By Jey Pandian published July 31, 2013

Create Precise Buyer Personas With SEO Data: A 10-Step Guide

buyer personas-targetIs your content underperforming? If so, your buyer personas may be to blame. Below I’ll show you why, and teach you how to build precise personas that will help your content hit the mark. 

Why content fails to resonate with a target audience 

Many marketers are disappointed when their content falls short of their expectations. And when it does, they usually look for a scapegoat: Maybe it was the writer? Or maybe it was the design of the piece? Or maybe it was the title?

But content — whether it’s for a website or other asset — often fails because it is built upon faulty buyer personas that, in turn, create a weak foundation. That’s because many marketers develop their buyer personas based on a hunch or guesswork.

The problem with conventional buyer persona development 

The conventional methodology used for persona research mines a variety of tools, such as the Google Keyword Tool, Google Trends, Google Insights, social media monitoring, and demographic and psychographic sources. When combined, this information helps marketers gain clarity about their target demographics.

The problem with this approach is that none of this information is definitive. Rather, it suggests possibilities — in fact, numerous possibilities. Because of that, persona development tends to be based on guesswork.

For example, let’s say research indicates that we have a female demographic whose household size is 3+, with ages from 25–54, and a total household income in excess of $300,000.

What might we surmise from this data? How might we define our buyer personas?

One content strategist might define the target thusly: “Her name is Michelle, she’s married, and both she and her husband are lawyers.

But another option is that the research represents a business owner who brings in that $300,000 alone, or that it is a combined figure that includes contributions from adult children who have temporarily returned to live at home.

Clearly, the research could be interpreted in numerous ways — which can lead marketers to make faulty persona decisions.

Precise personas are key to content success 

Wouldn’t it be great if you could precisely identify your target demographics for your content marketing efforts? Well you can, but to do so, the personas you develop need to be grounded in analytical data, not based on hunches.

To build precise personas, you need to dive into your website’s search data by using your web analytics tool(s). This will help you determine what people search for, why people search, how people search, and the actual searcher demographics. In addition, you need to go beyond the data insights collected in the conventional method, and actually integrate SEO into your persona research. Overall, the approach involves taking your website’s data and utilizes analytical rigor to deliver an effective content marketing strategy.

Sound like a lot of work? It is. But it has its rewards! In fact, this approach will help you identify every single persona that visits your website. It will also help you quantify the market size for each.

You can then use this information to feed your content strategy. Simply select the personas that you want to target — based on market opportunity — and then create the proper content with the appropriate maintenance schedule. And voila! You are on your way to producing more content that hits the mark!

The 10-step plan to build precise personas for your content 

The approach detailed here focuses on website content, but it can also be used to inform various assets, such as white papers, webinars, and eBooks.

Step 1. Download your website’s analytics data: The goal of this first step is to get a list of all the keywords/phrases utilized on your website by potential searchers.

To accomplish this, go to your analytics platform and download SEO keyword data from the search engines. Below are some process examples for two of the most popular platforms: Google Analytics and Omniture:

  • If you are using Google Analytics, go to Traffic Sources > Search > Organic, and set the correct date range in the upper right corner.
  • In Omniture, go to Traffic Sources > Search Natural Keywords, and download as much data as you possibly can. I like to get at least 1–5 years’ worth of data because the more information you have, the easier it is to connect the dots. 

Step 2. Establish themes: The goal of this step is to identify categories for your key phrases that represent different themes.

To start, take your big list of key phrases and read through it carefully. As you do, look for patterns and modifiers that will help you break down the data into themes. Study these key phrases and look for meaning behind the terms, such as key phrase intent. Your mission is to break up the list into small buckets of intent.

As you examine your list in this way, patterns will start to emerge from the data. For example, take a look at the following list of key phrases. How could you group the words together in a way that makes sense? (Note that the list is abbreviated for illustration purposes.)

key phrases-data

You’ll notice that some of the above key phrases have a city modifier, and that many of these are combined with other key phrases, such as a particular disorder or treatment. Given these patterns, we will group our key phrases into the following themes (buckets of intent).

data-three themes

Step 3. Group your key phrases into personas: The goal of this step is to pinpoint exactly who your targets are based on the definition of each key phrase.

