Can you offer some best practices for iOS campaigns?
Patrick McGuigan, Business Development, Mobile App Partners, Google: We are seeing many advertisers shift more iOS investment into UAC for ‘Action’ based campaigns; finding high quality users taking action at target CPA bids. Many advertisers increase their bids on iOS (compared to bids on Android) since iOS users generally have a higher lifetime value than that of Android users.
Since the search network is limited and Google Play inventory isn’t available for iOS, advertisers must include strong creative assets so they can win auctions on YouTube and Display networks. Having good creatives is critical for both platforms but it becomes even more important when it comes to iOS campaigns.
Since Android is the native platform for Google and has all placement and tracking options, collecting more signals is easier. iOS requires more patience – these campaigns must be given more time to collect data and optimize before comparing results.
How would you ensure strong quality and performance for iOS UAC?
Sharon Zimlichman, Head of Media Development US, Bidalgo: iOS performance can be attained through a number of premium properties. On iOS there are a lot of great signals from Search, YouTube and Admob/Display that help identify user intent; Google has the same great bidding options (and soon to come hints) to help achieve business goals. UAC uses numerous signals from inside and outside the app (time of day, signal strength) to deliver the best possible ad to the most relevant user at just the right time.
UAC’s Machine Learning algorithms continuously process over 300 million signal combinations to help find the best users possible across both Android and iOS. Moreover, UAC allows you to reach properties that you wouldn’t normally be able to reach on iOS such as YouTube – a valuable destination to find high quality users.
Do UAC algorithms behave differently if you have a 100x budget:bid ratio versus, say, a 500x budget:bid ratio on install campaigns?
Sharon Zimlichman, Head of Media Development US, Bidalgo: The mantra of machine learning is: the more data the better! The more input the machine has, the faster and more accurately it can detect patterns to predict future success. So with this in mind, the 500x budget:bid ratio would give much more data to feed UAC for faster results, more scale and better performance.
What performance metrics is UAC prioritizing when optimizing assets (CVR, CTR)?
Patrick McGuigan, Business Development, Mobile App Partners, Google: In terms of prioritization, there are wide performance metrics incorporated, including CVR and CTR, as well as auction-time signals such as device, location, time of day, language, and operating system. Asset performance across each network is also taken into consideration (YouTube, Display, Search). UAC optimizes assets according to their ability to hit specific campaign goals. If the Conversion Optimizer recognizes that an asset can achieve a specified goal, it will try to push and prioritize that asset in the rotation. In the first week, before UAC collects enough data on the asset, it will use other upper funnel signals like ad engagement (e.g. view rate on YouTube) as proxies for the desired events.