How to do content analysis yourself or what an A-grade text should be. Content analysis using the example of a study of the newspaper "Vecherny Rostov" Ready-made content analysis example

It has already been repeated many times that content is one of the significant ranking methods in both Yandex and Google. And in order to promote your website, you should pay great attention to its content. At the same time, take into account both the technical characteristics and the benefits that a potential client will gain by visiting your resource.

If you want your “brainchild” to delight you with good positions, excellent traffic and an influx of customers, conduct a content analysis of the site. Thanks to it, you will identify errors (maybe minor ones) that prevent your resource from conquering the coveted top. Well, let's figure it out?

Conventionally, content analysis can be divided into 3 parts:

Oh, how many sites there are in the world filled with unnecessary, uninteresting and simply “indigestible” texts, after reading which you are left with a slight (and sometimes not entirely) feeling of regret about wasted time. Remember, you need to fill the site with content not “out of the blue,” but constantly keeping in mind the thought: will this be important to my potential client or did I insert this paragraph for the sake of saying something? So here's what you should always keep in mind.

Unique Selling Proposition

Even if it seems to you that you are no different from your competitors, you can always find a feature that will set you apart from the “gray mass.” And it’s not only possible, but also necessary! After all, if you yourself don’t know why your company is worth contacting for a certain product, then how will the client understand it? People are always looking for something better/cheaper/more beautiful - this is an axiom. Here is the best/beautiful/inexpensive and there is a unique selling proposition.

We have already written about the types of USP, as well as how to compose it earlier - we advise you to read it. The main thing to remember is that a unique selling proposition should not be a deception, otherwise you will lose the client’s trust as if by magic: screw it up and it’s gone. After all, it’s unlikely that your customers will be pleased to learn that the loud slogan “on our website prices are 50% lower than those of competitors” is just a marketing ploy. In addition, remember that the phrases “high quality”, “highly qualified specialists”, “prompt delivery” are not a USP, but hackneyed cliches that should be gotten rid of long ago. What if we really have prompt delivery? - you can say indignantly. Great, then back it up with numbers and benefit the client. For example: We will deliver the product within 2 hours after placing the order. Voila - a clichéd phrase that has lost its value has turned into an excellent selling proposition.

Catch another example of a good USP suitable for the B2B sphere.

Sales text structure

Do you still think that by posting a huge “canvas” of albeit useful text, the user will willingly spend a good half hour reading it? But it’s not like that. Quantity is far from synonymous with quality. As practice shows, users most often leave the site in the first 10 seconds of viewing, when they realize that they cannot cope with the cumbersome text. Accordingly, all this negatively affects not only the client’s mood, because he did not find anything valuable for himself, but also on behavioral factors.

Take care of your potential clients, pay special attention to the structure of the text. The simpler the information is presented, the faster the signal “oh, this is exactly what I’m looking for” will be sent to the brain of potential customers. And you don’t need to be a distant relative of Pablo Picasso or have a design education to design the text beautifully. Icons, underlines, highlights, fonts - experiment and find your style. Just don’t overdo it; a riot of colors and volumes is also not the best option.

An example of an aesthetic and attractive placement:

And these are by no means all the secrets for correct text formatting. You will find detailed illustrated tips on how to make text readable for users.

Semantic uniqueness of the text

No, now we’re not talking about the uniqueness that can be checked through special services - we’ll talk about that later. We mean added value. What makes you different from hundreds of other sites with similar topics. In addition to developing a good USP, you can use other techniques. For example, be targeted at a specific target audience. Let's say that while your competitors are trying to promote their music portal aimed at all music lovers, you can “target” only jazz lovers. Yes, we have sharply limited the target audience, but fans of Ella Fitzgerald, Ray Charles and other cult personalities can bring you such traffic (if they find something interesting for themselves) that your competitors with a wide target audience have never even dreamed of. To learn about other ways to capture added value, welcome here.

No errors

Do you naively believe that the presence of typos and spelling/grammatical/stylistic errors does not affect anything? I'll have to disappoint you - how it affects you. You can’t even imagine how meticulously most users read texts, and, having found elementary shortcomings, ardent fans of the Russian language frown tragically, immediately close the page and never return to the site. It seems like a minor oversight, but it can undermine the company’s authority. And it will also have a negative impact on the bounce rate. Therefore, be sure to check your texts using the services glvrd.ru, orfogrammka.ru and the magic wand that knows the answers to all questions - the website Gramota.ru. And if you want to be 100% sure of the quality of your content, contact professional copywriters.

