Ad Networks Explained Part I: Which ones should you use for your UA campaigns?

In this blog post series, we will take a look at the current landscape of mobile advertising in gaming and share UA best practices for ad networks.

Can Sölömbaz
Can Sölömbaz
Data Scientist
Ad Networks Explained Part I: Which ones should you use for your UA campaigns?

Once upon a time, UA specialists were only spending their marketing budget on Facebook and Google without thinking too much about attribution or competition. Today, the marketing scene has evolved to a phase where UA teams have to allocate their budgets into a dozen ad networks and lock horns with hundreds of competitors to get the attention of their possible users.

After the deprecation of IDFA, the available holistic data the ad networks can access and the  accuracy of their machine learning (ML) estimations decreased significantly. It has become impossible to target the most valuable users on the iOS platform and soon, similar changes are expected to happen on Android. The new ad ecosystem already started to threaten certain business models.

In this blog post series, we will take a look at the current landscape of mobile advertising in gaming to understand how UA professionals can get the most performance from their marketing budgets and share UA best practices for each of the ad networks.

The first stop of our journey is the comparison of ad networks’ performance and find out which ad networks drove the most installs in this crazy year of mobile marketing.

Which network is winning?

According to the Performance Index of Appsflyer¹, which compares attributed installs in 2021 on consented users, 7 networks achieved the most number of installs in gaming.


TikTok For Business claims the #1 spot in SKAN Index, while other top mobile media sources like Facebook are still adapting to the new reality¹.

The non-self-attributed networks (SAN) have been on the rise for a while especially for gaming. The report supports the trend and shows that Unity, Ironsource, and Applovin continue to be top non-SAN networks for UA specialists. They also have not been as severely impacted by the new ATT paradigm as self-attributed networks.

Facebook is probably the most affected network by the deprecation of IDFA since their core ML algorithms highly depend on tracking activities in other apps. On the other hand, it is still a top channel for user acquisition since it still has a huge first-party data advantage.

One of the outstanding UA networks is TikTok for Business. Recently, it reached 1B monthly active users and it is now too big to ignore for mobile advertisers. The platform has unique virality effects which drive an impressive amount of organic installs when a campaign is successful. It contributed to one of UAhero's customer’s success in reaching #1 on the US top chart.

As a result, we can say that exploiting all the promising acquisition networks is a must for UA professionals. Each ad network has unique characteristics. Some of them have a massive reach potential while others are more suited to the changing landscape of privacy-first mobile advertising.

A brief intro to ad networks

Let’s take a look at the characteristics of ad networks to understand how they are suited to the new advertisement ecosystem.


Before exploring the specific characteristics of ad networks, we should talk about attribution. When using multiple ad networks at the same time, UA teams should not base their decisions and strategies solely on the reports of SANs. A probabilistic attribution or multi-touch attribution is necessary to avoid duplicate or fraud attributions and both methods are provided by MMPs. You can check our blog post on iOS 14.5+ to get informed on what SKAN attribution is and how you can benefit from it.

Network types

To explain it more clearly, we can divide the ad networks into 3 groups.

I – The ad networks which are not social networks: Applovin, IronSource, Unity Ads, Vungle…

II – The ad networks which are also social networks: Facebook, Snapchat, TikTok…

III – The ad networks that can’t be specified as either: Google Ads, Apple Search Ads…

For the first group, the common campaign types are App Installs and Retention. The most common targeting options on these platforms are geo, publisher, and device type. Marketers can determine their target cost per action (CPA) based on these granularities regardless of the platform type.

It is a known fact that there is a huge difference between the lifetime value (LTV) of users in different tiers of countries (tier 1, tier 2, etc.). Therefore, all countries should be targeted with different CPA objectives to reach optimal performance.

Also, if your team has enough budget, targeting publishers with different bids is crucial since a user coming from a casino game will have different behavior than a hyper-casual player. In addition, you may see a big difference in LTV between an iPhone 13 user and an iPhone 8 user. To offer granular bids, marketing teams have to spend a significant amount of budget but following this path can be the differentiator you need in such a competitive market.

Below, you can find the advertiser-publisher data from SKAdnetwork that show the distribution of installs².

For the social network group, there are more options to try and play with, however, keep in mind these networks have been impacted more by the new ATT paradigm. While targeting iOS devices, you should acknowledge the fluctuation in the current environment.

Below, you can find a table from Snapchat Ads that shows the expected impact of ATT on its platform capabilities³.

What we can expect from these social network platforms would be the loss of their retargeting and custom audience abilities for iOS (and possibly for Android in near future). However, social networks have huge data matrices for their users from signals in their first-party data. This means that marketers can still target users on demographics, engagement activities on the platform, and other detailed signals coming from the first-party data.

In social ad networks, usually, there are two options to determine targeting costs. Even though the naming can change, the first option is optimizing towards the lowest cost. When this option is chosen, marketers don’t have full control over their bids and CPA, however, choosing the lowest cost can improve campaign performance by giving algorithms a chance of collecting more data. The second option is optimizing towards target CPA and it can be used to reach the target ROAS for your campaigns. There is also a target ROAS option in some platforms but you need to share your revenue data with platforms to be able to use this targeting mode.

The last group has different behaviors than the first two groups we talked about. Google Ads is slightly similar to social networks in terms of target cost strategies. But, all the efforts in app install campaigns are made at the campaign level and marketers can only target geo and device type. The other network, Apple Search Ads, allows you to promote your app in the search tab and on top of the search results with the ability to target an audience of choice.

How can you reach your target goals?

User acquisition teams have to use different distribution channels and the most granular levels to get all the value in the ecosystem. You should automate your user acquisition decisions, accurately measure the LTV of users in the most granular levels and build bidding algorithms to reach optimal performance. As we mentioned before, the best way to do that is AI-Based Automation.

In the upcoming posts of this series, we will deep dive into best practices in the ad networks so that you do not miss any performance in your mobile campaigns. Stay tuned!


Stay updated
Get the latest news on UAhero and the mobile marketing world.
Thank you! Check your inbox to confirm your submission.
Oops! Something went wrong.
Your privacy is important to us, see how we handle your data in our Privacy Policy.