Chat GPT as a Keyword Research Tool: Case Study

ChatGPT can be an invaluable tool for keyword research. Its machine learning and language models give it the power to uncover great keywords that can be used to optimize SEO. We’ve seen some creative and innovative examples of how Twitter users have been able to use ChatGPT for this purpose. 

Today we’ll dive deeper into how ChatGPT can help you with keyword research and showcase the potential of the AI-powered platform.

Different Types of Keyword Research and Analysis

There are several ways to identify which keywords best suit your website. You can use keyword research and analysis to help you determine the right terms to target. 

Keyword Discovery: This technique involves figuring out the most popular and relevant terms your target audience uses. 

Competitor Analysis: Analyzing the keywords used by your competitors can help you determine which terms they are targeting. 

Search Volume Analysis: This method estimates the number of searches for a particular keyword or phrase. Keyword Difficulty Analysis: This process helps you determine the difficulty of ranking for a particular keyword or phrase. 

Long-Tail Keyword Research: Long-tail keywords are longer, more specific, less competitive search terms. Identifying these terms can help you create more targeted content. 

Keyword Clustering: Keyword Clustering involves grouping related keywords together to create more targeted content.

Can Chat GPT Fulfill Your Keyword Research Needs?

Typical keyword tools like Keyword Planner, Ahrefs, and Semrush are powered by huge databases that can be used to help you find the best keywords for any given topic. With these tools, you can explore related keywords or even longer-tailed keywords that contain your initial keyword search.


These tools likely use a combination of algorithms that can comprehend your search query and its related concepts, as well as data that reveals what people are searching for. Combining these two elements, these tools can provide you with the most relevant results with data.

ChatGPT, unlike traditional keyword research tools, utilizes natural language processing (NLP) to comprehend the meaning behind your query and suggest related keywords. This means it can provide you with a different type of suggestion, but it also means that it’s limited in the sense that it doesn’t have access to data from actual searches. 

Despite the fact that AI is limited when it comes to providing you with Data, it still can help you with keyword research and analysis by skipping the work you need to put in to find those keywords and phrases relevant to your target search term. In addition, it’s handy to find and filter your keywords as you want just by sending a custom prompt to the AI and having your data set ready within seconds. 

Let’s take a look into how Chat GPT fits into different types of keyword analysis and research needs:

1. Discover New Keywords

ChatGPT is an incredibly versatile tool for keyword discovery. All you have to do is ask it for relevant keywords. You can then narrow your keyword search with audience specifications or add filters to your query. 

For example, you want to find out long tail keyword keywords related to copywriting. You can prompt it like “Make a list of keywords related to ‘copywriting‘ that one might search on Google.


Now that you have some major keywords, you can find some long-tail keywords targeting beginner copywriters. The prompt goes, “adjust this list of keywords related to ‘copywriting’ with beginners in mind.

copywriting’ with beginners in mind

Again, suppose you want to follow the programmatic content approach by duplicating the keyword style and changing the variables according to your need. In that case, Chat GPT makes this entire process looks effortless. 

For instance, you can ask the AI to {Make a list of keywords by changing the state name for all states in United States ‘Best Copywriting Agency in California’}, and it will give you the exact list you want without you having to look for all the state names as the variables.

programmatic content

When brainstorming new keywords, Chat GPT is among the top tools even though it misses out on one key component, volume, which SEOs use to determine the keyword’s worth. 

It’s faster than most of the mainstream keyword research tools in finding out keywords and is highly flexible, and has great room for your creativity.

2. Competitor’s Content Analysis

Unfortunately, ChatGPT does not let you just input a competitor’s URL into a tool like Ahrefs to get a list of keywords they are ranking for. The only way to do a competitor analysis is to feed the text from their page into ChatGPT and ask it to identify the keywords.

This sounds like time-consuming work, and the output would still not be accurate. In addition, you will miss out on metrics like keyword competition, volume, and current trends. 

Regardless, you can still use ChatGPT to extract a list of long tail keywords from a selected text like this example; you can ask the AI to make a list of 20 long tail keywords from the following text: “{copied article}.” I know it sounds like a lot of work, and some tools can do it much better than Chat GPT, but Chat GPT is not a dedicated keyword research tool either.

We copied a full article about a streaming application called Kodi from a random webpage, and chat GPT gave us a keyword list that was not really impressive on what we wanted to find out: 


So we asked if Chat GPT could recheck this. The revised list is actually pretty impressive and closely related to what we gathered from Ahrefs and Semrush on this topic.

revised keyword extraction list

It would not be worth your time to use Chat GPT for competitor analysis. Instead, you should stick to the traditional keyword research tools because they offer more valuable information. In addition, this process is quite time-consuming, and keep in mind that there is a limit to the amount of text you can copy and paste at once before ChatGPT gives you an error message.

