The semantic core. What is it and how to collect it?
To attract visitors to your Internet site, you need to know and understand exactly what and how network users will search. In order to make it easier, a semantic core is compiled – this is the set of words or their combinations that can fully describe the direction and structure of the Internet resource. In simple words – the semantic basis.
When you start to compose an SC, the task is to answer the question – “what information do I carry?”. Since a successful business is built on customer focus, you always need to look from the outside, i.e. through the eyes of a potential visitor to the page.
Compiling a kernel will help to solve another important problem – sorting key phrases into sections of the entire site. When working with semantics, it is determined which page is suitable for a specific user question.
To solve this issue, you can use two methods:
The structure is created according to the analysis of search queries. Thus, the semantic core will be the foundation for building the resource structure.
The structure is determined, and only then, based on the created “grid”, is the analysis of the SQ, i.e. they are customized to the already created site.
Both the first and second method work successfully. However, any specialist will tell you that using the second one is not only more productive, but simply more logical. Since you will not depend on the situation and can easily influence it.
How is a resource hierarchy built?
First, let’s explain that the structure of a site is a hierarchical layout of pages. Here a certain number of primary problems are solved right away – the submission of information, the convenience and compliance with the requirements of search engines. To make it, you have the opportunity to use any convenient method – an Excel table, a graphic designer, or, in general, draw a piece of pencil on a sheet of A4 sheet. Most importantly, before you begin, answer yourself two questions to facilitate further actions:
- What do I want to tell visitors?
- Where will it be more convenient for them to watch this or that section with information?
We will tell you a little about what you need to know about the CS in order to effectively organize the semantic core, you need to not only understand what keys are, but also know which ones your target audience will use for search purposes.
They can be classified according to several characteristics. Based on them, CFs will be divided into groups:
This concept is the set of words or phrases that your potential visitors will use to search. For example, to prepare a shank, the hostess will write in the search bar “pork shank recipe photo”.
- low-frequency – are shown no more than 100 times per month;
- mid-frequency – a thousand times;
- high-frequency – all that is more than 1000, although some experts believe that it can include impressions of five thousand or even ten.
How is frequency estimated? It directly depends on the popularity of topics. What you need to know about it when compiling a semantic core? According to statistics, approximately 2⁄3 of all entered questions in the network are low-frequency. Therefore, AE should be as wide as possible and supplemented from time to time. This does not mean that the other groups can simply be ignored. Without them, anyway. But as the main tool for attracting visitors, use words with a low frequency.
Now we will understand the distribution according to needs:
- Information – people will enter them to look for the necessary information, articles, news and more. As an example, this may be such a request: “how to clean pans from soot”, “the latest news of show business in Russia”, “recipe for home face masks” and more.
- Commercial – the goal here is to buy something. Often found with such inclusions are “price”, “cost”, “buy”, “delivery”, “stock”, “cheap”.
- Others – those phrases by which to determine the desire and further actions of the user is extremely difficult. For example, a person enters the string “pizza”. What is needed is not clear. Order, go eat in a cafe, cook at home or find out the story of creating a dish and calorie content?
Many experts identify more navigation requests. They include, as a rule, the desire to find information about a particular site, the address or phone number of organizations, tracking packages and so on. If this type is not typical for your resource, you can skip it.
Using these classification methods, it will be easier to understand what people want and how to “spread” content throughout the site. Questions with intent can be attributed to sections of trading sites such as a catalog or product cards, and information related to blogs or news.
Also, do not forget that commercial issues have fairly high competition. Therefore, the wider the key, the more likely it is that the buyer will come to your pages, and not to the pages of competitors. If you are the owner of an online hardware store, then the query “buy a xiaomi phone” has less chance than a “buy a xiaomi 8 phone”.
Key phrases can consist of several parts – body, specifier, tail.
To make it clear, consider a specific example. Back to the pizza. As mentioned above, with one word the system is difficult to determine the intent of a particular user. Competition in search results in a particular case will be very high. If you own a pizzeria, then using this key will lead to inappropriate traffic. Why is this happening? The thing is that the key itself in this case consists only of the body.
Look at the phrase “order pizza.” It just determines the desire for action. Here we see both the body and the “order” specifier. It is the latter that can attribute the request itself to informational or commercial.
Pay attention to such examples:
- order pizza;
- make pizza;
- how to serve pizza.
It happens that the opposite desire is expressed – to sell.
Now we detail the question – “order pizza delivery”. Here the inclusion of all parts – the body is “pizza”, the qualifier “order” and the tail “with delivery”.
In two cases, we see that a person expresses a desire to buy a product, but the tail shows exactly how he wants to do it.
If you know the anatomy of the CS, you can easily derive an approximate algorithm for selecting a semantic core. Define the basic terms that are related to your activities and user needs. For the owner of the Italian food delivery service, these will be “ravioli”, “pasta”, “pizza”, “carpaccio”, “bolognese”, “carbonara” and others.
Once you have identified the basic concepts, you can do the “tails” and qualifiers. So you will increase the core competitiveness and be able to reach more people.
