks delicious, doesn't it?

But all the flavor is lost if you put the same phrase in quotation marks:

Not all users type without errors. Because of them, “crooked” key queries end up in the database. Including them in the semantic core is pointless, since search engines still redirect the user to the “corrected” query.

And it’s exactly the same with Yandex.

So we delete “crooked” requests (even if they are high-frequency) without regret.

An example of the semantic core of a site

Now let's move from theory to practice. After collection and clustering, the semantic core should look something like this:

Bottom line

What do we need to compile a semantic core?

  • at least a little businessman (or at least marketer) thinking
  • at least some SEO skills.
  • it is important to pay special attention to the structure of the site
  • figure out what queries users can use to search for the information they need
  • based on “estimates”, collect a list of the most suitable queries (Yandex.Wordstat + Wordstat Assistant, Slovoeb, Key Collector, Google Keyword Planner), frequencies taking into account word forms (without quotes), and also without taking into account (in quotes), remove “garbage” "
  • the collected keys must be grouped, i.e. distribute across site pages (even if these pages have not yet been created).

No time? Contact us, we will do everything for you!

At the moment, factors such as content and structure play the most important role for search engine promotion. However, how to understand what to write text about, what sections and pages to create on the site? In addition to this, you need to find out exactly what the target visitor to your resource is interested in. To answer all these questions you need to collect a semantic core.

Semantic core— a list of words or phrases that fully reflect the theme of your site.

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

Here is an example of a query core broken down into a structure:


By clustering I mean breaking your search queries down into separate pages. This method will be relevant for both promotion in Yandex and Google PS. In this article I will describe a completely free way to create a semantic core, but I will also show options with various paid services.

After reading the article, you will learn

  • Choose the right queries for your topic
  • Collect the most complete core of phrases
  • Clean up uninteresting requests
  • Group and create structure

Having collected the semantic core you can

  • Create a meaningful structure on the site
  • Create a multi-level menu
  • Fill pages with texts and write meta descriptions and titles on them
  • Collect positions of your website for queries from search engines

Collection and clustering of the semantic core

Correct compilation for Google and Yandex begins with identifying the main key phrases of your topic. As an example, I will demonstrate its composition using a fictitious online clothing store. There are three ways to collect the semantic core:

  1. Manual. Using the Yandex Wordstat service, you enter your keywords and manually select the phrases you need. This method is quite fast if you need to collect keys on one page, however, there are two disadvantages.
    • The accuracy of the method is poor. You may always miss some important words if you use this method.
    • You will not be able to assemble a semantic core for a large online store, although you can use the Yandex Wordstat Assistant plugin to simplify it - this will not solve the problem.
  2. Semi-automatic. In this method, I assume using a program to collect the kernel and further manually breaking it down into sections, subsections, pages, etc. This method of compiling and clustering the semantic core, in my opinion, is the most effective because has a number of advantages:
    • Maximum coverage of all topics.
    • Qualitative breakdown
  3. Auto. Nowadays, there are several services that offer fully automatic kernel collection or clustering of your requests. Fully automatic option - I do not recommend using it, because... The quality of collection and clustering of the semantic core is currently quite low. Automatic query clustering is gaining popularity and has its place, but you still need to merge some pages manually, because the system does not provide an ideal ready-made solution. And in my opinion, you will simply get confused and will not be able to immerse yourself in the project.

To compile and cluster a full-fledged correct semantic core for any project, in 90% of cases I use a semi-automatic method.

So, in order we need to follow these steps:

  1. Selection of queries for topics
  2. Collecting the kernel based on requests
  3. Cleaning up non-target requests
  4. Clustering (breaking phrases into structure)

I showed an example of selecting a semantic core and grouping into a structure above. Let me remind you that we have an online clothing store, so let’s start looking at point 1.

1. Selection of phrases for your topic

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

How to select queries to collect semantics from Yandex Wordstat

Go to the service, select the city(s)/region(s) you need, enter the most “fatty” queries in your opinion and look at the right column. There you will find the thematic words you need, both for other sections, and frequency synonyms for the entered phrase.

How to select queries before compiling a semantic core using competitors

Enter the most popular queries into the search engine and select one of the most popular sites, many of which you most likely already know.

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

At this stage, it is important to do the right thing: to cover as much as possible all possible words from your topic and not miss anything, then your semantic core will be as complete as possible.