Analysis of text technical characteristics for SEO benefits

So, we’ve talked about the “sublime” - it’s time to move on to statistical data, without which you can forget about the good positions of the site. Comprehensive work is also important here - sometimes even for a small mistake you can “get a belt” from the search engines. But let’s put philosophy aside until better times and get down to business.

Estimation of text volume

For a page to be indexed well, a text of at least 500 characters is required. But it’s not always worth focusing on this value - it all depends on the area of ​​business and the type of resource. For example, for an online store product card, 500 characters will be more than enough.

But for the main page, this volume is only enough to describe the activities of your company. The optimal indicator is a text of 1500-2500 characters. But even here you need to keep your finger on the pulse: if your competitors, who are already in the TOP, have more information than 4000 characters, then you should also increase your text, otherwise there is a risk of getting lost.

Technical uniqueness of text content

From article to article we repeat, like a mantra, that texts must be unique. To check, we recommend using the services TEXT.RU or CONTENT-WATCH.

However, even if you don’t find fault with your content, it’s worth protecting yourself from theft. Yes, yes, believe me, there are not only pickpockets: your competitors recognized your article as a masterpiece of copywriting skill and “borrowed” it for themselves. But the most annoying thing is that search engines may consider the text on your resource to be non-unique and exclude the page from the index. To protect your content, be sure to secure rights in Yandex.Webmaster. We have already written about how to do this.

Inclusion of keywords in the text

Soon the horror story about Baden-Baden will be on a par with the stories about the Queen of Spades and the “black, black city.” Only a caveman would not understand that we are now talking about the algorithm and, accordingly, the sanctions of Yandex, and not about the famous German resort. But jokes aside: before posting the text, be sure to check your nausea using the Advego service. The optimal figure is 9-11%.

Exceeding the value already threatens with overspam and the train to Baden-Baden flashing on the horizon. After posting text on the site, we also recommend tracking nausea and the number of keywords through the site pr-cy.ru.

The norm is from 4 to 8%. If the value is less, it is worth adding key queries to the text, otherwise the dream of the coveted top will remain a dream. When re-optimizing, on the contrary, remove several keywords, otherwise the subsequent struggle with filters will take a lot of time and effort.

And a few more words about uniqueness

Only now about the images. Original and unique pictures can bring you additional traffic: search engines will index the images and display them in search results. We talked about the benefits of photographs and proper work with them in one of our articles, so today we will not focus on this.

Correct spelling of title title

One of the most important tags, because it is the one that is reflected in the search results, forming part of the snippet. Looking at the title, people decide whether they should visit this site or whether it’s better to surf the Internet. Therefore, you should be very careful when creating a title.

1. Maximum length- no more than 70 characters. Search engines also index long titles, but in the search results they appear cut off, which looks unaesthetic and unattractive.
2. Using keywords. If you still think that by stuffing the title with keywords you will immediately get to the top 10, then we are now going to really, really disappoint you. Believe me, the fewer keywords are used in the title, the more weight they acquire. Therefore, use 1-2 important queries.
3. Relevance. The title should reflect the content of the page. Therefore, if your title contains the key query “prices” or “services”, but this information is not in the text, then the potential client will leave the site after 7-10 seconds, muttering dissatisfiedly under his breath. Do you need it?
4. Whole phrase. Remember that you are targeting (most often) a person who speaks Russian, so the headlines should be a solid, catchy phrase. And a set of words in the style of “buy a green corset is beautiful and inexpensive” is better left for Ali Express.

Analysis of text compliance with reader expectations

Well, the final block, which is very often ignored. And in vain. To prevent your site from falling in search results, you should not forget about 3 important points.

Determining Page Relevance

Let us remind you that relevance is the correspondence of a page to a key query. Dilution of relevance can lead to loss of page weight, deterioration in behavioral factors and, as a result, a decrease in traffic. Therefore, the verification should be taken very seriously. Yes, this is a labor-intensive process that requires time and patience, but who said that website promotion is fast, fun and easy?

Determining the relevance of a page can be done manually by entering a query into a search engine and clicking “Search on site.”

