What is LSI copywriting texts?

Perhaps every copywriter who writes search engine optimization (SEO) texts have ever heard of the concept of LSI copywriting. Or LSI texts. If you still have not heard – also not bad, because today you will get a big portion of new information and can safely add a 20-30% price to your texts.

On the good side, there is a whole Wikipedia article about latent semantic indexing (that’s how the acronym is deciphered), but it is written too scientifically and without examples. So We decided to make a simple, clear, accessible, and easy-to-understand article tutorial on what LSI texts are and how to write them correctly. In other words, We want to share information with you today which can be immediately applied in practice and get the result. Are you ready? Then make yourselves comfortable, we are starting!

What is LSI copywriting

At the dawn of the 2000s, search engines (Google, Yahoo) and others did not work as they do now. They analyzed how the page of the site meets the user’s request, counting the number of occurrences of keywords. As a result, the optimizers “pumped” keywords texts to the tie. And for the sake of credibility, also singled out them in bold font. It was unrealistic to read such materials, but they were well ranked (took the leading places) in the search results.

An example of a text that at the dawn of the 2000s was ranked well and now is guaranteed to fall under the filter.
An example of a text that at the dawn of the 2000s was ranked well and now is guaranteed to fall under the filter.

Over time, search engines have become more advanced. They have learned to analyze the Intent (the user’s intention, the reason why he enters a particular search query) and select pages with the answer based on the type of search query: commercial (when a person wants to buy something) and information (when he wants to learn something). For information queries, the principle of text analysis based on latent or latent semantic indexing (LSI).

The essence of the principle: search engines, analyzing texts on the site, take into account not only keywords (so-called, semantic core), but also associated words that reveal the topic (so-called, thematic core): synonyms, associations, hyponyms, hyperonyms, related concepts, etc. The larger and richer the topic core, the more valuable the text looks for search engines and the higher its ranking. That’s why many SEO-specialists when giving the authors the terms of reference (TOR), indicate not only the keys but also the subject dictionary – the same accompanying words to activate the LSI-factor.

Let’s look at an illustrative example. Imagine that you have a high-frequency (HF) golf request. And you are writing an article. Based on the searches alone, the search engine cannot understand what you are writing about: whether it is about the game, or a vehicle. Therefore, the robot during the indexing will begin to analyze the surrounding text (its thematic core). For example, if we have it as in the figure below, it will immediately be clear what we are talking about.

The same is true for more substantive keywords. When a user enters a “rule of the game” query, one of the weight factors of the ranking is a developed thematic core to give a comprehensive answer to a person’s question, reducing the search time to a minimum.

In other words, the direction of LSI copywriting is based on writing texts for search queries, adding to them the maximum number of “close” in the sense of (thematic) words to develop an associative core, so it would be more useful than competitors.

“Tail” LSI queries

The main reason why LSI texts are actively used in search engine promotion is not only page ranking by main queries. There is another subtlety, which provides individual sites up to 50% of additional traffic (audience from search engines). And this subtlety is called “tail queries”.

Tailing queries – these are long queries, which are usually entered like 1-10 times a month. Or even less frequently: once or twice every 2-3 months. For example: “Which is the best laptop for a copywriter: Asus, Dell, Lenovo or MacBook and why.” There are even such requests, and they are not shown in the Google Keyword Planner tool.

And now imagine that there are two articles. Both are optimized for the query: “Which notebook is better for the copywriter”. But one describes the lineup of Acer, MSI, Xiaomi, and HP, and the other describes models from the query’s tail. In your opinion, which one will be higher in terms of the “tail” suggested above? Naturally, with all other conditions being equal, the answer is the second one.

The main strength of the tailing requests is that there are a lot of them: hundreds, thousands, tens of thousands per month. And, of course, there is no point in optimizing or “sharpening” the page for each of them separately. But! The more elaborate your thematic core, the more tail queries you capture automatically, and thus, the more people can it bring to your site. Here it is, the true power of LSI copywriting: to get an audience even when the main requests are rabid competition.

By the way, if you look at the realities of SEO promotion, you will notice that novice optimizers paired with copywriters are trying hard to “sharpen” the page for the main keys. However, in case the project is new or there is a high level of competition, it is not reasonable, because this way you can wait for the results for a very long time.