Applying to our example, we need to create a list of the following phrases/keywords:

  • Cloth
  • Shoes
  • Boots
  • Dresses
  • T-shirts
  • Underwear
  • Shorts

What phrases are pointless to enter?: women's clothing, buy shoes, prom dress, etc. Why?— These phrases are the “tails” of the queries “clothes”, “shoes”, “dresses” and will be added to the semantic core automatically at the 2nd stage of collection. Those. you can add them, but it will be pointless double work.

What keys do I need to enter?“low boots”, “boots” are not the same thing as “boots”. It is the word form that is important, not whether these words have the same root or not.

For some, the list of key phrases will be long, but for others it consists of one word - don’t be alarmed. For example, for an online store of doors, the word “door” may well be enough to compile a semantic core.

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

2. Collecting queries for the semantic core

For proper, full collection, we need a program. I will show an example using two programs simultaneously:

  • On the paid version - KeyCollector. For those who have it or want to buy it.
  • Free - Slovoeb. A free program for those who are not ready to spend money.

Open the program

Create a new project and call it, for example, Mysite

Now to further collect the semantic core, we need to do several things:

Create a new account on Yandex mail (the old one is not recommended 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 click the “save changes” button below. More details in the screenshots.

We select a region to compile a semantic core. You need to select only those regions in which you are going to promote and click save. The frequency of requests and whether they will be included in the collection in principle will depend on this.

All settings are completed, all that remains is to add our list of key phrases prepared in the first step and click the “start collecting” button of the semantic core.

The process is completely automatic and quite long. You can make coffee for now, but if the topic is broad, for example, like the one we are collecting, then this will last for several hours 😉

Once all the phrases are collected, you will see something like this:

And this stage is over - let's move on to the next step.

3. Cleaning the semantic core

First, we need to remove requests that are not interesting to us (non-target):

  • Related to another brand, for example, Gloria Jeans, Ecco
  • Information queries, for example, “I wear boots”, “jeans size”
  • Similar in topic, but not related to your business, for example, “used clothing”, “clothing wholesale”
  • Queries that are in no way related to the topic, for example, “Sims dresses”, “puss in boots” (there are quite a lot of such queries after selecting the semantic core)
  • Requests from other regions, metro, districts, streets (it doesn’t matter which region you collected requests for - another region still comes across)

Cleaning must be done manually as follows:

We enter a word, press “Enter”, if in our created semantic core it finds exactly the phrases that we need, select what we found and press delete.

I recommend entering the word not as a whole, but using a construction without prepositions and endings, i.e. if we write the word “glory”, it will find the phrases “buy jeans at Gloria” and “buy jeans at Gloria”. If you spelled "gloria" - "gloria" would not be found.

Thus, you need to go through all the points and remove unnecessary queries from the semantic core. This may take a significant amount of time, and you may end up deleting most of the collected queries, but the result will be a complete, clean and correct list of all possible promoted queries for your site.

Now upload all your queries to excel

You can also remove non-target queries from semantics en masse, provided you have a list. This can be done using stop words, and this is easy to do for a typical group of words with cities, subways, and streets. You can download a list of words that I use at the bottom of the page.

4. Clustering of the semantic core

This is the most important and interesting part - we need to divide our requests into pages and sections, which together will create the structure of your site. A little theory - what to follow when separating requests:

  • Competitors. You can pay attention to how the semantic core of your competitors from the TOP is clustered and do the same, at least with the main sections. And also see which pages are in the search results for low-frequency queries. For example, if you are not sure whether or not to create a separate section for the query “red leather skirts,” then enter the phrase into the search engine and look at the results. If the search results contain resources with such sections, then it makes sense to create a separate page.
  • Logics. Do the entire grouping of the semantic core using logic: the structure should be clear and represent in your head a structured tree of pages with categories and subcategories.

And a couple more tips:

  • It is not recommended to place less than 3 requests per page.
  • Don’t make too many levels of nesting, try to have 3-4 of them (site.ru/category/subcategory/sub-subcategory)
  • Do not make long URLs, if you have many levels of nesting when clustering the semantic core, try to shorten the urls of categories high in the hierarchy, i.e. instead of “your-site.ru/zhenskaya-odezhda/palto-dlya-zhenshin/krasnoe-palto” do “your-site.ru/zhenshinam/palto/krasnoe”

Now to practice

Kernel clustering as an example

To begin with, let’s categorize all requests into main categories. Looking at the logic of competitors, the main categories for a 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 gender/age, such as simply “shoes”, “outerwear”.

