How to Analyze Social Media Audience Segmentation
by Infegy Research Team on October 18, 2021
See our updated 2023 version here.
If identifying or analyzing online audience segments is part of your day-to-day, for a brand, business, or agency, you need to use more than your typical online analytics.
You need accurate data - not assumptions. And you also need actionable data. The number of Twitter followers or how many likes a tweet gets doesn’t help.
Audience segments can be focused on more than just age and gender. You’ll want to look beyond demographics and analyze psychographics -- how people think and feel, how they buy, and their emotions, interests, and hobbies.
Thankfully, solutions exist, such as Infegy's social listening tool, that can help you dive into these psychographic metrics to learn more details about audiences.
One thing we’ve found when talking with audience segmentation experts is that many brand teams still miss out on crucial audience data with their social media analysis. They often fall into some audience segmentation traps. Here's how to avoid them!
In this article, we’ll cover 5 key things to remember when creating audience segments that are often overlooked. Here they are:
- Base audience segments off what they say, not who they follow
- Look for psychographic details beyond just brand affinities
- Build audiences from multiple channel sources, not just Twitter
- Combine audience segments for a more specific view
- But avoid going too specific with audience segments
Let’s dive into each of these 5 considerations in-depth:
1. Base audience segments off what they say, not just who they follow
Most social platforms allow you to see audience patterns based on whether or not they follow a brand or entity on a social channel. This tells you next to nothing.
If you’re building audience segments, you need more than just follower data.
You need insight into how these audiences express themselves: their lifestyles, their values, their beliefs, and their opinions. Your audience segmentation research should focus on how people think and feel as well as their behaviors.
When you analyze and build audience segments, you can build them based on their conversations, not just whether or not they, at one time or another, clicked the "like" or "follow" button.
Take a look at the differences in the two example audience segments:
Ages 18-34, male, in the U.S. and follows Nike and Adidas on social media
Ages 18-34, male, in the U.S. and who mention sneaker culture, sneaker collecting, #SneakerHead, shoe influencers, or that they plan to buy new shoes
Social listening insights allow you to better create more accurate audience segments based on: who they are, what they’re interested in, brands they love and talk about, topics of interest, how they typically buy, and lifestyles.
People don’t base their lives around brands. So, let’s look beyond the brand to see how people describe themselves.
2. Look for psychographic details beyond just brand affinities
Your typical audience segment might look something like this:
25-34 years old, male, making $50-75k, who loves outdoor adventure brands like The North Face, REI, and Patagonia.
Paying attention to the last part of that description, you can see that the details tell you a little about who this audience segment is, but it doesn’t go far enough.
It’s great to list out possible brands that the audience segment likes, but you need more. The reason is, you can learn a lot more about the audience segment than just topics relevant to brands they love if you have the right analytics tools.
For example, if your audience segment loves outdoor adventure brands, you might include that they have an interest in outdoor adventure or outdoor winter sports. But if they are interested in these things, that does not mean that’s all they’re interested in.
For our custom audience segment of The North Face brand advocates, we included important psychographic and lifestyle information that is important to know about these audiences:
In this audience segment, you see under “interests” some topics that make sense like skiing, sports, and travel. But you also find some things you might otherwise miss:
- They’re interested in humor, pop culture, and food and drink
- They’re gamers! ...one of their top-mentioned keywords is games
- Their leading age range is 25-44, but they are also the youngest members of Gen Z
- The themes they express center around quality, intent, promotion, and discounts
- They also mention climbing, hiking, photography, travel, and nature. They appear to enjoy hiking, outdoor photography, and nature hiking
Look how much more you can learn about these consumers and their lifestyles by looking at what they say, not just information related to a specific brand.
Building social personas based on social media following is a dangerous trap for analysts and strategists. Another is relying on only one channel. Let’s look at that.
3. Build audiences from multiple channel sources, not just Twitter
Consumers use different platforms to talk about different things. They express themselves differently and present themselves to the world differently based on the channel.
This concept underscores a vital consideration when building audience segments. Understand that people portray themselves differently on each channel.
Most social analytics tools rely way too heavily on data derived from Twitter. If you’re building audience segments with social data, Twitter doesn’t tell you everything you want to know.