For a resource that already exists when collecting KP, you can use the Google Analytics and Yandex.Metrica analytics systems. With their help, you can easily find out what phrases that relate to your direction are most often used by the audience. All statistics data can be found in the webmaster.
Competitive Web Space Analysis
Competitive sites can be a great source of examples and ideas for keywords. For one specific page, you can find them with your own hands. To determine them, it is enough to study the content information or check the content of the meta tag in the page code. Also, special services that conduct semantic analysis will be good helpers. Many use Advego. To analyze the entire site, use the tools of a comprehensive analysis of competitors. As a rule, such tools require payment in the form of a monthly subscription.
Stages of the selection of components for semantics
- determination of the base using which people will search for your goods and services;
- expansion of SC;
- removal of inappropriate search phrases.
At the very beginning, the BC gathering will resemble a meeting of crazy scientists. Because the most effective method is good old brainstorming. Stock up on coffee and energy, get together with colleagues and write everything that comes to mind or shouts out in a furious impulse. Take every breath. You will discuss and filter later, because if you start to do it right away you can miss important points. As they say: doubt is the enemy of genius.
Most of the keys you write down at this stage will be low frequency. To get high-frequency or with medium frequency, you will need to begin to expand the KP.
Wordstat is one of the most popular and simple tools for selection. This service will help to analyze all the keys that were recorded earlier. As a result of the work done, you will receive a list of phrases for all written basic keys. Please note that there will be one thousand requests.
Removing excess and inappropriate
This is the time to work, because the process not only requires a lot of effort and attention, but also time. All work is carried out exclusively by hand. What phrases need to be removed?
- mention of competitors or their brands;
- goods and services that are not present in your catalog, and you do not want to supplement their assortment;
- exclude the words “cheap”, “inexpensive”, “discount” if you are not currently reducing product prices so as not to damage your reputation and attract users with inappropriate behavioral characteristics;
- KW that are duplicated;
- Inappropriate locations. For example, you are engaged in pizza delivery in Kiev, then phrases with the name of other cities do not suit you;
- typos and errors inquiries. Fortunately, search engines are endowed with the right intelligence and are able to understand a person who is searching for information, even if he enters a proper name with a small letter or makes a critical error in another word.
After the lists for the entire database are compiled, you can proceed to grouping.
KW grouping and relevance map building
There is such a thing as clusters of semantics. This is a combination of phrases thanks to which the target audience can find your site. In appearance it is a kind of structure built on the basis of many levels. How many steps there will be in one cluster directly depends on how broad the topic you are conducting your activity. Clusters are divided into orders.
Since we have already talked about the delivery of Italian dishes, we will consider examples on this topic.
So, in this case, “dishes of Italian cuisine” is a cluster of the first level. The second level or, more simply, subcategories will be generalized names like “pasta”, “pizza”, “drinks”, “soups”, “pastries”, etc. Theoretically, the number of clusters can reach up to 5. But in this case there will be only two. and then goods cards will go.
It is very important to work out exactly the second group in order to be competitive and get on the leaderboards. The first group is determined at the time of the brainstorming.
At the second level, KP should be expanded with the help of words that determine what actions a person plans to perform.
Well, now it’s time to talk in detail about tools that will help in collecting semantics.
If you are a professional in Internet marketing, then you definitely can not do without Key Collector. Especially if the SC of your site is very large.
How does the program help?
- selection of key phrases occurs through Wordstat;
- Parses search hints
- removes inappropriate due to the presence of stop words;
- requests are filtered according to frequency;
- looking for doubles that are hard to detect with the naked eye;
- defines requests that may be relevant in a particular season;
- collects statistics from other services and sites;
- Searches for relevant pages.
- The tool has many useful functions, it works in automatic mode. However, the program costs money. The plus is that on the network you can find a sufficient number of absolutely free analogues, which are no worse.
Service from the creators of the previous application. It also parses, detects the frequency and collects keys through Wordstat. But it’s absolutely free. There are, of course, some difficulties in use. For example, in Yandex it is better to use a spare direct account, since the system can put a block for automatic requests.
Selection of KW from Yandex service
If SC is small and you are new to optimization and promotion, then Wordstat will be the best solution. The service is absolutely free and fairly easy to learn.
- allows you to see statistics on impressions of the keyword itself or phrases that contain it. It makes it possible to conduct an analysis not only according to general information, but also according to mobile users;
- displays information on location sections;
- in history, you can observe the dynamics of popularity growth or its decline depending on seasonality, everything is displayed in a simple and understandable schedule;
- statistics will be displayed exclusively for the offers or their interpretations indicated in the search. To get complete information, the main word is written in quotation marks;
- key forms are not displayed in statistics, so an exclamation mark should be written in front of it. For example, “! Pizza”.
- doesn’t take into account negative keywords. To see the data for the phrase “Christmas cake”, you need to enter the minus – “Christmas cake-recipe”;
- display data using the selected preposition. Put a “+” in front of him to see the full picture;
- statistics can be displayed for entire groups. To do this, write the groups in brackets, and put a dash between the keys;
- binding to certain regions;
If you are ready to start creating groups and analyzing manually, then Wordstat is the most suitable way.