That’s why advanced optimizers and copywriters for such tasks focus on low-frequency keywords (which request up to 500-700 times per month), LSI promotion and “tail” requests. So you can get the first results in 2-3 weeks and develop the site much faster. But the main thing: a copywriter can promote his own site or blog, even without the help of an SEO-specialist.

LSI texts and SEO-copywriting

And here We have a very important clarification to make: LSI texts are not a substitute for SEO-copywriting. It is just one of its attributes in the realities of modern search engine optimization. Yes, when you know how to make a thematic core it’s a good help, and in some cases, you can collect “tail” traffic on low-frequency queries without competition. But to achieve the maximum effect in solving the problem, the basics of SEO are a must-have.

How to write LSI texts?

The process of writing text (most often articles) is slightly different from the process of writing an ordinary article. And the main difference is that before developing the text, you think through the thematic core. This is very important because both the LSI factor and the number of “tail” keys will directly depend on it.

To do this, We will now show you a very simple but powerful technique. This technique is based on the method of associative thinking, and it can be used not only in copywriting but also in naming, slogan development, as well as in solving a number of other, more complex tasks.

Create an associative core (LSI-core) of the first order

A thematic or, as it is also called, the associative core is not created at once but in stages. At the first stage, we take a local semantic core (keyword list) for the text and write it out. Or we reduce it to an intellect card. This will be much easier and more visible in the future. There still will be minor queries.

Next, for each keyword (or set of keywords if there are many) we create a cloud of associations, synonyms, hyponyms, hyperonyms, and other related words. These are the so-called LSI queries of the first order. If you have enough words, and the topic is not particularly competitive, you can stop there.

If the queries are similar, the thematic cores will largely coincide. Then it’s better to group the keywords to save time and save yourself some extra work.

How do I collect key queries on LSI technology?

LSI queries are words and phrases on which search engines can better understand the meaning of the text. They are divided into two groups:

Synonyms of words included in queries from the semantic core.
Additional key phrases that are associated with keys from the semantic core. For example, for “Golf” query such LSI queries can be words “Caddie”, “Par”.
We have figured out what LSI keywords are. Let’s see, in what sources can we find them?

  1. Dictionary of synonyms

Use special dictionaries to find synonyms for words from key queries. Use this information carefully so as not to add outdated words or synonyms that do not fit the context.

  1. Search engine prompts and tools
Keyword planner in LSI

When you enter a query into the search string, search prompts are automatically displayed. They may contain additional keywords that are worth adding to the text. Look for tips in Yahoo, Bing and Google and write the ones that match the subject of your text. And of course, DO NOT forget about the Keyword Planner tool, this is your main instrument to gather LSI core.

  1. Viewing requests from the block “Searches related to”

The lower part of the search results contains a block listing queries that are searched together with a given phrase. This block may contain LSI words.

There is no need to write down all the words you found in the prompts, WoKeyword planner and other sources. All you need to do is write out the most appropriate words that will allow you to explore the topic in greater depth. Note that all tools may contain phrases that are completely unrelated to the topic of your article. Such phrases should be filtered out.

Create a second order thematic core

We have LSI first order queries. Next, we take each of these words and make our associative mini-cores for them.

Important: the second order core must be directly or indirectly related to the first order core, i.e. if you include all words that relate to Golf but have no relation to the game, there will be little sense.

Vocabulary and common sense play a major role in the development of such associative cores. It is not so easy to define them programmatically. Sort of look-a-like systems, but they are either complex and paid or primitive or far from perfect and inferior to the method of associative thinking. It is good to connect neural networks here. Unfortunately, online services for the selection of LSI words based on neural networks are not yet available. That’s why in LSI copywriting the higher level of erudition and reading by the author – the better.

Add LSI phrases

For most tasks, the first two stages are enough. But we are not looking for easy ways, so we will strengthen the strategy and can make a set of additional stable phrases (LSI phrases), which are used in natural speech.

A fragment of a cloud of associative phrases looks like this: play golf, where can I find golf stores, learn to play golf, golf rules.


The main question is how effectively it will be ranked. But the main convenience of this article is that you can see the result in the nearest future, since you may be the only article that solves a very specifically asked question.

If the materials will occupy high positions in search and attract traffic – good. If not – then you need to review LSI queries, to study once again competitors and “re-assemble” the text. Plus, work on behavioral factors.

Rate article
Add a comment