How to create a buyer persona with Google Analytics
People laughing

Written by Bartek Bezemer

I want to help you get more out of your online marketing by giving you insider tips and combine them with market trends to help you better reach your target audience.

May 24, 2020

The audience section in Google Analytics can prove a worthy source for learning more about our most valuable and loyal customer.

Google Analytics is a treasure trove of data, but I didn’t find myself frequently using the audience section over the years I’ve been using the tool. Yes, I used it for looking at how the divide was between mobile and desktop and get a global picture of the genders visiting my website, but I didn’t dive into it so deep that I would be able to create a buyer persona. But today, we’re going to challenge ourselves.  

In this post I’ll create a buyer persona for the Google Merchandise Store, which will be an interesting account for us to explore. This analytics account is freely available for everybody with a Google account. I like using this Analytics account since you as a reader can follow along and have a sense of how a richly filled Google Analytics account looks like. It’s difficult to analyze if you have limited data to work with. At least, that’s how I felt back in the day when I started out with new accounts or those with limited traffic. 

I’m going to break down the different steps I would be taking in real life to help you better understand the methodology used to create a buyer persona through your own account. I read a lot of posts just giving pointers, but how would a seasoned online marketer approach this task when given? That’s what I want to aim at and help you become more confident using Google Analytics for these purposes.

Why limit ourselves to Google Analytics?

There are a multitude of (free) tools at our disposal that we can use to form a more complete picture of our customers. But, not always is there enough resources available, may it be time or money, to perform such thorough research. Additionally, it’s very tempting to use other tools, whilst Google Analytics has such an extensive library of reports we could be using. 

As I said in the beginning of the post, I’ve overlooked many parts of Google Analytics over the years, since I only needed to use a few reports to cover my day to day tasks. Only when I needed a specific question answered did I focus myself on reports within the tool not used frequently. I realized there was so much data buried in there, you can spend days just lurking around one section. 

I’m going to use the Audience report which houses some very interesting reports such as Demographics, Interests, Geo and Technology. While those reports won’t give the complete, full-blown picture of our most valuable customer, we can get a very long way as you’ll see in this post. We can determine the age of our most valuable customer, where they live and through which device they purchase. Through this we’ll be able to get a view on how our customers interact with our website and our products. 

Limitations of Google Analytics

Of course, I’m not fully delusional about its capabilities. As we will see later in the post, a well formulated persona is more than just a set of data points. It takes into account abstract elements such as brand preference, favorite offline media channels, education and occupation. 

This data is unfortunately not available within Google Analytics, although some can be found in the Google Ads audience section, for this post we do not have this at our disposal, that’s why this post will be geared towards creating a high level view of our most profitable customer group. Believe me as I say, that even a global overview is more than most well established companies have. 

What is a buyer persona and how is it used?

First we need to pin down what a buyer persona actually is, and the answer is actually pretty straightforward. A buyer persona is the ideal customer for your business. A buyer persona is a blend of current customers based on data analysis and part fictional characteristics, all distilled into a personification of who she might be. 

We use a buyer persona to find our most lucrative customers and tailor a product and message that resonates with this audience across all our media channels. Creating a solid buyer persona helps you in structuring your content, funneling it to one core topic that steers towards one main goal you need to fulfill with your website. 

I hear you say, “But I have more than one type of customer.” Of course, you don’t have to limit yourself to one persona. It’s best to have at least two or even five buyer personas. This will prevent you from creating one message that only rings true to a fraction of your entire customer base. Having multiple persona’s will also help you identify which channels you have to add to your marketing mix to reach all of your desired audiences.

Remember that hoarding data and doing extensive research is great and all, but even with heaps of it, making meaning out of it is a major challenge. I’ve met people so skillful at hoarding data, but paralyzed when having to make a decision. A good buyer persona is able to move your customer from an abstract concept, a collection of numerical variables, into a person. A person like you and me. Somebody who is in search of a solution and seeks help. A person who will learn about your how, why and what. 

Another important note I want to give before we go into greater detail, is that while Google Analytics is free, you might find that without extensive knowledge of graphic design or not having a specialist in your team, creating assets for your newly defined audience can become an expensive endeavor. In no means do I want to scare you away from finding the right audience for you to improve your sales, but marketing is hardly ever free. 

What makes a good buyer persona? 