We group the semantic core using Excel. Open our file and act:

  1. We break it down into main sections
  2. Take one section and break it into subsections

I will show you the example of one section - men's clothing and its subsection. In order to separate some keys from others, you need to select the entire sheet and click conditional formatting->cell selection rules->text contains

Now in the window that opens, write “husband” and press enter.

Now all our keys for men's clothing are highlighted. It is enough to use a filter to separate the selected keys from the rest of our collected semantic core.

So let’s turn on the filter: you need to select the column with queries and click sort and filter->filter

And now let's sort

Create a separate sheet. Cut the highlighted lines and paste them there. You will need to split the kernel in the future using this method.

Change the name of this sheet to “Men’s clothing”, a sheet where the rest of the semantic core is called “All queries”. Then create another sheet, call it “Structure” and put it as the very first one. On the structure page, create a tree. You should get something like this:

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

For ease of use and navigation through your clustered semantic core, provide links from 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 you need to methodically separate the requests manually, simultaneously deleting what you may not have been able to notice and delete at the kernel cleaning stage. Ultimately, thanks to the clustering of the semantic core, you should end up with a structure similar to this:

So. What we learned to do:

  • Select the queries we need to collect the semantic core
  • Collect all possible phrases for these queries
  • Clean out "garbage"
  • Cluster and create structure

What, thanks to the creation of such a clustered semantic core, can you do next:

  • Create a structure on the site
  • Create a menu
  • Write texts, meta descriptions, titles
  • Collect positions to track dynamics of requests

Now a little about programs and services

Programs for collecting the semantic core

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

  • Yandex Wordstat Assistant is a plugin that makes it convenient to select queries from Wordstat. Great for quickly compiling the core for a small site or 1 page.
  • Keycollector (word - free version) is a full-fledged program for clustering and creating a semantic core. It is very popular. A huge amount of functionality in addition to the main direction: Selection of keys from a bunch of other systems, the possibility of auto-clustering, collecting positions in Yandex and Google and much more.
  • Just-magic is a multifunctional online service for kernel compilation, auto-breaking, text quality checking and other functions. The service is shareware; to fully operate, you need to pay a subscription fee.

Thank you for reading the article. Thanks to this step-by-step manual, you will be able to create the semantic core of your website for promotion in Yandex and Google. If you have any questions, ask in the comments. Below are the bonuses.

Considering the constant struggle of search engines with various link factors, the correct structure of the site is increasingly coming to the fore when carrying out search engine optimization of the site.

One of the main keys for competent development of the site structure is the most detailed elaboration of the semantic core.

At the moment, there are quite a large number of general instructions on how to make a semantic core, so in this material, we tried to give more details on exactly how to do it and how to do it with minimal time.

We have prepared a guide that answers step by step the question of how to create the semantic core of a website. With specific examples and instructions. By using them, you will be able to independently create semantic cores for promoted projects.

Since this post is quite practical, a lot of different work will be done through Key Collector, since it saves quite a lot of time when working with the semantic core.

1. Formation of generating phrases for collection

Expanding phrases for parsing one group

For each group of queries, it is highly advisable to immediately expand it with synonyms and other wording.

For example, let’s take the request “swimwear” and get various other reformulations using the following services.

Wordstat.Yandex - right column

As a result, for the initially specified phrase, we can still receive 1-5 other different reformulations for which we will then need to collect queries within one group of queries.

2. Collecting search queries from various sources

After we have identified all the phrases within one group, we move on to collecting data from various sources.

The optimal set of parsing sources to obtain the highest quality output data for RuNet This:

● Wordstat.Yandex - left column

● Yandex + Google search suggestions (with search by endings and substitution of letters before a given phrase)

Clue : if you do not use a proxy in your work, then in order to prevent your IP from being banned by search engines, it is advisable to use the following delays between requests:

● In addition, it is also advisable to manually import data from the Prodvigator database.

For bourgeoisie we use the same thing, except for data from Wordstat.Yandex and data from Yandex PS search suggestions:

● Google search suggestions (using endings and substituting letters before a given phrase)

● SEMrush - corresponding regional database

● similarly, we use import from Prodvigator’s database.

