Redefining customer personas: Beyond demographics and stereotypes 

When was the last time an ad made you stop scrolling out of pure interest? How often do you see one that makes you think, “Hey, I could really use that” or “That’s really cool”?

Personally, it’s not often, but it did happen today. I spotted an ad on Instagram showing phone cases with classic art designs on the back. Not only were they paintings I knew and loved, but the cases fit the same model that was in my hand.

However, the next ad I saw was trying to sell me some strange, new workout equipment that quite frankly I had no interest in. The only exercise it inspired me to do was to swipe past the ad as fast as possible.

Why did these ads have opposite effects? The answer is simple: targeting.

While the first ad encompassed my interests in art and catered to my phone— specific things the advertiser targeted based on my online behaviour and device type — the latter likely only targeted my demographics. And why my demographics? Because I’m a male in my early to mid-20s, and there’s a stereotype that this segment of people is into working out.

The goal of good marketing is to reach the right people and give them something they’ll care about for the benefit of an organization. That can solely be done by taking the time to understand your audience. Relying on stereotypes isn’t a shortcut, it’s a wrong turn.

Stereotypes? Not a good strategy.

Marketing has a problematic history of perpetuating harmful stereotypes, at times even playing a role in their creation.

In the early 80s, advertisements such as ones for the Commodore 64 were universally appealing, highlighting their usefulness for the entire household without gender bias. Yet, by the late 80s, the narrative shifted. Ads from Apple questioned, "What kind of man owns his own computer?", and IBM's marketing predominantly featured men. As a result, tech organizations missed out on broader engagement opportunities and contributed to harmful stereotypes that still exist today.

The tolerance for heavily biased marketing is thankfully decreasing. After stories of job ads excluding women, rental ads excluding select races and more, changes came in 2022 when Meta removed the ability to target based on race, health causes, sexual orientation, religious practices and political beliefs. Luckily, these updates don’t pose an issue if your marketing relies on modernized customer personas.

However, due to user privacy concerns, some platforms are removing other, less newsworthy interest categories. As more devices block 3rd party cookies, platform pixels become less effective at connecting people to interests, leading to fewer targeting options.

For example, one recent removal from Meta included the ability to target users “in or regularly in a location.” Now, it’s only possible to target users “that are or have been” in a location, which could include those who are just travelling to the location and not part of the target audience. You can account for this change by excluding all other locations in your ad set.

Despite Meta’s attempt to reduce biased advertising, algorithmic extrapolation is still at play. If your ad is targeting many more than it could reach, Meta will use past behaviour and trends data to determine who will see the ad based on its goal. At times, this is problematic. Research has shown that the algorithm will disproportionately deliver ads featuring young women to older men, which one researcher called the “Creepy Old Man Effect.”

If you really know your audience, you may not want to rely on the algorithm. Here are three ways you can reduce algorithmic extrapolation:

1. REFINE YOUR AUDIENCES

Utilizing more targeting can minimize your audience size and allow you to re-engage them more often. If you have an audience size of 5,000 and enough of a budget to reach them with a frequency of five or six times, your ads will be seen by the right people often.

 

2. BUILD YOUR AUDIENCES

Use broader targeting on a video ad that ONLY people in your market would be interested in. Then create a “retargeting audience” of people who watched 75% or more of the video. This way, you can use your videos as a filter to build the perfect audience yourself. Now you can run ads to a much smaller, proven audience. These retargeted ads often have more direct messaging, making them more efficient at moving your audience through a sales funnel.

 

3. TEST YOUR AUDIENCES

Instead of your ad set targeting a big group of people ages 15–55, you could create three sets for ages 18–34, 35–44, and 45–55. This way you can bypass a platform’s bias and get your own data on which of these age groups engage with your ads most. You can use this same approach with other demographics, interests, and so on.

With that said, these algorithms have billions of dollars put into them, so sometimes they are more effective. You can always test your targeting against a set that allows for algorithmic extrapolation and see for yourself what really works. Ultimately, it’s your call!

Course-correcting your targeting.

This isn’t to say that you should never use demographics such as age or gender in your marketing, but rather it’s important to think critically about when it makes sense to. Of course, it’s hard to get away from these entirely because they are baked into the data provided by marketing research companies. Still, your brain is your best tool in evaluating and applying those findings.

For instance, if you’ve determined (through survey data, A/B testing, interviews etc.) that your ideal customer profile is 25-35-year-old males from South America who love soccer, it could make sense to keep targeting them. In fact, learning about where they’re from can help with tailoring your messaging, and you can learn more about tailoring content here. The best ads show your audience that you understand them. It’s all about empathy.

Another source of impersonal targeting is promotional DMs (Message Ads) on LinkedIn. They’re often sent to a very wide audience and consequently feel spammy. A better approach would be to break the audience into subgroups, then personalize those DMs to better resonate with each audience. You can learn more about this approach and other digital outreach personalization tactics with our guide.

Here's a good way to see if you’re getting niche enough: name the audience.

A traditional customer persona might be named “Downtown Gen Zers.” In today’s highly fragmented world, it’s not a good idea to treat all Gen Zers as one entity, including the ones living in the same area. There’s an ocean of different viewpoints and experiences among them, so thinking of them as the same at such a broad level doesn’t really make sense. Even if you believe your product or service would be useful to all Gen Zers that live Downtown, you’re better off separating the segments and customizing your messaging accordingly.

A more nuanced persona name might be “Social Networkers.” This audience of twenty-somethings live in apartments, eat at Mexican restaurants, attend university, read e-books and shop at Zara. Interested in learning more about them? Well, you can! They’re segment 47 from Environics Analytics’ PRIZM®.

At Show and Tell, we use audience data from companies like these and break them down into meaningful segments. For instance, if we identify a subgroup that likes new technology particularly more than the others, we might create ads for them that use AR capabilities. Or if we find data showing that the audience drives blue hatchbacks, we might have the ad feature this vehicle in some way – but we might not. Ultimately, it depends on the client, audience, placement and findings. Like I said, your greatest tool is your brain.

As you can see, zeroing in on niche segments can help focus marketing efforts. With this in mind, let’s revisit that workout equipment example from earlier.

Assuming they only targeted males, they’re missing out on potential sales from people of other genders who are into working out. If they changed the ads to reach all genders, yet limited the scope to those who work out, buy online, and so on, they’d attract more people who care. They could even set up multiple audiences to monitor which ones interact with the ad the most, then either reevaluate the other audiences, try new ones, or simply focus their attention on the most successful segment. Learning as you go is a huge part of improving these personas and driving results for your business.

Tip: Meta will automatically test which of your ad variations performs the best and focus on promoting those versions. While this can be useful, it might cause your audience to burn out on the ad quicker or prevent them from seeing a diversity of people in your sets. To prevent this, consider pausing the most “successful” version in favour of the other ones at times. You’d be surprised how often these ad sets end up performing very well.

When it’s all said and done...

The best way to communicate with others is by remembering that they’re people, just like you and me. They’re individuals with lives. Consider their pain points. What kinds of things are they saying online? What makes their lives harder, and how can you help? If you take the time to really learn about who you’re trying to motivate and what problems they’re trying to solve, you’ll see better performance and enjoy the work a lot more. It becomes a reminder that you’re making a wholesome contribution to their lives and not just trying to make them do something.