Mobile app install ads: Get low CPIs on Facebook with this powerful tactic

By Mike Sonders

Mobile app install ads are a very effective method for acquiring mobile apps users.

In particular, the extensive targeting criteria available for Facebook ads–including location, mobile device, demographic, interests, and more–allow app publishers to focus their ad spend on the mobile users who are most likely to install.

To get the most for their money and achieve low costs-per-install (CPIs), app publishers shouldn’t guess about which criteria to target for their mobile app install ads; they should save time and money by learning from the lessons already established by their app competitors.

Today we’re going step-by-step through a straightforward and easy method for doing competitive research to find the optimal audience to target with your mobile app install ads on Facebook, without having to guess or go through needless, costly ad testing.

Here’s what we’re going to cover; you can click a link to jump right to that section:

Why not use Facebook Audience Insights to inform your mobile app install ads?

Audience Insights is Facebook’s freely-available intelligence tool. You can use it to uncover deep demographic, interest, and behavior insights on the fans of millions of different products (including apps).

Audience Insights is a logical place to start when you’re looking for intelligence on an app competitor’s audience.

But you’ll run into several notable shortcomings with this tool:

Audience Insights is missing data

Even if an app’s Facebook fan page has a lot of fans, that doesn’t necessarily mean that Facebook will make data for that app available through Audience Insights.

Coupon app Flipp–one of the fastest-growing apps in the U.S. over the past month, with over 5 million monthly active users–has almost 40 thousand fans on its Facebook page as of this writing.

Yet there’s no audience data available for Flipp through Audience Insights:

Audience Insights doesn’t have specific data on apps

Even when Audience Insights does have data on a particular product or brand, the data often isn’t limited to fans of the app; it includes insights on the product or brand’s entire fan base, even if the product or brand includes web-based, desktop-based, or brick-and-mortar offerings.

Within the same brand or product, the qualities of app users can be quite different from those of other users.

So, for the purposes of figuring out how to target Facebook app install ads, you can’t jump to any strong conclusions based on data that includes anything but a brand’s app audience.

Popular travel app Hipmunk exemplifies this issue. Hipmunk has a Facebook fan page for its website product, but doesn’t have a specific fan page for its mobile app.

Subsequently, Audience Insights has competitive data on Hipmunk’s entire fan base on Facebook, but doesn’t provide specific insights on the Hipmunk app audience. This doesn’t help us.

Audience Insights tells you about fans, not users

This final issue affects any “insight” that you glean from Audience Insights.

And that issue is: Liking an app is not the same as using an app.

It takes very little effort to “Like” a fan page on Facebook, and the motivations for doing so (e.g., you want to follow news related to the page) can vary dramatically from the motivations for installing an app.

To get the best bang for your ad-spend buck, you don’t want to target the types of people who just say they like a direct competitor’s app. You want to target the types of people who actually use the app.

Those users of your competitor’s app are the types of people worth targeting with your ad spend, because they’re more likely to engage with your app (and good app engagement metrics are critical to driving revenue).

Uncover the perfect targeting criteria for your mobile app install ads on Facebook

Now you know that Audience Insights can be a useful tool when analyzing the audiences of your app competitors, but it’s far from ideal.

Let’s talk about a much more powerful approach for uncovering an optimal target audience for your mobile app install ads by analyzing your main app competitor.

Use an intelligence tool that specializes in mobile apps

A mobile app intelligence tool like SurveyMonkey Intelligence (sign up for a 14-day free trial) provides detailed demographic and interest insights on the audiences of your competitors, much as Facebook’s Audience Insights does.

But unlike Audience Insights, SurveyMonkey Intelligence bases its data on the actual usage of iOS and Android mobile apps on smartphones in the United States.

There’s no noise in the data from “fans” who don’t use the app or from insights on non-app products.

By using SurveyMonkey Intelligence to uncover the qualities of your competitor’s audience, you can quickly identify optimal targeting criteria for your Facebook app ads to acquire a relevant, likely-to-engage audience at a low CPI.

Identify your closest app competitor

One of the keys to making this approach effective is choosing the right competitor to analyze.

The more similar a competitive app is to yours in terms of value proposition and features, the better that app will help you uncover your ideal app audience.

When you target people who are similar to the users of a directly-competitive app, they’ll be likely to find your app and its value proposition appealing.

For the sake of example in the rest of this post, let’s imagine that you’re launching a calorie-counting app.

Looking at the Calorie Counter apps within SurveyMonkey Intelligence (which allows you to view apps by sub-category), you’ll see that the “Calorie Counter & Diet Tracker” app by MyFitnessPal leads the market in terms of market share, downloads, and users.

Given this app’s value proposition and the substantial size of its user base (which suggests it has a proven market and has found its core audience), Calorie Counter & Diet Tracker should be an excellent competitor to analyze for your planned calorie-tracking app’s Facebook app install ad targeting.

Analyze the key demographics of your competitor’s audience

Once you’ve identified the best app competitor to analyze, it’s time to uncover the demographic qualities of that app’s audience–including location, gender, age, education, and income–using SurveyMonkey Intelligence.

To identify the most exciting ad targeting opportunities, you’re looking for the demographic segments in which your competitor app’s audience isdisproportionately high relative to that segment’s representation among all U.S. smartphone users.