An adequate buyer persona for you to work with abides to a certain set of characteristics. A buyer persona consists at least of the following aspects:

  • Age 
  • Gender
  • Income 
  • Occupation
  • Location
  • Education
  • Interests 
  • Media consumption
  • Opinions
  • Pains 

Another important addition to all these traits is the customer journey. Having a customer journey is not mandatory to the buyer persona, but will help you a lot in further tailoring a message for each channel and optimizing your sales funnel for increased revenue. Doing a customer journey analysis requires another approach and different reports, which can also be found in Google Analytics. There are other tools out there from Google themselves which highlight different buyer journeys across industries and regions around the world. 

How to report the buyer persona? 

Before we go into detail about gathering data and creating the actual buyer persona, we have to look at how to report upon the buyer persona. A lot of good research can fall flat through its presentation. We want to prevent our presentation from being left untouched. The following topics should return in the reporting and presentation of your persona:

Problem definition
Try to be as specific as possible why the persona was necessary and its investment. A lot of times the persona will be considered as a bonus, but as you and I know the buyer persona is a vital part of marketing and makes or breaks any communication you are trying to set up. The problem definition can also be interpreted as the introduction part of your presentation. You are free to play around with the beginning stages of your report and presentation. 

Define the scope of the research. How many personas will be investigated and how will they be implemented within the company. 

State the sources you’ve used for your research. In this case it is Google Analytics, but you can also use your database, customer surveys etc. This will prevent unnecessary questions being raised, concerning the validity of your data. Mention the time period that is analyzed.

This will be the meatiest part of your report. You will state the different findings of your data. Data speaks for itself and proves you have done a thorough job and that the buyer persona is a result of carefully selected and interpreted data. You might find this boring to elaborate upon, but remember that a persona has no direct ROI and if you want to get your management on board, you have to show and prove to them it was worth the effort. 

Now you’ve come to the actual stage where you present the buyer persona. You want to turn this into an actual person and pull it out of the factual and into the realm of the abstract. The persona section should be approached as an interview. Let the persona speak for itself. It’s a blend of factual and fictional. How would the persona describe him or herself when asked?

I like thick reports, something I can dig into and pull my marker through. I love to be as elaborate as possible. But you might find you work best within more compact formats, such as a powerpoint presentation or be creative and invite the people who best represent your buyer persona and let them tell who they are or a video interview. You can go in any direction and create complementary documents. 

The core principle remains, that it should give you and your company enough insights into the ideal customer where multiple departments can benefit from. From customer service to sales, all should be able to understand your findings and be able to implement it into their daily routines. 

Creating a persona for the Google Merchandise Store

Through the Google Merchandise Store Analytics account I’ve found the following results for 2019. I took the whole year as it will give us a good overview of the most recent customers. I took new website users as my reference point to prevent double values. Remember there are a multitude of approaches for analyzing and segmenting an account, such as by revenue, by conversion rate etc. I want to tap into high traffic generators to create a global persona. 

We are going to lay the foundation for our persona in this exercise. That’s why I’m going to add multiple layers of characteristics to our metrics so you can follow along and see how our persona takes shape. I don’t want to cram all the data into one large table as you won’t see the different steps we have to take to boil it all down to a usable buyer persona. Unfortunately, we will miss some obvious aspects that are necessary for a full buyer persona such as pains and gains, but as I’ve said a global overview is just as valuable. 

There are a lot of regions on this mighty earth, so we’re going to focus on the top 10 most popular regions by new user count. We see the US, India and UK generate the most new users to the website.

CountryNew Users
United States244,638
United Kingdom29,596

Region divided by revenue
When breaking down the amount of new users per country, we do see that having a lot of traffic from a certain region, but that won’t translate into revenue. No revenue was made from France and Taiwan despite their relatively high numbers of new users coming to the website. The most valuable countries for us are the United States, United Kingdom and Canada. 

CountryNew UsersRevenue
United States244,638$ 58,311.35
India47,433$ 170.42
United Kingdom29,596$ 1,342.36
Canada21,162$ 1,453.32
Japan15,240$ 152.09
Germany15,073$ 57.38
France13,291$ 0.00
Taiwan12,461$ 0.00
Spain10,860$ 74.98
Australia10,211$ 52.77

Not entirely unsurprising is EN-US and EN-GB the most frequently used language. 

LanguageNew Users

We see that the biggest age group is between 25 to 34 years old, followed by 18 to 24 years old and after that, the third largest group is 35 to 44 years old. Beyond these age groups dwindle in numbers fast and will be less relevant for creating our most valuable persona’s.

AgeNew Users
18 – 2451,027
25 – 34115,669
35 – 4446,694
45 – 5418,087
55 – 646,851

Males are well, or overly, represented on the website compared to females. 

GenderNew Users

Gender divided by Age
We can see that males between the age group 25 to 34 years old are coming to our website in great numbers. This group is followed by the ages 35 to 44 and as third the group 18 to 24. 