In addition, if your site already collects search traffic, then for a general analysis of search queries in your topic, it is advisable to download all phrases from Yandex.Metrika and Google Analytics:

And for a specific analysis of the desired group of queries, you can use filters and regular expressions to isolate those queries that are needed for the analysis of a specific group of queries.

3. Cleaning queries

After all queries have been collected, it is necessary to carry out preliminary cleaning of the resulting semantic core.

Cleaning with ready-made lists of stop words

To do this, it is advisable to immediately use ready-made lists of stop words, both general and specific to your topic.

For example, for commercial topics such phrases would be:

● free, download, …

● abstracts, Wikipedia, wiki, ...

● used, old, …

● job, profession, vacancies, …

● dream book, dream, ...

● and others of this kind.

In addition, we immediately clear all cities of Russia, Ukraine, Belarus, ….

After we have loaded the entire list of our stop words, we select the option for the type of occurrence search “independent of the word form of the stop word” and click “Mark phrases in the table”:

This way we remove obvious phrases with negative words.

After we have cleared away obvious stop words, we then need to review the semantic core manually.

1. One of the quick ways is this: when we come across a phrase with obvious words that are not suitable for us, for example, a brand that we do not sell, then we

● opposite such a phrase, click on the indicated icon on the left,

● choose stop words,

● select a list (it is advisable to create a separate list and name it accordingly),

● immediately, if necessary, you can select all phrases that contain the specified stop words,

● add to the list of stop words

2. The second way to quickly identify stop words is to use the “Group Analysis” functionality, when we group phrases by words that are included in these phrases:

Ideally, in order not to repeatedly return to certain stop words, it is advisable to include all marked words in a specific list of stop words.

As a result, we will get a list of words to send to the stop word list:

But, it is advisable to quickly look at this list so that ambiguous stop words do not appear there.

This way you can quickly go through the main stop words and remove phrases that contain these stop words.

Cleaning up hidden duplicates

● sort by descending frequency for this column

As a result, we leave only the most frequent phrases in such subgroups, and delete everything else.

Cleaning phrases that do not carry much meaning

In addition to the above-mentioned word cleaning, you can also remove phrases that do not carry much semantic meaning and will not particularly affect the search for groups of phrases for creating separate landing pages.

For example, for online stores, you can remove phrases that contain the following keywords:

● buy,

● sale,

● online store, … .

To do this, we create another list in Stop Words and add these words to this list, mark them and remove them from the list.

4. Grouping requests

After we have cleared out the most obvious garbage and inappropriate phrases, we can then begin grouping queries.

This can be done manually, or you can use some help from search engines.

We collect results for the desired search engine

In theory, it is better to collect by the desired region in Google PS

● Google understands semantics quite well

● it is easier to collect, it does not ban various proxies

Nuances: even for Ukrainian projects, it is better to collect results from google.ru, since the sites there are better structured, therefore, we will get much better results for landing pages.

Such data can be collected

● and with the help of other tools.

If you have a lot of phrases, then to collect data, search engine results will obviously need proxies. The combination of A-Parser + proxies (both paid and free) shows the optimal speed of collection and operation.

After we have collected the search results, we now group the requests. If you have collected data in Key Collector, then you can further group phrases directly in it:

We don’t really like how KC does it, so we have our own developments that allow us to get much better results.

As a result, with the help of such grouping we are able to quickly combine requests with different formulations, but with the same user problem:

As a result, this leads to a good saving of time for the final processing of the semantic core.

If you do not have the opportunity to collect results yourself using a proxy, then you can use various services:

They will help you quickly group queries.

After such clustering based on the search results, in any case, it is necessary to conduct a further detailed analysis of each group and combine those that are similar in meaning.

For example, such groups of requests ultimately need to be combined onto one page of the site

The most important: Each individual page on the site should meet one user need.

After processing the output semantics in this way, we should get the most detailed structure of the site:

● information requests

For example, in the case of swimsuits, we can create the following site structure:

which will contain their title, description, text (if necessary) and products/services/content.

As a result, after we have already ungrouped all the queries in detail, we can begin to collect in detail all the key queries within one group.

To quickly collect phrases in Key Collector we:

● we select the main generating phrases for each group

● go, for example, to parsing hints

● select distribute into groups

● select from the drop-down list “Copy phrases from” Yandex.Wordstat

● click Copy

● and begin collecting data from another source, but for the same distributed phrases within groups.

Eventually

Let's look at the numbers now.