3. Search Volume Analysis

ChatGPT (March 14 version) can provide you with keyword volume estimation from Google Keyword Planner. Although Chat GPT is not designed to provide this kind of information, you can ask the AI to provide you with a set of keywords with their respective volumes and competition data. 

It looks like Chat GPT can access or read data from Google keyword Planner; although, when I asked the AI about geotargeting it indicated that the data represents Global search volume, which is what we will cross-check shortly.

If you ask, “Can you provide me keyword volume and competition data for 10 long tail keywords related to this term {Copywriting}” – here’s what Chat GPT gave us:


When I attempted to verify the keyword volume and competition using the keyword planner, I found that the data didn’t exactly match what ChatGPT provided. 

We’re not entirely sure how ChatGPT is obtaining these keyword volumes, but it’s possible that different tools and algorithms may yield different results. Similarly, the competition data didn’t match my findings on the keyword planner, but the competition information corresponded for certain countries.

keyword verification on Google Keyword planner

4. Long Tail Keyword Research

You may have noticed that I requested the tool to generate a list of “long tail” keywords, and it delivered exactly that. However, it’s important to note that the accuracy of these keywords in terms of matching real-world search queries from people cannot be guaranteed. 

Nevertheless, the tool provides a lot of relevant output that you can actually use to guide your content creation process. You can instruct the tool to help you write and piece together information during the content creation time based on the keywords it suggests.

In that case, Chat GPT is a faster and easier solution than other keyword research tools just to find some potential long-tail keywords and blend them into the content so your output is more relevant if we consider content creation with AI support. 

To find long tail keywords about any topic, you have to pass the prompt as “make a list of {count} long tail keywords for this term {your term} where minimum word count per keyword is 5”, setting up word count as a filter here to avoid shorter keywords.

And here’s what we got:

5. Keyword Clustering

When grouping similar keywords or phrases into clusters based on their semantic relevance or topical similarity, Keyword clustering is the technique to go for. You can create more focused and relevant content around specific themes or topics by grouping related keywords together. 

This helps you to create organized content and a well-structured website that benefits the visitors to find what they’re looking for in the site website. In addition, focused keyword clustering can increase click-through rates across the content because of their relevancy. 

By grouping related keywords together, it becomes easier to identify gaps in your site’s content, making it easier to target new keywords.

Keyword clustering is easy with tools such as Ahrefs and Semrush, but with Chat GPT, it’s faster and has more freedom to tweak it the way you want. For example, you want to create a cluster of informative and commercial keywords and their volume for your main term, ‘Copywriting.’

We asked Chat GPT: “Make a list of long tail keywords for Copywriting and cluster them based on keyword search intent as well as include search volume.”

Make a list of long tail keywords

After we provided the AI with a list of long-tail keywords related to copywriting, we requested it to present the data in a table format. The table can be very useful in planning your website’s content without manually extracting keywords from various keyword research tools. 

However, it is important to note that the accuracy of these keywords may not be entirely reliable, so it is important to exercise caution when using them as the basis for your content planning.

keyword with table

And YES, you can easily copy and paste the table created by Chat GPT to your preferred text editor or software. This means you can quickly generate any table you need by simply providing the relevant data as a prompt and asking the AI to create the table. It’s that easy!

6. Chat GPT Prompts for Title Creation

Titles are important in case you want to guide your content production team to have a list of Titles per keyword with AI assistance; you have to ask Chat GPT to include Titles that are optimized for SEO and convey the catchiness and content intent.

We asked Chat GPT to add a catchy article title for all these keywords in the same table that matches the keyword in the title.

Ultimate guide to copywriting keyword on chatgpt

There are so many fantastic opportunities to explore with Chat GPT! You can use it to research, plan, create and verify your content production, which is incredible. Of course, if you’re looking to create content using AI, it’s essential to be careful and avoid any issues with duplication or plagiarism. 

That’s where Originality.AI comes in – it’s the perfect tool to protect your content from being detected as AI-generated and prevent accidental plagiarism with other content on the web.

We’re working on putting together a list of prompts you can use on Chat GPT to take your content production to the next level with AI support. So stay tuned to to get access to this excellent resource! And while you’re there, check out other articles and information on online business development. It has many ideas and insights to help you build your next idea.