Take your newly created list of themes, and start examining the key phrases in each grouping. Define the most likely intent for each grouping, and ask yourself if it is possible for a user in the group to search for every single bucket you’ve created.

If the answer is no, then create a decision tree for all possible user combinations. This will provide you with more clarity when you decide to build out your content.

For example, who do you think uses the terms shown in the themes above? The data indicates that there are most likely three main types of individuals, along with a fourth variant type:

  1. Individuals who are searching for a particular treatment in a particular city
  2. Individuals who are not patients who are researching for another individual
  3. Individuals who are potential patients who are researching a particular disorder
  4. Individuals who are concerned about the side effects of a particular treatment

But overall, there are really only two main types of individuals that are likely to use these terms: 

  • Persona #1: A potential patient
  • Persona #2: A researcher, such as an individual who might be researching information on behalf of a patient — like a patient’s parent or caregiver

Before you move on to the next step, be sure you have established your key personas, based on the information you have analyzed.

potential patients - data


Step 4. Identify the search funnel: This step is fun. Your goal here is to uncover the specific order of search engine queries that are utilized by each of your personas.

To accomplish this, take the buyer personas you created in the previous step, and cross-reference them with your key phrase list. Now see if you can determine the first, second, third, (nth, etc.) search for each persona. (Bear in mind that search engine users will initially search for the most generic version of their key phrase before drilling down and using specific language to find the object of their search.)

This step will help you achieve two objectives:

  1. Figuring out how many potential customers are present at different stages in the search journey
  2. Identifying the kind of content needed to move them to the next stage, and ultimately, to becoming a customer

For instance, the individual search funnel in the ‘potential patient’ persona shown in the example above might resemble the following:

individual search funnel

Step 5. Calculate the opportunity: Your goal at this point is to calculate the potential opportunity for each key phrase in the search engines, should you decide to build out your content based on the data you have just compiled. To do so, you’ll need to create a spreadsheet like the one far below.

calculate the opportunity

  • To start the process, take a look at the exact match volume for your key phrases for a given month. This is the local search volume taken from Google’s Keyword Tool.
  • Next, you need to normalize the search traffic based on Google’s market share. To do so, divide a key phrase’s exact match volume by 0.66. This will give you the normalized volume.
  • Now multiply the normalized volume by 0.4. This will give you an estimate of the search traffic for a Position 1 result if you were to rank for the target key phrase.
  • Enter the actual search volume — you can take it directly from your analytics platform. Just fill in the blanks in your spreadsheet for each key phrase for a given month.
  • Potential opportunity can then be calculated by subtracting the actual traffic volume from the normalized volume shown in Position 1 (Column 4). Fill in the cell with “not applicable” if this number is negative.

Once you complete your spreadsheet, you’ll be able to identify your best areas of opportunity. For example, in the table below, you can see that of the nine key phrases shown, the largest opportunity lies in Row 2: “laser hair removal san diego.” The other key phrases have some volume, but not much, so we would cut them from the personas in Step 3. 

Step 6. Define your competitors: Now you need to turn your attention to your competitive landscape. Your goal in this step is to create a competitive snapshot of the first page in Google, Bing, and Yahoo (or other search engines you might use) for your search terms.

To start, take your high-volume key phrases and target terms that convert into leads, and load them into a rank tracking tool like Conductor. (Alternatively, you can manually Google them, but be sure personalization and web history are turned off so you get non-personalized search results.) How does your competition rank for each of your target key phrases?

In addition, you need to do a little research on your SEO competition to get a sense of their characteristics: Are they mom-and-pop shops? Are they big chains? Are they local business listings, like Yelp or Foursquare?

Detail your findings on a spreadsheet — it should look something like the example shown below. (Note: There’s no need to include any competitors that aren’t ranking for your top terms.)