We know from our research that consumers turn to Twitter more for customer support purposes and less to describe themselves in identifiable ways that would be useful for audience segmentation. However, on Instagram, Pinterest, and forums, consumers often provide more crucial information for your research.
When we built Gen Z audience segments for our Gen Z purchasing behavior report, we found that our audience segment discussed purchases they made or planned to make on other channels, forums like Reddit:
You see here that conversations about purchases from Gen Z occur primarily on forums, Instagram, and Pinterest. But if you were like most, you’re probably looking at Twitter (shown as Microblogs) the most because it has the most conversation.
Another great example: when we looked at how people talk about beauty and cosmetics for our Beauty Report, you can see the varying ways people discuss their beauty preferences, routines, and lifestyles, based on the channel.
In summary, Twitter is not the only place consumers turn to discuss experiences with brands and products. It’s key for you to analyze other channels when building your audience segments with social media data.
4. Combine audience segments for a more specific view
In today’s high-access, mass-information online world, you are likely targeting several different types of target customers.
For instance, if you’re in the entertainment industry, you might be looking to target younger consumers, but you might also be interested in people who are big Netflix superfans.
Using social listening research, you could combine different audience segments to get more specific with your audience segmentation.
By using the right search parameters within your research to pinpoint how people talk online, you could narrow it down to very niche audiences.
Here’s the audience segment for Gen Z Netflix watchers who also happen to speak both English and Spanish:
We did this in several different ways using capabilities available to us in our social listening tool:
- We used our tool’s built-in age-segmentation filter feature
- We searched online conversations for self-identifying terms to find super fans of Netflix and Netflix programming
- We searched for self-identifying conversations that mention that they speak both English and Spanish in posts that use either language
It’s incredible how far you can go to drill down narrow audience segments based on specific criteria using social listening data. However, keep one last thing in mind:
5. ...But avoid going too specific with audience segments
We’ve seen some wild attempts at audience segments around here. Take this one for example:
Females, age 23-40 who love the outdoors, want to be overachievers at work but also like to have fun with their friends. And they like animals.
Trying to be this specific doesn't tell you what you need to know about your audience segment because people don't express themselves like this on social media. (When's the last time you tweeted "I want to be an overachiever at work!"?)
Instead, it’s important to remember how people talk online. Then, look for ways people describe themselves and their lives that have to do with your brand or business.
When building audience segments, you aren’t looking for one person; you’re looking for a group that best reflects a typical person. This is how an audience segment will help you target, market, and sell your product or service.
Audience Segmentation is a tricky process. There are many different ways to go about it, and no two audience segment templates will look the same.
But the most critical factor in creating quality audience segments is having the correct data. To get the best data for your audience analysis, you’ll need to keep the above 5 critical factors for audience segmentation.
Social listening provides you with the best path to meet these criteria. It can drastically improve your research for audience segmentation, buyer personas, and targeting.
Get in touch with an Infegy rep to learn more about our social listening insights for audience segmentation!
- March 2023 (2)
- February 2023 (4)
- January 2023 (2)
- December 2022 (3)
- November 2022 (5)
- October 2022 (3)
- September 2022 (3)
- August 2022 (2)
- July 2022 (1)
- June 2022 (1)
- April 2022 (1)
- March 2022 (1)
- January 2022 (1)
- December 2021 (1)
- November 2021 (2)
- October 2021 (1)
- June 2021 (1)
- May 2021 (1)
- April 2021 (1)
- March 2021 (1)
- February 2021 (1)
- January 2021 (2)
- November 2020 (1)
- October 2020 (2)
- September 2020 (1)
- August 2020 (2)
- July 2020 (2)
- June 2020 (2)
- May 2020 (1)
- April 2020 (1)
- March 2020 (3)
- February 2020 (2)
- January 2020 (2)
- December 2019 (2)
- November 2019 (1)
- October 2019 (1)
- September 2019 (2)
- August 2019 (2)
- July 2019 (2)
- June 2019 (2)
- May 2019 (2)
- April 2019 (1)
- March 2019 (2)
- February 2019 (2)
- January 2019 (1)