Read More

10 keyword tools for YouTube To Increase Traffic [2021]

The keyword tools for youtube, also known as a keyword research tool, are the foundation tools used to discover new keyword opportunity for YouTube videos.

Read More

Content Writing vs Content Marketing – Case Study

Content writing is one of the steps that has to happen within content marketing. Content marketing consists of a series of steps, including business strategy, keyword research, content writing, publishing, and marketing the created content. Too often, the process of content writing gets mistaken for a business doing content marketing.

Read More

Does The Holy Grail for Content Marketing Exist Yet? – CognitiveSEO Review

What is the “holy grail” for Content Marketing? The answer is:  The ability to predictably say that if you target a keyword with a specific criteria, and produce an article that meets certain specifications, you will be “guaranteed” to rank in the top 10 of Google.  

Today we look at how CognitiveSEO can help in this quest!

The “Holy Grail” of Content Marketing might be a long shot, but we are always using our unique position of publishing A LOT of quality articles on great sites with Content Refined to try and come closer to understanding the variables that will rank your content with certainty on search engines. With the sheer number of articles published, along with access to data science talent, we were able to perform a multivariable regression analysis that we think it is worth pursuing.

In this article, we are going to talk about the results from one specific tool that can provide more metrics about both the keyword and content quality compared to almost any other tool on the market. 

CognitiveSEO provides an amazing array of metrics for data driven content marketers to dig into and analyze against the performance of their content. Where you would need data from several different tools such a SEMRush, MarketMuse and Ahrefs etc., CognitiveSEO consolidates a lot of that functionality into one super awesome tool. A great solution for those who have too many ongoing SAAS subscriptions already (I’ve been so guilty of this). 

So by using just CognitiveSEO, what are the metrics that would “guarantee” a ranking in the top 10? Come on a stats adventure with us.

Indiana Jones And The Last Crusade GIF - Find & Share on GIPHY
GIF – Indiana Jones

Stats Disclaimer – Before we get dig too deeply into these numbers, it’s important that we share what some of the statistical performance was for these. 

The Keyword Monthly Volume, Content Performance, Article Word Count and URL Character length are the most valuable variables to analyze. Analyzing  50+ regression scenarios,12.42% r square is the best regression result available using a majority of only CognitiveSEO’s Variables. The best single regression variable, content performance, achieved an r squared value of 4.4%. In most successful multi regression analysis, a variable ” anchor” is needed to drive line of best fit. This seems to be missing as the most successful single regression variable only has 4.4% correlation to google rankings, making it hard to fit the rest of the data to variable with little fit. Actionable insights are still achieved and are explained below. 

Actionable Insights: 

If you are trying to create content that will predictably rank on google, according to CognitiveSEO, you want to follow these guidelines. 


Minimum Conditions:

  • Keyword difficulty should be at least 17
  • Content performance score should be at least  51
  • The readability score should be at least 47
  • Minimum article length should be 250 words

Ideal Conditions: 

  • Keyword difficulty should be at least 17
  • Content performance score should be at least 85
  • The readability score should be at least 65
  • Minimum article length should be 1400 words

I checked to see if one of my best ranking pages met this criteria, and check out my findings below! 

My article “How to Build a PBN” is a great example of this. This article does really well–it has 10 keywords ranking in the top 3 spots on Google and it has been picked up for Google’s featured snippet for this article. 

In this case all of the “Ideal” Criteria were met!

  • Keyword difficulty should be at least 17
    • For my Keyword it was 37
  • Content performance score should be at least 85
    • Our content performance score was 97
  • The readability score should be at least 65
    • Ours was 75
  • Minimum article length should be 1400 words
    • We went big with this one and it was 14,743 words long! 


So what does this mean? Have we “cracked the code” of content marketing? According to this data analysis (which has pulled on hundreds of articles written by Content Refined) we are able to predict that if the ideal conditions of an article are met, you should be able to rank in the top 10 of Google. We’ve given you the Holy Grail–so give CognitiveSEO a try for yourself and really step up your game in the world of content marketing.

Read More

SECockpit Review New Features – Reviewed by My Content Marketing Company

Between my portfolio and the work that ContentRefined does we end up using LOTS of tools. In the past couple of weeks we have been in talks with Sam the CEO and founder at SECockpit reviewing the new features of SECockpit.

Learn more about Content Refined in this deep dive review of Content Refined as a business.

Maddie and I did some free training with Sam at SECockpit that I hope that will be of value to people – watch it here.