You can see that a lot of results for the term “laser hair removal san diego” are dominated by Google+ Local, followed by Yelp, and individual websites. You can also tell that it is possible for a Yelp, Facebook, and Yahoo! Local listing to all rank on the first page.

competitive landscape review

Step 7. Create a competitive matrix: Your goal in this step is to see how your own website’s content stacks up against the content on your competitors’ websites.

To make this happen, visit the websites of all the competitors you uncovered in Step 6, and take a note of every type of asset class on their sites, such as FAQs, videos, articles, About Us pages, etc. What do you see in terms of commonalities? What type of content have your competitors created that you do not have on your own website? In other words, where are you behind the curve?

Plot your findings in another table (as shown below). Here, Row 1 shows all the different types of content that you would find if you visited each of these websites. Row 2 compares your website’s content to the websites that rank in the Top 10 for “laser hair removal san diego.” The green cells indicate that the content is present; the red cells indicate where content is missing.

competitive matrix

Step 8. Perform a website content audit: Your goal in this next step is to take inventory of your existing website content. This will help you integrate your SEO and content strategy. A basic way to do this is to create a spreadsheet like the one below.

website content audit

Simply list each asset class that can be found on your website, give it a unique identifier, and classify it by type of asset (PDF, video, images, audio, etc.). In addition, include a column where you list the URL of each piece of content, so you can make note of where it is located on your website. You can also add an ROT score (redundant, outdated, or tired) for each asset. 

Step 9. Look at the big picture: Your goal in this step is to combine the information from Steps 7 and 8 so you can identify weaknesses in your competitors’ content strategies. (Note that while this step is very time consuming, it will be well worth it in the end!)

Take the competitive matrix and your website content audit and merge them together to produce one table. It should be categorized by the different asset classes you identified in the website content audit. An example of the merged spreadsheet for the “Treatment Type” asset class is shown below.

big picture

Now repeat the process for each of your different asset classes (in our example, these would be: Tattoo Types, Removal Process, Reviews, FAQs, About, Before & After, and Maps). 

10. Prioritize your opportunities: This is the most exciting step! Your goal here is to figure out where you can get the most value when creating content for your website, and to set your content marketing priorities based on this information.

  • To start, create a spreadsheet categorized by your personas.
  • Now add in the asset classes and competitive information from Step 9. Your asset classes should align with each persona. For example, if you refer to Step 6 you’ll note that the key phrases listed for the Potential Patient persona are all about types of treatment. Given that, the Potential Patient persona aligns with the asset class, “Treatment Type,” as shown below.
  • Now enter the opportunities you uncovered for each persona earlier. This is done by totaling the numbers found in the “Potential Opportunity” column from Step 5. For instance, the table below reflects an opportunity of 261 visits for the Potential Patient persona, and five visits for the Researcher persona.
  • Once you finish filling in the table, highlight your best opportunities. But be sure to focus on the low-hanging fruit by prioritizing those opportunities that will drive the most visits with the least amount of work.

prioritizing your opportunities

So for our example, which would be the easiest opportunity?

The Potential Patient persona shows that “” has a little bit of content on the topic of Treatment Type. However, the green cells indicate that four competitors have content in this area as well, and they seem to have the advantage. But note that the content for two of those competitors is outdated or redundant. Whereas the Researcher persona shows that four competitors have relevant targeted assets for the topic of Patient Education, but “” has none.

Should you reinforce your existing topic (Potential Patient persona) for 261 visits, or fight an uphill battle against your competitors (Researcher persona) for a measly five visits? The answer should be obvious: Pick the Potential Patient! It doesn’t make sense to target Treatment Type for the Researcher persona because you would have to create assets that are better than what your competitors have, and you would only get five visits for your effort.

Note, though, that even if a key phrase or asset class has less than 100 visits per month, you should still take a look at your analytics and see if it converts. If the answer is yes, it might make sense to build out content for this phrase to help you capture and convert prospects. However, if the data shows it does not convert, you would want to ignore this traffic.