For the topic “swimwear”, we initially collected more than 100,000 different queries from all sources.

At the query cleaning stage, we managed to reduce the number of phrases by 40%.

After that, we collected the frequency for Google AdWords and left for analysis only those with a frequency greater than 0.

After that, we grouped queries based on Google PS results and we managed to get about 500 groups of queries within which we had already carried out a detailed analysis.

Conclusion

We hope that this guide will help you collect semantic cores for your sites much faster and better and will step by step answer the question of how to assemble a semantic core for a site.

Good luck collecting semantic cores, and as a result, quality traffic to your sites. If you have any questions, we will be happy to answer them in the comments.

(78 ratings, average: 4,90 out of 5)

Hi all! When you run a blog or content site, there is always a need to compile a semantic core for the site, cluster or article. For convenience and consistency, it is better to work with the semantic core according to a well-established scheme.

In this article we'll consider:

  • how the semantic core for writing an article is collected;
  • what services can and should be used;
  • how to correctly enter keys into an article;
  • my personal experience in selecting SY.

How to collect a semantic core online

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

In this article, I will collect SYNOPSIS on the topic “how to lay laminate flooring.” In a similar way, you can use these instructions for compiling a semantic core for any topic.

  1. If our article is on the topic “how to lay laminate flooring,” then we will enter this query to obtain information about the frequency in wordstat.yandex.ru.

As we can see, in addition to the target request, we received many similar requests containing the phrase "lay laminate", here you can eliminate all unnecessary things, all the keys that will not be discussed in our article. For example, we will not write about similar topics such as “how much does it cost to install laminate flooring”, “The laminate was laid unevenly” and so on.

To get rid of many obviously inappropriate requests, I recommend using operator "-" (minus).

  1. We substitute a minus, and after it all the words are off topic.

  1. Now, select everything that remains and copy the queries into Notepad or Word.

  1. Having inserted everything into the Word file, we go through it with our eyes and delete everything that will not be disclosed in our article. If there are false queries, then a keyboard shortcut will help you check for their presence in the document Ctrl+F, a window opens (sidebar on the left), where we enter the search words.

The first part of the work is done, now we can check our Yandex semantic core template for pure frequency, the quote operator will help us with this.

If there are few words, then this can be easily done directly in Wordstat by substituting the phrase in quotation marks and finding the pure frequency (quotes indicate how many requests there were with the content of this particular phrase, without additional words). And if, as in our example of the semantic core of an article or website, there are a lot of words, then it is better to automate this work using the Mutagen service.

To get rid of numbers use the following steps with a Word document.

  1. Ctrl+A— to highlight the entire contents of the document.
  2. Ctrl+H— calls up a window for replacing characters.
  3. Substitute in the first line ^# and click “replace all” this combination will remove all numbers from our document.

Be careful with keys that contain numbers, the above steps may change the key.

Selection of semantic core for a website/article online

So, I wrote in detail about the service. Here we will continue learning how to compile a semantic core.

  1. We go to the site and use this program to compile a semantic core, since I haven't seen a better alternative.
  1. First, let’s parse the pure frequency; for this we go through “Wordstat parser” → “mass parsing”


  1. We paste all our selected words from the document into the parser window (Ctrl+C and Ctrl+V) and select “Wordstat Frequency” in quotes.

This process automation costs only 2 kopecks per phrase! If you are selecting a semantic core for an online store, then this approach will save you time for mere pennies!

  1. Click send for verification and, as a rule, after 10-40 seconds (depending on the number of words) you will be able to download words that already have a frequency in quotes.
  1. The output file has the extension .CSV and is opened in Excel. We begin to filter the data to create an online semantic core.


  • We add the third column, it is needed to display competition (in the next step).
  • We set a filter on all three columns.
  • We filter the “frequency” column “in descending order”.
  • Everything that has frequency below 10 - deleted.

We received a list of keys, which we can use in the text of the article, but first it is necessary to check them for competition. After all, if this topic has been covered far and wide on the Internet, does it make sense to write an article on this key? The likelihood that our article on it will reach the TOP is extremely low!

  1. To check the competition of the online semantic core, go to “competition”.


  1. We begin to check each request and substitute the resulting competition value into the corresponding column in our Excel file.

Price checking one key phrase is 30 kopecks.

After the first top-up, you will have access to 10 free checks every day.