See the first page that appears? She is relevant. In order not to get confused in keywords, we display all the data in a table. Yes, in this case it is worth being a pedant. =)

The manual method is perhaps the most reliable. But if your temperament does not allow you to scrupulously check each request, then luckily for you there are several services that will check the relevance of pages in a couple of minutes: Topvisor, All Positions or Site Analysis.com.

It looks something like this:

We generate a report in the same way, only now it will take not 2 hours to prepare it, but a maximum of 20 minutes. If the smart machine has not found relevant pages for all queries, we recommend that you take the time and check them manually.

Defining Landing Pages

Just a second for educational purposes: a landing page is the one that the user lands on directly from the search. And if the relevance of the page is determined by His Majesty’s search engines, then you yourself are responsible for the “landing”. But it’s not in vain that we struggled so much with compiling the table - we apply the results in practice.

However, there are often cases when Yandex and Google disagree and identify different relevant pages for the same key query. Here's a clear example:

Naturally, we cannot promote two pages using the same keyword, so we start from positions in the search results. For the query “haircut in a beauty salon,” the site ranks 192 in Yandex, and 213 in Google. It is logical that we will choose the page /strizhka/ as a landing page. We are sure that you will want to rise in search, so put all your efforts into promoting this article. And over time, Google will take pity and accept this page as relevant.

Tracking changes in relevance

Have you decided that once you’ve sorted out the relevant and landing pages, you’re done? Not so, everything can change. And not for the better for you. It is imperative to monitor the change of relevant pages - the services that we have already named today - Topvisor and All Positions - will help you with this. After each check, they will provide you with data and let you know how often the items change.

Constant jumps can lead to cannibalization of key queries, and then you will have to say goodbye to dreams of thousands of customers. We have already told you how to prevent this.

Instead of a conclusion

In the article, we highlighted the algorithm for conducting website content analysis, which you can easily handle on your own. But, as they say, you won’t be satisfied with just theory =). We invite you to consider for the main page from A to Z.

However, if you are not sure that you can handle website promotion on your own, and the idea of ​​​​a flow of clients does not let you sleep peacefully at night, then welcome to us. We will audit the site, identify errors and eliminate them. After all, in our field, as in a university: first the owner works for the site, and then the site pays the owner with a gold coin. Or come to us for comprehensive content marketing, you can see the tariffs.

Content analysis is a method of quantitative analysis of text documents, widely used in the social sciences and humanities. Its essence is to reliably determine the meaning and direction of a specific text message by counting semantic units.

Instructions

The object of research can be any documents containing text messages. In particular, newspaper articles, public speeches of public and political figures, books, answers to questionnaires, diaries, official statements, etc. Content analysis can be carried out either manually or automatically. The second option is used to study large amounts of text data and requires computational and special statistical programs.

To conduct independent content analysis, without the use of automatic processing tools, you first need to determine the array of data with which you will work. For example, if you intend to analyze press coverage of a regional election campaign, then the necessary sample will be all newspaper publications on this topic for the selected period.

The second step in the content analysis procedure is the selection of semantic units that are directly related to the problem being studied. Separate words and phrases that carry a key semantic load can act as a semantic unit. For example, in the context of an election campaign, such units can be the names of candidates, the phrases “modernization of the economy,” “business development,” “struggle for power,” etc. Moreover, the selected semantic units should be characteristic of all texts under study.

The next stage is one of the most important in the entire content analysis process. It is a codification of units of text. Its essence is to develop rules for correlating semantic units with a list of analysis categories. The result of the codification stage is the development of a codifier, which includes not only a list of observed indicators, but also data about the document in which they are present. If we are talking about newspaper articles, then the name of the publication, city, date of issue, format, number of pages, page of placement, etc. will be taken into account.

After forming a sample of messages, selecting semantic units and creating a codifier, we proceed directly to the text analysis itself. In practice, this is expressed in the compilation of a dictionary in which each observation (semantic unit) belongs to a specific type or class in accordance with the rules of the codifier. After that, a quantitative calculation of the use of all semantic units is made. An important point is also to assign specific ratings (positive, negative or neutral) to key mentions. In other words, a fairly fine ranking is required. Scaling by pairwise comparison or the so-called Q-sort method are traditionally used as ranking methods. You can learn more about both of these methods in textbooks on applied sociology or political science.

Good afternoon, readers and guests of the site. Ekaterina Kalmykova is with you. Surely every blogger would like to know how high-quality his content is, whether readers like it, whether it is easily perceived and whether it is understood at all.

