Vermilion Design + Digital Vermilion

Open Menu
 

5 Dos and Don’ts of Meta Audience Targeting

I’ll start by acknowledging the obvious: of the ever-growing list of digital platforms to advertise on, Meta (a.k.a. Facebook and Instagram) isn’t the sexiest. And right now, most blog posts would rather turn their attention — and your clicks — to newer or trendier darlings like TikTok, Reddit, or streaming audio. 

But Meta remains a go-to advertising platform for brands and organizations, especially those just starting out with paid social in their media mix. With its enduring and enormous usership, various ad formats, and audience targeting options, we find Meta continues to provide insights that can make you a more prepared marketer.

1. DO: Carefully consider ad placements on Facebook vs. Instagram.

Because Meta owns Facebook and Instagram, they’d like you to liberally spend your ad dollars on both. Therefore, when setting up a new campaign within Ads Manager, the default setting is “automatic placements.” That means Meta will place your ad in any available inventory (in-feed, right rail, Audience Network, etc.) across the two platforms based on what their algorithm deems most valuable with respect to your campaign goals. 

With over 40 years of advertising experience, at Vermilion we see extremely varied behavior between the two. A thoughtful and considered approach to actual ad placement is very important. 

What is the key action you want your audience to take? For example, we often see users are more likely to click away from Facebook (resulting in higher CTR and lower CPC), while Instagram users tend to share or save a post for later instead of immediately leaving the app (resulting in a higher engagement rate, but lower CTR). Consider which channel makes the most sense for your overall campaign goals.

2. DO: Build audiences in Ads Manager early in the planning process.

Success relies on understanding your target audience(s). Strategists will outline audiences during campaign and media planning, but consider taking one additional step during this pre-launch phase: Build your audience within Ads Manager by creating a new campaign and ad sets (use placeholders where needed, as this can be archived later). Doing so may reveal learnings that lead you to proactively adjust creative or media dollars for an even stronger launch.

For example, let’s say our primary audience is women in the U.S. between 20 – 40 years old with an interest in yoga and meditation. With Meta’s automatic placements enabled by default, Ads Manager indicates there is an estimated audience size of 23 – 27 million. When filtering to Instagram placements only, that range drops to 18 – 21 million; Facebook comes in at 15 – 18 million. 

That means our audience is about 20% larger on Instagram — not a small difference! Relay this information to the strategist and creative team and together, consider: Should we allocate 20% more budget to this platform to reach our full Instagram audience opportunity? Will highly visual assets perform better given the nature of top-performing creative on Instagram? Should we lean more into Story format content? These are all valid questions to discuss based on the audience size discrepancy between the two platform placements.

3. DO: Separate key audiences by ad set to better report on them.

There are many helpful dimensions within Meta’s built-in Ads Reporting tool that can parse out and report on: age, gender, state, placement, etc. However, there are others you might assume would be available but aren’t, such as city, zip code, interest type, and custom audience type.

Continuing our example from earlier, let’s say you’re charged with determining who is more likely to engage with the ads: women interested in yoga or meditation. If we include both interests in the same ad set, we can’t tease them apart in the Ads Reporting tool. Instead, set up two separate ad sets from the outset that are identical except for the unique interest targeting, so you can compare performance between them. Want to avoid audience overlap? Add exclusion targeting (i.e. ad set 1 would include yoga and exclude meditation; ad set 2 would include meditation and exclude yoga).

Keep in mind that oversegmentation can split your budget too finely and result in more expensive CPMs. The key is to identify early which aspects will be important to report on and strike the right balance with help from an experienced digital marketing strategist.

4. DON’T: Rely solely on interest and demographic targeting.

Interest- and demographic-based criteria like we described above are the foundation of audience targeting. However, the effectiveness of this approach has eroded over time. Rightful concerns over data privacy translate to more limited targeting options, such as Meta’s announcement to remove many interests earlier this year. 

Beyond that, we tend to see stronger results when we add another layer to our targeting, such as lookalike modeling. Leveraging first-party data (e.g. a list of customers, email subscribers, or donors) is becoming increasingly critical to success. One approach is to create a lookalike audience from your first-party list within Ads Manager, followed by layered interest targeting to dial in the audience size.

That said, be wary of Meta’s “detailed targeting expansion” setting. This checkbox, which is automatically enabled for certain campaign objectives (e.g. Reach objective campaigns), allows Meta’s algorithm to serve your ads to people it deems likely to respond. Specifically, Meta states: “If our system finds better performance opportunities outside your defined audience… we may expand your audience further.” Of course, you have no way of knowing exactly who or why this expanded audience was selected.

In some cases, this automation may benefit performance — especially if your goals include a lower CPM or increased reach. But, it can be concerning if your budget or audience targeting requires more control. If your campaign touches on sensitive topics, you may want to avoid letting Meta freely expand your reach as it could place your ads in front of antagonistic eyeballs. Negative commentary can flare up quickly on social ads, causing your team to spend precious time and energy on community management.

5. DON’T: Forget to follow post-click performance in your website analytics.

While there is much to analyze within the platform itself, an equally important piece of the puzzle lies beyond the ads. What does the user journey look like after they click and visit your website? In most cases, this is where Google Analytics (or another website analytics tool) comes into play.

Urchin Tracking Modules (or UTMs) are your ally. Use a tool like Google’s URL Builder to strategically pass information from your ads to Google Analytics to go beyond simply “facebook/referral.” For example, use “source” to identify Facebook vs Instagram (if you separated by ad set as mentioned above), so you can assess the behavior of each group once on your website. Use “content” to identify different promotional offers to understand which compels more visitors to purchase or convert. These insights can illuminate audience patterns and learnings to inform and strengthen future campaigns. There’s no wrong way to use UTMs, as long as your team agrees on naming conventions and uses them consistently (pro tip: Google treats capitalized and non-capitalized UTM parameters as separate sources).

———

The next time you build out a campaign, we hope these tips help level up your Meta advertising game. Remember: like all aspects of digital marketing, the platform is constantly evolving. UI interface updates, new and removed settings, and even the digital landscape at large affect how and why marketers leverage Meta. Embrace the uncertainty of the moving target and take satisfaction in knowing you’re continuously learning.

Let's grow the good.

Let's Talk