Gender MaleNew Users
18 – 2431,853
25 – 3474,748
35 – 4435,653
45 – 5412,696
55 – 644,164

Females between the age group of 25 to 34 years old are most represented on our website, followed by 18 to 24 year olds and as third 35 to 44. 

Gender FemaleNew Users
18 – 2418,953
25 – 3440,639
35 – 4413,893
45 – 545,347
55 – 642,677

Gender divided by age and revenue 
As we zoom in on the revenue generated per age group, we see that the greatest spenders among males are the ages 25 to 34 years old, second the age group 35 to 44 and as third 45 – 44. 

Gender MaleNew UsersRevenue 
18 – 2431,853$ 480.14
25 – 3474,748$ 1,544.52
35 – 4435,653$ 1,117.87
45 – 5412,696$ 1,036.82
55 – 644,164$ 358.84
65+2,291$ 34.99

In the female group we see our biggest spenders are between the ages 25 to 34 years old, followed by 55 to 64 and 35 – 44 years old. The second age group, 55 to 64, is interesting as it is a group not well represented on the website (11.48%), but are responsible for 28.42% of the revenue in this gender group. Increasing the amount of users in this gender and age group would mean a healthy sales bump. 

Gender FemaleNew UsersRevenue 
18 – 2418,953$ 253.36
25 – 3440,639$ 1,003.81
35 – 4413,893$ 949.97
45 – 545,347$ 243.83
55 – 642,677$ 1,018.11
65+1,638$ 112.97

New users by device 
We keep running into surprises when analyzing the account. The most new users to the website come from desktop devices. I would have expected to have more 50/50 divide between mobile and desktop traffic. 

Device category New users
Mobile 176,624

New users by device by revenue 
Another surprise concerning the devices. We see that mobile traffic, despite generating 30.45% of the traffic, is responsible for 49,61% of the revenue. Desktop accounts for 37.92% of the revenue, but generates 70.25% of the traffic. 

Device category New usersRevenue 
Desktop389,093$ 24,025.64
Mobile 176,624$ 31,426.16
Tablet14,289$ 7,899.07

Device usage and revenue by gender 
I want to see how the divide is by device by gender. We might uncover that a particular gender likes to shop on a particular device. We see that both genders favor mobile devices as their platform of choice for purchases.  

DesktopNew UsersRevenue
Male125,077$ 997.22
Female67,519$ 324.25
MobileNew UsersRevenue
Male40,925$ 2,378.13
Female19,158$ 1,278.07
TabletNew UsersRevenue
Male2,499$ 1,629.67
Female1,646$ 495.55

We can already get some contours of our persona, who we will make two off, one male and one female. We can determine that our male persona is between 25 to 34 years old, lives in the United States and prefers using a mobile device. Our female persona is also between 25 and 34 years old, from the United States and uses a mobile device for purchases. 

You might wonder how come that despite the other countries generating masses of traffic, they don’t generate as much revenue. This can be due to shipping costs that outweigh the cost of the product. We also need to take into account that Google is a US company, which will obviously play a role in who will visit the website. This is beyond this report and it might be true that when optimizing the shop our persona will change over time. Same can happen when adjusting the marketing strategy. 

In-Market segments by gender
The in-market segment will show us the interests users display when browsing related products. We can go for the affinity category, which is a step higher in the funnel, but these groups are harder to convert and are more suited for generating low intent traffic. 

We see that from the male group, the most new users come from the in-market segments Software/Business & Productivity Software, Consumer Electronics/Mobile Phones and Employment

In-Market SegmentNew Users
Software/Business & Productivity Software44,778
Consumer Electronics/Mobile Phones44,563
Financial Services/Investment Services33,292
Business Services/Business Technology31,796
Business Services/Advertising & Marketing31,643
Travel/Hotels & Accommodations29,814
Employment/Career Consulting Services30,115
Travel/Air Travel27,105
Business Services/Business Technology24,954

For the female group, the most in-market new users come from Employment, Travel/Hotels & Accommodations and Business Services/Advertising & Marketing. Although the difference is more evenly spread, compared to the male counterparts. 

In-Market SegmentNew Users
Travel/Hotels & Accommodations22,539
Business Services/Advertising & Marketing21,413
Software/Business & Productivity Software20,164
Employment/Career Consulting Services19,374
Travel/Air Travel17,251
Apparel & Accessories/ Women’s Apparel16,011
Apparel & Accessories 14,892
Business Services/Business Technology13,315
Home & Garden / Home Decor12,358

In-Market segments by gender by revenue
We’re going to take a deeper dive into the in-market segments, but just like we saw with the revenue divided by age, the amount of new users doesn’t tell the complete story yet. 