SEO Pro Tip: You don’t necessarily want to pick the personas that have the highest search volumes, such as terms with more than 10,000 visits, as these will often be extremely competitive and difficult to execute. Low-hanging fruit means a focus on search volume below 500 exact match searches per month. 

Congratulations! You have now successfully integrated your SEO with your persona research and your content strategy. Now it’s time to put it all to work. Simply create a timeline and start executing your priorities!

How have you used personas in your content marketing? What methodology did you use? Share your tips here. Find this article helpful? Please share it!

For more information on creating buyer personas, read CMI’s audience-focused how-to guide. 

Cover image via Bigstock

Author: Jey Pandian

Jey Pandian is an Organic Search Director for Catalyst, a leading search engine marketing firm. Jey has over a decade of SEO experience, and oversees SEO strategy and implementation for Fortune 1,000 brands. Previously, Jey was the SEO lead for, and managed SEO operations for large enterprises, including Samsung Electronics and Nokia. Follow him on Twitter @jeypandian.

Other posts by Jey Pandian

  • christian

    Excellent post Jay!

    One question tough, when you describe step #1, you mention about getting potential searchers from GA, my question is since we are talking about “potential traffic” wouldn’t be also beneficial to get data from WMT based on impressions? people that are interested in the service your are selling but didn’t arrive to your site (missed opportunity).

    • jeypandian

      Hello Christian,

      Yes, you can. Ideally – the more data you can get your hands on, the better and more cohesive your strategy.


      • Morlette

        Can you send me the template as well?


  • Rajiv

    One of the best post. Can you send me the template used to develop persona by following the 10 steps

    • jeypandian

      What’s your email Rajiv?

      • Kashif

        I’m interested in the template too 🙂

        • Manoj Juyal

          This 10 step guide seems interesting. I would like to try it. It would be great if you can send me a template Jey.

  • Bryce Propheter

    This has almost nothing to do with personas, yet the title is how to create precise buyer personas? What’s precise about them? “Potential Buyer” is about as generic as you can get. The point of personas is to understand your customers deeply so you know what content they need.

    I agree with your intro about false assumptions leading to bad content decisions, but your 10-step plan doesn’t get the job done, either.

    Will this help your SEO efforts? Yes. Is it good advices for people wondering which keywords to target? Yes. Is it a good practice? Yes.

    But it is not a “10-step guide to creating precise buyer personas.”

    • jeypandian

      Hello Bryce,

      The precision stems from the use of Analytic data to drive the creation of Personas. When we look at organic keyphrase data, we gain insights into the mind of the customers and this in turn, helps inform the content creation.

      Does that make sense?


      • Bryce Propheter

        But that does not make a persona. It’s a general target audience, at best. You have not identified any way to actually know things like the age, education level, occupation and demographics of the people searching for these phrases. What you’ve identified helps with SEO efforts only, not marketing and business development on a greater scale, like personas are designed to do.

        • jeypandian

          It does make a persona because the data is from terms that people use to “search.” Who is more relevant than your website’s audience?

          Taking a note from the same vein, who is the best indicator of your target demographic other than your website’s audience?

          Is it Nielsen, Kantar and media sampling data? Or is the real gold in your own website’s data?

          The keyword data is set in stone, you can’t argue with it. For the other mentioned items – where’s the proof? You can estimate that X people are located here and Y people are this age but these figures are based off data from other websites – not your own.

          How do you trust this information to accurately generate personas?

          Also, what helps SEO impacts the entire marketing funnel including business development. Do you disagree?

          • Bryce Propheter

            I’m not arguing with you that pulling in “target demographic” data from Nielsen and Kantar is the way to build personas, either. I’m simply stating that what you have created is not a persona.

            A persona is a single person – a ficticious character created from REAL customer data that any business should have. There are plenty of great guides on creating marketing personas by doing USER RESEARCH (of which, keyword research is a part.)

            What you are creating is a conglomerate of multiple people labeled as “potential customer.” That doesn’t really help with bigger questions.

