A Guide for Choosing the Cannabis Keywords for Marijuana Company

Img source: pixabay.com

The most effective way to promote websites has been and remains organic search. At the same time, any SEO promotion strategy begins with building a semantic core. How to do this to promote the cannabis website in the TOP and attract targeted visitors, we will tell you in the article.

What Is the Semantic Core

Img source: jeffbullas.com

The semantic core is words and phrases that reflect the subject matter and structure of the site. Using these queries, users are looking for you on the Internet, so it is crucial to determine what potential customers are interested in and what words they use when searching for the information they need. The better the site (and each of its pages individually) answers these queries, the more it reveals the topic, the more search engines will love you, and the more targeted visitors you can attract.

But this does not mean that it is enough to stuff each page with keywords, and the search engines will move the site forward. The algorithms have become smarter, and now you can easily get banned for this. To have a good result, you need to look at promotion like a marketer. Of course, the best option is to contact a professional agency like mjseo.agency that specializes in local SEO for cannabis brands and will conduct the research for you.

At the same time, you can collect a semantic core by yourself. First of all, you need to understand what users will be typing into a search engine to find information that is posted on your site. This is the starting point from which you will start when building a semantic core. And also, when you collect the core, it is crucial to correctly distribute the search phrases across the pages, determining which of them most accurately answer specific queries.

Separation of keywords by popularity

Img source: unsplash.com

First of all, keywords are divided by popularity. According to this criterion, there are 3 categories:

  1. High-frequency: the frequency of displaying phrases is from 1,000 per month.
  2. Medium-frequency: the frequency of displaying phrases is up to 1,000 per month.
  3. Low-frequency: frequency of displaying phrases is up to 100 per month.

And again: you cannot solely rely on these numbers. Here you need to look at the subject matter and competitors. For some projects, the frequency of displaying low-frequency queries can be increased to 1,000 per month, medium-frequency ones — up to 5,000 per month, and high-frequency ones — up to 10,000 per month.

At the same time, it might be useful to assess the keyword competition. This is a marker of ranking difficulty. It allows you to understand how many webpages you need to leave behind to rank with a particular keyword. Usually, it depends on the popularity of a certain keyword and competition in the industry.

Img source: pixabay.com

Parsing competitors

Find competitor sites. The more you find, the better. Examine these sites for optimization possibilities. Usually, you can immediately see if the site is being served or if it has been done just as if it is not worth your attention.

Parse all selected sites using Screaming Frog or a similar service. You need data from the title and H1 fields. As a rule, for most sites, these tags are made according to templates. And it will not be hard to remove unnecessary information from the tags and leave only keywords. Process the title and H1 of competitors. Collect all this into one big spreadsheet.


It is essential to divide keywords by geo and remove geo requests if you plan to promote in several states. If you are engaged in local SEO, you should manually add locations you are interested in. These might be countries, states, or cities. Sometimes, it is also reasonable to add neighborhoods or neighboring cities.

Stop words

After dealing with geo, you need to compose your own dictionary of stop words. Collect your personal list of negative words, such as tootsie roll, get stoned, and so on. The list will gradually expand, making it easier for you to filter in the future. Do not be lazy to add frequent junk words to it. For each topic, such a list of stop words is unique. Therefore, you should not look for some kind of universal lists and create your own.

Img source: pexels.com


You need to separate brand and product queries from pure category ones. To do this, you create your own list of branded stop words. This includes various spelling options for brands, transliteration, and misprints. Filter out the list of queries by brand, placing all such queries in a separate folder “Brand queries.”


Separation of info requests. Create your own stop-word vocabulary to separate informational queries from commercial ones. This includes words like reviews, top, rating, best, etc. Put information in a separate folder “information requests.”

Primary structure

Drawing up the primary site structure is necessary to simplify further filtering. By creating a site structure, you also make it easier for yourself in the future to parse and process requests because they will be collected from about the same topic/group of requests. And it is much easier to weed out waste from one such batch than trying to remove it from one large common mass.

The more detailed the project structure you create, the easier it will be for you to work. This structure is only needed for parsing and filtering. What you will implement on the site will be very different from this preliminary structure.

Img source: pexels.com


Here, you work with the structure, sorting requests into folders, simultaneously getting rid of waste. This is the longest and most tedious process, but it needs to be conducted. In parallel, you will find geo, information, brand, and product queries, and your stop word dictionaries will be replenished. The structure will become more detailed and understandable. The project will begin to take shape; you will finally understand how much you have missed.

In general, an understanding of how it all works, what is convenient and what is not, will already begin to take shape. At the same time, you will see how many new options for writing queries you already know exist, but you did not even know about them.