For instance (and you’ll see this specific example illustrated shortly), U.S. smartphone users are approximately 50% female. If your competitor’s app has an audience that’s 72% female, then it’s safe to say women make up a disproportionately high percentage of that app’s audience. That’s a strong indicator you should test targeting your very similar app’s ads to women.

In the following examples, we’re using SurveyMonkey Intelligence to compare the results for the Calorie Counter & Diet Tracker app to those of Facebook, since Facebook is used by a majority U.S. smartphone users and is therefore a decent proxy for the U.S. smartphone user population.


Looking at the geographic distribution of users of the Calorie Counter & Diet Tracker app, it’s apparent that the Midwest part of the U.S. represents a disproportionate percentage (26%) of the app’s audience relative to its representation among all smartphone users (20%, as represented by Facebook) and among the the country’s population (21%).

Since users in the Midwest are disproportionately represented in the Calorie Counter & Diet Tracker app audience, that’d make the Midwest region a compelling targeting option for the ads for your imaginary new calorie-counting app.

Note: While the South represents the largest contingent of users (37%), the percentage is actually consistent with the percentage of the country’s population that lives in that region (38%). It’s also a smaller representation than the South has among all U.S. smartphone users (41%).

So, while the South is a large addressable market (which is exciting if you have money to burn on exposing a lot of ads to a large audience trying to find your customers), smartphone users there don’t show a particular affinity for calorie-counting apps compared to users in the Midwest.

For that reason, you’re better off testing your Facebook mobile app install ads against the Midwest region, first. You’ll likely have more efficient ad spend resulting in lower CPIs.


In the case of gender (where there are only two options in Facebook targeting), you only have to see that women far outnumber men in using the Calorie Counter & Diet Tracker app to know where you should focus your ad spend.

And this gender-split isn’t typical of the U.S. smartphone user population. Mobile apps users in the U.S., on average, are almost evenly split between men and women, reflecting the composition of the entire U.S. population.

Since Calorie Counter & Diet Tracker’s users skew heavily female, you know this is a compelling demographic insight to include in your ad targeting.


The MyFitnessPal’s calorie-counting app isn’t particularly favored by the young or by seniors (50+). 18–49 year olds are over-represented among the apps users relative to their representation among all smartphone app users.

Unfortunately, this is quite a large age segment to target. The broader the targeting, the more inefficient the ad spend. Where possible, make your targeting more precise.

In this example, since Facebook offers lots of flexibility in terms of targeting by age, it would definitely be worth experimenting with smaller age segments within this large 18–49 segment.

Facebook ads expert Jon Loomer recommends segmenting your ad targets into age groups of 5–10 year ranges. In this case, you could try different tests targeting 18–23, 24–29, 30–35, 36–41, and 42–49 years old, or some variation.


A disproportionately high number of holders of college and advanced degrees use the Calorie Counter & Diet Tracker app:

As you’ve gathered by now, you should definitely prioritize targeting people with college and advanced degrees for your imaginary new calorie counter app.


People with incomes between $50 and $100 thousand are disproportionately represented among the users of the Calorie Counter & Diet Tracker app (38%) relative to their representation among all U.S. smartphone users (25%).

As mentioned in the previous section regarding age demographics, it’s probably worth testing less-broad segments among this target segment. Facebook provides income targeting segments of $50–75 thousand and $75–100 thousand.

Take your ad targeting on Facebook to the next level with interest data

One of the distinctive features of Facebook app ads is the ability to target by interest.

Again, you don’t have to guess about the interests of the audience for a new app. You can improve the targeting of Facebook mobile app install ads using behavioral interest data from SurveyMonkey Intelligence.

This behavioral data tells you which apps the users of your competitor’s app are more likely to install. In other words, the data tells you that apps in which your target audience has an interest.

Include apps they’re interested in

Continuing our example of a new calorie-counting app, you can see that the users of the main competitor app Calorie Counter & Diet Tracker are 8 times more likely than the average smartphone user to install the Voxer Walkie Talkie Messenger app.

Likewise, Calorie Counter & Diet Tracker users are also more likely than average to use other MyFitnessPal apps like MapMyRun and MapMyWalk.

To focus your ad targeting for the imaginary calorie-counting app, you could tell Facebook that your ad’s audience must include people who have indicated an interest in any one of these apps. (Remember that not all the apps will be available as targeting options.)

Exclude apps they’re not interested in

Similarly, you can tell Facebook to exclude people from your ad’s audience. To do that, we return to SurveyMonkey Intelligence’s “Propensity” tools to see which apps are less likely to be used by the users of our competitor’s app vs. average smartphone users.

In this case, you’d definitely wouldn’t want your ads to appear in front of people who “Like” 8 Ball Pool, Madden NFL Mobile, and TuneIn Radio. You could keep expanding your “exclude” list on Facebook Ads with any of the apps from this “negative propensity” list.


You should always test different variations of your mobile app install ads, including their targeting. But there’s no reason to start from a blank slate or to make uninformed guesses about your ideal target audience.

Use a mobile app intelligence tool like SurveyMonkey Intelligence to analyze the demographics and interests of your closest app competitor’s audience to guide the targeting of your first set of Facebook ads.

The more precisely you can use this data to find your ideal target audience on Facebook, the lower your costs-per-install will be.

Finally: When running mobile app install ads on Facebook, remember to exclude people who have already installed your app.

This post originally appeared on October 11, 2016 on the blog of SurveyMonkey Intelligence, a provider of competitive intelligence for the mobile app industry.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.