In locations and regions in which Google is the dominant, it will be logical to use its planner to compile the semantic core. You can search for keywords on sites, topics. Also, the application will allow you to plan traffic, create new advanced queries or their combinations. There is a corresponding function on the main service section. Do not forget to indicate the region you are interested in for more accurate information.
The search engine is the third in the ranking, so you can’t miss statistics on it. Here, information is provided taking into account the gender and age of the audience, you can also see data for the previous month.
Paid SEO tool that has many useful abilities in its arsenal.
With its help, you can analyze and select questions of different frequency. The application successfully works with both Google and Yandex. Shows the following information:
- The presence of competitors on a scale of one to hundreds.
- Calculates the estimated price of the contextual conversion click.
- The graph displays popularity in dynamics.
- Shows “rivals” by topic in search engine results and ad units.
- In a special section, you can choose the extension, gives not only tips, but also similar phrases in meaning. Also here you can observe all the results from the TOP for the entered proposal. You can track the increase in the popularity of competing organizations in the interest of the KW.
- In one of the sections you can get a content plan, as well as phrases with a long “tail”. Capable of generating new clusters.
By itself, it costs money, but it has a nice feature to do a free check of ten KW per day. To receive such a bonus, a new user just needs to make one top-up. Searches for CFs that have low competition at high frequencies. The analysis is carried out automatically by thirty leaders in the issuance list for the query of interest on Yandex.
Again, not a free analysis assistant. If you enter the base, he will pick up options for entire offers, analyze the frequency, price of transitions, as well as the issuance in real time and competitors.
The following results will be visible:
- Price for Google Edwards.
- The frequency of important words in general and each individually.
- Regional redistribution.
- Allows you to determine if there is a difficulty in taking a place in the leaders for a specific request. Convenient is not it? You can immediately understand whether it is worth working further with the key.
Paid service that searches for KW on popular sites such as Google, eBay, youtube and others. The version without payment will only display the list, and for the money will give advanced information, i.e. will show the dynamics and others.
It will help not only pick up, but also conduct clustering. It has a basic tariff and a free trial period, the capabilities of which are very limited. It works in automatic mode. All information collected is amenable to export to a convenient document. Gathers hints from popular search engines and allows you to expand the semantic core.
There is a useful option for copywriters, which allows you to create TT taking into account the necessary KW.
An assistant who does clustering. For a fee. After you register, you can use the free test package to determine if it suits you or not. Generates terms of reference and key reference texts.
A tool for SEO professionals. Here, too, you need to pay, but, as in previous applications, there are free actions with restrictions on functions. SC is collected based on its own databases, collecting data from analytics systems. In addition to semantics, it is possible to carry out division into clusters according to the results of TOP search engines. In addition, there is the ability to collect according to regional affiliation, commercial degree and subject matter.
What to do with SC?
When you start planning content, then a table with a ready-made semantic core should become for you a kind of guiding map and the main source for introducing ideas. Lists with trial names of pages and sections of the site will be decisive in the needs of your target audience. A key that is included in a text or title does not have to be highly popular. Its main task is to reflect the need and intention of the user. The search phrase should answer your question for your copywriter – “what you need to write about”.
Remember that “stuffing” the filling immediately with all the keys is unreasonable. The most important thing is the potential customer and we focus on it. Therefore, you need to answer all the questions that he supposedly can ask. Focus on completeness and full disclosure.
Back to the past or how not to do to be in trend
Let’s keep up with the times. The world does not stand still and SEO optimization too. Once upon a time, promotion specialists entered the COP in such a way that the text spam was even scary, and this was considered normal. Now, because of such “dense” indicators, search engines can reject the Old. Let’s dwell on these points in detail and analyze what should be avoided.
- Do not exclude work with those who have too much competition. Yes, they are unlikely to help you get into the TOP, but can be a great addition to the content plan.
- Low-frequency phrases. This is also an idea for effective content, although previously, professionals considered them garbage and a waste of money.
- Formulas and coefficients. In the end, semantics is a branch of linguistics that studies semantic meanings. Therefore, we will leave all kinds of mathematical definitions to specialists in mathematics. If you start to actively use the same KEI, you can miss a lot of useful things. No formula can correctly characterize human behavior.
- Do not create pages for requests. Well, really, you have nowhere else to spend money. Remember our pizza? “Order pizza” and “pizza delivery” essentially mean the same person’s desire to eat a delicious homemade tortilla. So why then two sections on this topic?
- Automation. Everyone wants faster and easier. But this is not always good. A robot is also a robot in Africa. Applications and services for collecting nuclear weapons still require manual rechecking. Do not lose sight of this so that there are no critical errors. Only the human brain can really assess and analyze the need and importance of the collected keys and phrases.
- Concentration only on CF collection. If you are creating a new and small-sized project, you don’t need to unnecessarily monitor your competitors and collect tips. Just use the statistics. And only when you feel that the number of ideas for filling has come to naught, begin to search and use the alternative in the form of special opportunities.
Perhaps, we have examined the topic of composing the semantic core in more than detail. Remember that if you are not ready to work on your own or are not sure that you can go through the whole process in a quality manner, specialists will always come to the rescue!