Intro: Who we are (Content Refined)

  1. How we use keyword research
  2. The way we used SECockpit in the past (including sharing our procedure!)
  3. New features of SECockpit
  4. How this will improve our systems

Final thoughts: Who benefits from the new SECockpit functions? (And more promo details)

Intro to Content Refined

If you’ve been following this blog for a while, you know a few things about me:

  1. I love data.
  2. I love to systematize procedures.
  3. I love using that data to refine systems and procedures.
  4. I love winning online.

At Content Refined, I think it’s pretty safe to say at this point that we’ve got content creation down to a science.  We’re on top of trends, including what’s relevant in digital marketing now, what clients want, what clients don’t know that they want but actually need, and the data behind how it all works.

We’ve tested a lot of different keyword tools over the past few years even buying one and bringing it in house; we’ve used SEMRush, Ahrefs, KeywordFinder, CognitiveSEO, MarketMuse, SECockpit, LongTailPro, and more.  Of those ones, we’re still using SEMRush, Ahrefs, MarketMuse, and SECockpit on a daily basis at Content Refined. They’re integral to our systems.

Today’s focus is specifically on SECockpit, because they’ve just updated a ton of features and our teams are super excited about some of them.

  1. How We Use Keyword Research

Keyword research is the first step in our content creation process.  We look for opportunities within our clients’ niche to rank for certain keyphrases.  There are a few metrics we generally stick to (bear in mind that these are flexible site-to-site depending on where you’re starting from with your sites relative strength):

  1. Monthly search volume of over 500
  2. Competition score of under 30%
  3. Average domain authority of top competitors is under 40

If our keyword matches all of those metrics, it’s probably a good opportunity to build an article around.

Over the years, we’ve adapted additional strategies depending on our clients’ needs.  Some examples of these strategies include the following:

  • Leveraging MarketMuse’s new ‘Research’ and ‘Questions’ functions to build relevant article titles that can compete with the existing top results and also perform with voice search results (i.e. Siri and Alexa)
  • Leveraging MarketMuse’s ‘Word Count’ functions to quickly assess how long our articles need to be in order to compete
  • Leveraging SEMRush’s ‘Competitor Analysis’ to find opportunities from competitors’ ranking articles
  • Leveraging Ahrefs’ ‘Also rank for’ and ‘Questions’ lists to seek new phrases that are closely related to the niche

It’s good to have all of these tools in our wheelhouse so that we can compete to the best of our abilities with any given website.

One shift in thinking about competition for keywords is to move beyond just looking at the number the KW research tool provides but also analyzing the strength of the competitions articles. If all metrics say a keyword will be easy to rank for but the top 10 current articles are all beasts written by experts and 5000+ words long…you are going to have a tough fight regardless of how “low competition” the tools saw the keyword is!

2. The Way We Used SECockpit in the Past

It’s a real process, so here’s our actual internal SOP that we’ve been using for completing keyword research with SECockpit.  It involves having access to SEMRush as well, but it’s not completely necessary to have that access.  You can do your research for your website with just SECockpit.

Basically, we search for a fairly general keyword, then filter the metrics to suit our needs.  We go into each keyword that looks like a good option to assess the top competitors’ Domain Authority.  If necessary, we’ll run an audit on a competitors’ website through SEMRush as well.

Here’s what it looks like when we search for a keyword in our client’s niche using the old version of SECockpit.  In this example, I’m looking for phrases like “travel backpacks” and “suitcases”, but this of course can be custom tailored to your specific niche:

SECockpit keyword research review

So you can see that it’s a little outdated looking.  That’s okay, though. We check off all the necessary boxes and begin our search.

The results can take a while to come up, so just be prepared for that.  Usually our Project Manager or Keyword Specialist (whoever is doing the keyword research) will have some other work to do in the meantime while they’re waiting for results.  I’d give it a solid 5-10 minutes to fully search, update, and load.

Here’s what it looks like when the results come up:

SECockpit keyword research review

We would also apply our specific metrics to get to the ideal monthly search volume and competition score.

From this list, we can go in keyword by keyword to see more details:

SECockpit keyword research review

I’ve highlighted the Domain Authority list in red there.  We want to see that the majority of those sites have a low-moderate domain authority.  An average of about 40 or below should be a good option for us.

From there, we can create a title based around the keywords that suit our metrics.  In this case it could be something like, “Best Travel Backpack with Wheels for 2019 Adventures”.  We would drop all the information into our Client Dashboard. When our keyword research is complete, we send it to our client to see if they want to change anything up before it goes over to the writer.

Here’s an example of a content dashboard that you can use with your teams.  It’s pretty simple:

content strategy system content refined

So, now that I’ve walked you through that existing process, it’s time to share the exciting new updates that SECockpit just came out with.