For example, I have always been interested in this question, since I write all the articles on the blog myself and I want to know whether I express my thoughts in an accessible way for other users. The fact that Anton reads my articles and evaluates them positively is understandable. He might as well have said something against it. I would have been left without dinner :)

But what do blog readers, independent experts, so to speak, think about the text? How to do content analysis yourself and whether it can be done - that’s what’s really interesting. And it is on this topic that I want to invite you to discuss with me.

To begin with, I want to tell you a funny incident that happened to me this morning. I needed to donate blood early, and so, waking up in a not very good mood from the upcoming event, I went to the hospital. There is a queue, nowhere to sit, a lot of people, everyone is gloomy and dissatisfied as usual in the morning, including the nurse doing the test samples. Well, okay, I think all hospitals are the same and everyone imagines the atmosphere in them in the morning.

In general, I donated blood, left the hospital, got on a minibus and went home. When leaving, I hand the driver a fare of 13 rubles. We have such a tariff in our city. To which the driver smiles, takes only 10 rubles and says: “I have a discount for traveling in shoe covers.” 🙂 And I just noticed that I never took off my shoe covers after the hospital. The whole minibus laughed. Overall, this driver just made my day. The bad mood immediately disappeared. And I returned home joyfully. Here's the story.

Well, now let's return to the topic of the article. What distinctive features does a good one have? Let's try to do a little analysis and identify them.

One glance is enough to determine what type of text we are reading: good, good or excellent. We skim over a passage of text and understand it. But when you need to write a text yourself and evaluate it objectively, difficulties may arise.

So, what are the distinguishing characteristics that I highlight?

An A-grade text should:

Knowing these characteristics of the text and using them in practice, your texts will always be excellent. Articles will stand out against the backdrop of numerous information on the Internet, and your resource (website or blog) will have unique content that both visitors and search engines love.

Photos on the topic

Content analysis is a method of quantitative analysis of text documents, widely used in the social sciences and humanities. Its essence is to reliably determine the meaning and direction of a specific text message by counting semantic units.

Just follow these simple step-by-step tips and you'll be on the right track.

Quick step by step guide
So, let's get down to action, focusing on the result.

Step - 1
The object of research can be any documents containing text messages. In particular, newspaper articles, public speeches of public and political figures, books, answers to questionnaires, diaries, letters, official statements, etc. Content analysis can be carried out either manually or automatically. The second option is used to study large amounts of text data and requires computer technology and special statistical programs. Having done this, move on to the next steps.

Step - 2
To conduct independent content analysis, without the use of automatic processing tools, you first need to determine the array of data with which you will work. For example, if you intend to analyze press coverage of a regional election campaign, then the necessary sample will be all newspaper publications on this topic for the selected period. Having done this, move on to the next steps.

Step - 3
The second step in the content analysis procedure is the selection of semantic units that are directly related to the problem being studied. Separate words, names, phrases that carry a key semantic load can act as a semantic unit. For example, in the context of an election campaign, such units can be the names of candidates, the phrases “modernization of the economy,” “development of small business,” “struggle for power,” etc. Moreover, the selected semantic units should be characteristic of all texts under study. Having done this, move on to the next steps.

Step - 4
The next stage is one of the most important in the entire content analysis process. It is a codification of units of text. Its essence is to develop rules for correlating semantic units with a list of analysis categories. The result of the codification stage is the development of a codifier, which includes not only a list of observed indicators, but also data about the document in which they are present. If we are talking about newspaper articles, then the name of the publication, city, date of issue, format, number of pages, page of placement, etc. will be taken into account. Having done this, move on to the next steps.

Step - 5
After forming a sample of messages, selecting semantic units and creating a codifier, we proceed directly to the text analysis itself. In practice, this is expressed in the compilation of a dictionary in which each observation (semantic unit) belongs to a specific type or class in accordance with the rules of the codifier. After that, a quantitative calculation of the use of all semantic units is made. An important point is also to assign specific ratings (positive, negative or neutral) to key mentions. In other words, a fairly fine ranking is required. Scaling by pairwise comparison or the so-called Q-sort method are traditionally used as ranking methods. You can learn more about both of these methods in textbooks on applied sociology or political science. Having done this, move on to the next steps.

Step - 6
The content analysis procedure is completed by quantitatively counting the data obtained and calculating the arithmetic mean of the scale for each case. The resulting average scores are then ranked in a certain way.