            For example, a fully fleshed out persona for a B2B company might describe the person’s other responsibilities, who they report to (are they the actual decision maker?), their research methodology and countless other things that helps the company better understand their customers.

            This information is retrieved from talking to REAL customers and is therefore just as accurate as search data (more so, actually, as search data doesn’t tell you how profitable the customer is that used those search terms.)

            The 10 steps here are a great starting point, but they will by no means help you create a fully functional buyer persona.

          • jeypandian

            Ah, in that case, I do agree. It is a starting point. I’m not saying use this as a solve all method but rather as a way to strengthen the preexisting persona research.

            However, on the search data front, one can see customer profitability at the keyword or individual level, most platforms can do it, but I will agree that real customer data trumps everything.

          • David

            I think it’s just semantics and a better term would be ‘search personas’.

  • aruegger

    Great post, very thorough.

    • jeypandian

      Thank you Andrew 🙂

  • Anne Ryan

    This is an interesting approach to creating content drivers (not so much personas) – kudos to the process and thought put into it. The only problem is it’s based off of the keyword themes extracted from Organic Search data in Google Analytics – which means you’re only using data based off of the keywords driving organic traffic to your site. To increase visibility and extend your organic reach, using organic search themes is a good place to start to see what’s working, but the next step should be keyword research to identify new keywords that will inspire new content drivers.

    At the end of the day, the process should be to first develop personas by identifying responsibilities and pain points; outline their buying journey and the kinds of topics and channels they engage; create a target keyword list by persona; and then a content playbook that outlines each persona, their target keywords, content topics, and channels for distribution. From there, you create an editorial calendar and you’re on your way.

    This allows you to always be working towards your audience and new opportunity, verses the process outlined in this blog, which is to build off of only what’s working. And if what’s working isn’t very much, then your digital market share will suffer.

    • jeypandian

      I agree completely. I used to work for a Sales company many years ago so that note about the pain point resonates with me. Themes are definitely the way to go for competitive terms.


  • Justin Howley

    Takes the sometimes vagueness of content marketing and really establishes a good practice of prioritization of interest. Great post.

    • jeypandian

      Thank you Justin 🙂

  • Donovan Panone

    I think this is a great process for identifying an SEO strategy and figuring out opportunity gaps you are missing out on. But it’s also based on traffic you are currently getting and misses out on traffic you are not.

    Also, keywords often point to “what” someone is looking for, but not necessarily “why”.

    BUT having said that, I think this can definitely be one valuable piece of many to help create a rich, accurate persona. This is real data, not theoretical “I think our audience really cares about X”.

    • jeypandian

      I agree. Thank you Donovan 🙂

      • Alessandro Gonzalez

        How can you reach the traffic you are missing out? That’s my question.

  • Ann Smarty

    Truly is a guide. Thank you for laying that all out for us.

    • jeypandian

      Anytime, thank you Ann.

  • Manoj Juyal

    This 10 step guide seems interesting. I would like to try it. It would be great if you can send me a template Jey.

    • jeypandian

      Manoj, please send me your email address – mine is available on the net 🙂

      • Manoj Juyal

        Thanks Jey. I’ve emailed you at your gmail ID 🙂

  • nathanjoynt

    Thanks Jey! This is a fantastic resource for every SEO and content marketer to use in their audit/strategy process. Bookmarked and shared.

    • jeypandian

      Thank you Nathan 🙂

  • Jakub Velicka

    Jey, maybe you could add the template for download at the end of the article 🙂

    • jeypandian

      That’s a good idea. I’ll see if I can add it. Thank you for thinking of it Jakub.

  • Carmen Rane Hudson

    This was intense!! I think I’m going to have to walk my way through it step by step on each website and do it to understand it better. Still, I really appreciate it. I’d always felt like “creating a buyer persona” seemed like an exercise in guesswork. I’ve done it, but I didn’t ever feel like I was basing it on much more than what I was trying to achieve. So! This post has taught me the most of any I’ve read today.