We’ll have to update that SOP for the new SECockpit features, because I already know it’s going to simplify our lives a lot.

3. New Features of SECockpit

Right off the bat, it’s clear that they did a ton of updating to the overall layout of the platform.

Here’s how that main page looks when you’re searching for a keyword:

SECockpit keyword research review

You can see it’s a lot cleaner looking.  All the functions are the same on that pop up, but it doesn’t feel so overwhelming and cluttered.

The other big thing design-wise is that they’ve made it mobile-friendly.  Here’s a screenshot from my phone:

SECockpit keyword research review

For digital marketers who are constantly travelling or running to different conferences/meetings, this is huge.  You can quickly pull up SECockpit on your phone and find your data that way. Really impressive.

Now, here’s the most exciting part to data nerds like me.

Here’s how it looks when those keyword results come up in the new platform:

SECockpit keyword research review

Notice anything different?

There are a few things.

  1. You can move the metrics around based on what’s most important to you.
  2. You can get rid of any metrics that are irrelevant for your purposes.
  3. There are new filters that are going to change our internal systems at Content Refined.

If you click on that little gear in the top right, all of these new options come up:

SECockpit DA avg DA max keyword research review

For our purposes, those ‘DA Avg’ and ‘DA Max’ filters under the Off-Page options are amazing!

Now, instead of having to go into each keyword one by one, we can actually just filter the Domain Avg. to be less than 50!

SECockpit DA avg DA max keyword research review

Game changer.  Now we have a list of keywords that meet all of our desired metrics.  Monthly search volume, competition score, and low-moderate domain authority.

Some of these other filters are really exciting for us, too.  Did you notice they now have Adsense Value? This is going to be huge for a lot of our clients:

SECockpit DA avg DA max keyword research review

I also really like some of the content specific filters like “Title Keyword”, which shows how many of those top 10 competitors have the exact keyword in their article title.  This gives us an edge in content creation:

SECockpit DA avg DA max keyword research review

Clearly, some massive improvements have been made here.  This was already a great tool for us, but now, especially with the Domain Authority and Adsense Value filters, we’re going to be able to deliver so much more relevant data to our clients.

4. How This Will Improve Our Systems

Our teams will still go in and look at each keyword individually.  It provides more insight on our competitors for each keyword.

Here’s what it looks like when you go into an individual keyword in the new platform:

SECockpit DA avg DA max keyword research review

We see the exact information for each of our top competitors.

We can also see the Google Interest Trends for our keyword over the previous 12 months:

SECockpit keyword research review google

With this information, we can make the best decisions for keyword opportunities for our clients.


If we wanted to pull some quick data and give our clients a bigger overview of a bunch of keywords at once, it is now so much easier to do so.  By adding in those extra filters, we can pull data straight from those results into a spreadsheet.

It would look something like this:

SECockpit keyword research review

Right away, we have a list of 20+ viable keyword options.  That’s just from one keyword search!

Now, keep in mind that these results will vary from site-to-site.  Not every niche is going to have great keyword opportunities with one search.  Sometimes you have to get creative with it. Find an angle that works for you, and/or flex the metrics until it’s more appropriate for your niche or field.

Final Thoughts: Who Benefits from the New SECockpit Functions?

Well clearly, we do.  I think that was made obvious.

But seriously, these are huge improvements for anyone who really wants to more predictably rank in Google.  Through using this real data, you gain so many advantages over your competitors. A lot of people will blindly produce content that they think is relevant to their audience, but they don’t actually have the real insights to show what content is necessary for their niche or field.

If you invest in consistent, quality, data-driven content marketing, you will see the results.  The measures that we go to at Content Refined have proven that time and time again. It’s even working for our internal marketing efforts, as you can see here:

SECockpit keyword research review

Consistency and quality are key when it comes to content marketing.

So, definitely check out the new features of SECockpit and let me know your thoughts.  I want to hear feedback from all the data nerds like myself.

SECockpit Offer:

Also, insider tip: They’re actually doing an exclusive offer that’s only valid until the end of March.  April 1st, prices are going up. There’s a 30 day no-risk guarantee, and once you sign up you’ll be locked in at this lower rate for life (unless you cancel at any point).  Take advantage! Here’s the link they sent me.

ContentRefined Offer:

If you have any questions about our processes or want some insights on how we can help your content strategy, reach out to our Co-Founder and CEO, Madeleine Lambert.  Her email is, or you can book a quick virtual meeting with her here.

Read More