We see a pretty equal divide of revenue by the in-market segment in our male population. Nonetheless we can take some pointers from the new users group. The top three in-market segments by revenue in our top 10 segments by new users is Employment, Consumer Electronics/Mobile Phones and Employment/Career Consulting Services. 

In-Market SegmentNew UsersRevenue
Software/Business & Productivity Software44,778$ 703.56
Consumer Electronics/Mobile Phones44,563$ 753.02
Employment42,708$ 829.52
Financial Services/Investment Services33,292$ 642.70
Business Services/Business Technology31,796$ 339.59
Business Services/Advertising & Marketing31,643$ 285.86
Travel/Hotels & Accommodations29,814$ 548.63
Employment/Career Consulting Services30,115$ 734.21
Travel/Air Travel27,105$ 707.66
Business Services/Business Technology24,954$ 270.99

In the female group we see a clear winner, and not by a small margin either. For the amount of revenue generated the Travel/Hotels & Accommodations in-market segment is most lucrative. This segment is followed by Apparel & Accessories/ Women’s Apparel and Software/Business & Productivity Software.

In-Market SegmentNew UsersRevenue
Employment25,232$ 391.60
Travel/Hotels & Accommodations22,539$ 1,624.47
Business Services/Advertising & Marketing21,413$ 169.36
Software/Business & Productivity Software20,164$ 633.00
Employment/Career Consulting Services19,374$ 498.96
Travel/Air Travel17,251$ 518.77
Apparel & Accessories/ Women’s Apparel16,011$ 732.99
Apparel & Accessories 14,892$ 409.37
Business Services/Business Technology13,315$ 506.99
Home & Garden / Home Decor12,358$ 483.13

From our in-market segments we can determine that there is a mixture between leisure and professional browsing behavior. We see that males gravitate more towards technology related websites and women more to travel and apparel.  

Social media behavior
Another important part of effective persona creation, is knowing the media consumption. Within the acquisition report of Google Analytics, we can see which social media channels refer traffic to our website. This report is not enabled by default and should be enabled. 

We see that by far, Youtube is the biggest source for generating pageviews on our website. This doesn’t come as a full surprise as Google is very active with a diverse set of channels for all its products on Youtube. Although I find it surprising that Reddit is on the low end, but maybe because it has no commercial motive and is not the right place for products such as Google Merchandise. 

PlatformPage views
Google Groups1,336

Just like with our mail audience, we see that Youtube by far generates the most page views on our website. Youtube is by far the most popular among women and other channels barely make a dent in the amount of pageviews. 

PlatformPage views
Google Groups276

Creating the buyer persona

We can never fully create a buyer persona without going deeper into the more abstract details, as mentioned at the beginning of this post, such as buying behavior, pains and gains. But we’ve come a long way, from knowing nothing to having the rough edge of our two most loyal customers. Let’s have a look. 

From the analysis we’ve discovered several findings:

  • Our buyer persona is a male between the age 25 – 34, which we call Adam
  • Lives in the United States
  • Likes to follow technology news and is career minded
  • Uses Youtube, Google Groups and Twitter as his go to platforms
  • Explores and buys from our website on a smartphone 

For our female persona we can define the following: 

  • Our buyer persona is a female of the age 25 – 34, which we call Susan
  • Lives in the United States
  • Likes to follow employment and travel and is career minded
  • Youtube is her platform of choice
  • Explores and buys from our website on a smartphone 

We see that the buying behavior of our persona’s are pretty well aligned, but see a difference in interests. Where males tend more towards technology news, women like employment related topics and travel. Which is highly likely consumed through Youtube whereas our male counterpart has a more different palate of social media behavior. This can be either a blessing or a curse. 

We can target this persona on multiple platforms, but as they are far and wide, creating highly targeted ads will be difficult. Our female persona can be efficiently targeted through Youtube ads which are, before the investment of the video content, fairly cheap compared to let’s say search engine advertisements. 


Creating a buyer persona is extensive and actually an organic process which keeps being optimized and redefined. In this post we’ve seen that while not being able to create a full fledged persona, with some clever reports of collected website data and a spark of imagination we were able to see how old our customers are, which regions of the world they are from, which social media they use and which topics they are interested in. This alone gives us pointers what message to craft on their preferred channels. 

Why Spotify struggles in Japan

Japan is one of the largest music markets in the world, but streaming service Spotify keeps struggling. Spotify had a global market share of 32% in...

You may also like

Share This