    • jeypandian

      Thank you very much Carmen 🙂

  • HiltonB1

    Jay – delightful and the type of thorough step-by-step most folks (mea culpa) seldom do…or worse, do infrequently. The notion of a quarterly review/refresh/update of the template seems astute.

    Add my name to the list of folks asking for the template.

    • jeypandian

      Thank you Hilton. Will do. Cheers

  • Tamar Weiss

    This is one of the best step-by-step practical guides I’ve ever seen for targeting your audience. Thanks for sharing!

    • jeypandian

      Thank you very much Tamar 🙂

  • Susan Oakes

    It is an interesting article. The one question I have is in the step where you advise to download your website’s keywords. With the increase in Keyword Unknown what do you advise to work through as it means you will not know what they are searching for when landing on you site. You can make an educated guess by looking at the landing page, but the exact words are not known.

    • jeypandian

      Hi Susan, The Keyword unknown/not provided should only be 50% or less – there should still be data available for you to see the words.

  • Bruce Washington

    Thanks for the posting for tips

  • Adriana Spencer

    I wouldn’t say the post has nothing to do with persona , as @Bryce Propheter argues. It starts with “persona” but from the point 5) downwards it looks like the plot has been lost. See the sequence: calculate the opportunity, define your competitors, create a competitive matrix, perform a website content audit, identify the weaknesses of your competitors, prioritize your opportunities. Nothing about “ the busy mom that has to cook faster, blog faster and get some sleep” ( sorry, just an example).

    I think the key lies in one of the lasts paragraphs:

    “You have now successfully integrated your SEO with your persona research and your content strategy.”

    This is what the article is about. From this angle , I find it excellent !

  • Tony Zambito

    What I see described here is something which can complement buyer persona development. But folks need to understand buyer personas provide a qualitative value which can help in developing a robust picture of buying behavior. How you described conventional buyer persona development is incorrect. The way you describe it naturally supports the rest of your article. It is mixing search analytics and profiling with the term buyer persona. The exercises you describe help to further understand buyer persona archetypes and insights but the purposes of buyer personas and buyer profiling are different.

    Buyer personas are qualitative researched based and based on real people qualitatively interviewed- not gathered solely from Google analytics mining and mining of demographics/psychographics as you mentioned. They serve the purpose of understanding oftentimes unarticulated and not-so-obvious patterns of insights – which I add – data and analytics cannot uncover. While the term buyer persona has gained popularity – it is also prone now to incorrect definitions and notions of what they are designed to do.

    I like your 10-steps – very good advice for content marketers. However, I see this as buyer profiling which should not be confused with buyer persona development. There is an important distinction and content marketers are served best when they do both well.

    • jeypandian


      It is clear to me that you possess a great deal more knowledge on this topic than me. What resources might I use to learn the fundamentals of your craft before studying the resources on your website to further my knowledge?

      Best Regards,
      Jey Pandian

  • Diaz

    Excellent article Jey!!

    You got me thinking.. now I need to tweak this a bit to fit my specific needs.

    I just wanted to add my 2 cents, for the step 5:

    I’m trying to develop a formula to calculate the opportunity, and so far I got this:

    (Priority * # of Searches * Term Trend) / (# of Visits * Competence)


    Can be a number between 1 and 10 and it dependes on the goal of your business… Could it be that you have some products that give you bigger earnings? Or a specific search that shows very clear signals of purchase intent?

    # of Searches and # of Visits are self-explanatory, however, the reason I multiplied by searches and divided by visits is that the percentage of visitors you get from each search its easily calculated by (# of visits / # of searches), and the lower the percentage the more you want to focus on that specific search

    Term Trend: Its a floating point number between 0 and 2. Low values like 0.3 and 0.5 tell us that the query has been decreasing in searches throughout time, and high numbers like 1.5 indicate us the opposite.

    Competence: A somewhat subjective index that can range from 1 to 10.

    What do you think?
    Any feedback would be really appreciated 😀