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

It's worth writing an article:

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

Compiling a semantic core online using low-competition phrases will give you traffic. If you have a new website, it will not appear immediately, you will have to wait from 4 to 8 months.

In almost any topic you can find MF and HF with low competition from 1 to 5; for such keys it is realistic to receive 50 visitors per day.

To cluster semantic core queries, use , they will help you create the correct site structure.

Where to insert the semantic core in the text

After collecting the semantic core for the site, it’s time to write key phrases into the text and here are some recommendations for beginners and those who “don’t believe” in the existence of search traffic.

Rules for inserting keywords into the text

  • You only need to use the key once;
  • words can be declined according to cases, changed places;
  • you can dilute phrases with other words, it’s good when all the key phrases are diluted and readable;
  • you can remove/replace prepositions and question words (how, what, why, etc.);
  • You can insert the signs “-” and “:” into the phrase

For example:
there is a key: “How to lay laminate flooring with your own hands” in text it might look like this: “...in order to lay laminate boards with our own hands we will need...” or so, “Everyone who tried to lay laminate flooring with their own hands...”.

Some phrases already contain others, for example the phrase:
“how to lay laminate flooring in the kitchen with your own hands” already contains a key phrase “how to lay laminate flooring with your own hands”. In this case, it is allowed to omit the second one, since it is already present in the text. But if there are not enough keys, then it is better to also use it in the text either in the Title or in the Description.

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

Contraindications for writing keys

  • You cannot separate the key phrase with a comma (only as a last resort) or a period;
  • You cannot enter the key into the text directly so that it will not look natural (not readable);

Page title and description

Title and Description- this is the title and description of the page, they are not visible in the article, they are shown by Yandex and Google when search results are displayed to the user.

Writing rules:

  • the title and description should be “journalistic”, that is, attractive for clicking;
  • contain thematic (relevant to the request) text, for this we enter key phrases (diluted) in the title and description.

Are common character requirements at the plugin All in one SEO pack, the following:

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

You can check your creation or that received from for plagiarism using.

With this, we have dealt with the topic of what to do with the semantic core after compilation. In conclusion, I will share my own experience.

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

You may think that I am trying to sell you something that is not believable. In order not to be unfounded, here is a screenshot of statistics for the last, (but not the only) year of this site, How I managed to rebuild my blog and start getting traffic.

This training in compiling a semantic core, although long, is effective, because in website building the main thing is the right approach and patience!

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

In contact with

If you know the pain of search engines’ “dislike” for the pages of your online store, read this article. I will talk about the path to increasing the visibility of a site, or more precisely, about its first stage - collecting keywords and compiling a semantic core. About the algorithm for its creation and the tools that are used.

Why create a semantic core?

To increase the visibility of site pages. Make sure that Yandex and Google search robots begin to find pages of your site based on user requests. Of course, collecting keywords (compiling semantics) is the first step towards this goal. Next, a conditional “skeleton” is sketched out to distribute keywords across different landing pages. And then articles/meta tags are written and implemented.

By the way, on the Internet you can find many definitions of the semantic core.

1. “The semantic core is an ordered set of search words, their morphological forms and phrases that most accurately characterize the type of activity, product or service offered by the site.” Wikipedia.

To collect competitor semantics in Serpstat, enter one of the key queries, select a region, click “Search” and go to the “Key phrase analysis” category. Then select “SEO Analysis” and click “Phrase Selection”. Export results:

2.3. We use Key Collector/Slovoeb to create a semantic core

If you need to create a semantic core for a large online store, you cannot do without Key Collector. But if you are a beginner, then it is more convenient to use a free tool - Sloboeb (don’t let this name scare you). Download the program, and in the Yandex.Direct settings, specify the login and password for your Yandex.Mail:
Create a new project. In the “Data” tab, select the “Add phrases” function. Select your region and enter the requests you received earlier:
Advice: create a separate project for each new domain, and create a separate group for each category/landing page. For example: Now collect semantics from Yandex.Wordstat. Open the “Data collection” tab - “Batch collection of words from the left column of Yandex.Wordstat”. In the window that opens, select the checkbox “Do not add phrases if they are already in any other groups.” Enter a few of the most popular (high-frequency) phrases among users and click “Start collecting”:

By the way, for large projects in Key Collector you can collect statistics from competitor analysis services SEMrush, SpyWords, Serpstat (ex. Prodvigator) and other additional sources.



 

It might be useful to read: