November 24, 2017
PPC App Advertising Heroes! Somehow it is already nearing November’s end, and you know what that means – that’s right, it’s the official sunsetting of AdWords mobile app install campaigns (did you think I was going to mention the holidays?) Google announced this change back in August, alerting users that on November 15 all other existing app campaigns on Search, Display, and Youtube would stop running.
Universal App Campaigns were launched in mid-2015 as part of Google’s efforts to utilize machine-learning in its advertising product offerings. The goal is for the machine-learning system to analyze, in real time, hundreds of millions of signals to put your app in front of the best possible users. We have seen Google’s push for automation in campaign management and optimization over the past few years, and most recently at their Google Marketing Next 2017 conference. According to Google, by using machine learning Universal App Campaigns can drive 140% more conversions that other Google app promotion products.
Hopefully by now you’ve become acquainted with the differences between the old mobile app install campaigns and the Universal App Campaigns (UACs) you should be using for your app promotion ads (if not, this post by Shannon Glass succinctly breaks down the differences).
As you create your new UACs, I’ve identified the 3 best practices to keep in mind to ensure that your efforts, combined with Google’s spectacular machine-learning capabilities, make for the optimal App Advertising endeavors that meet your (or your client’s) goals.
1. Pick the Right Campaign Goal
There are two possible campaign goals to choose from for your UACs – “Install Volume” and “In-App Actions.” You want to identify the type of customer you’re targeting with your campaign, and assign a value to each conversion (new user installs or in-app conversions such as level-ups or purchases). Both goals are going to optimize according to a set target CPI, but how you determine your target CPI should depend on which goal you’ve selected.
a) Install Volume
Choose install volume as your campaign goal if you want to focus on finding as many new users as possible to install your app. This goal is ideal for new apps, or apps with new updates/features.
When you optimize for install volume, Google recommends setting a campaign budget that is 50 times greater than your target CPI. For example, if your target CPI is $ 2, you should set a campaign budget of at least $ 100 according to Google. You also should ensure your campaign is not limited by budget, because Google’s algorithm needs enough conversion data to learn and optimize.
b) In-app Actions
Choose in-app actions as your campaign goal if you want to find users who are more likely to complete conversion actions like in-app purchases. With Universal App Campaigns, then, it is important to properly set up conversion events so you can track these actions and so Google’s algorithm can optimize for your unique conversion goals.
Google recommends choosing an action that occurs at least 10 times per day, and suggests a campaign budget of at least 10 times your target CPI (so, if your conversion action is a level-up in a game, you want to make sure that your app is bringing in at least 10 of these per day…and if your target CPI is $ 2, then your campaign budget should be at least $ 20.
Remember that you can have multiple campaigns for different goals! One of the ways you can focus on one campaign’s goals while continuing to run campaigns with another goal is to utilize your control over your target CPI and campaign budgets. Therefore:
2. Set the Right Bids and Budgets
Say you’re running two UAC campaigns – one with an “Install Volume” goal and one with a goal for “In-App Actions.” According to Google’s recommendations, you know that you need a target CPI and campaign budget that allow for Google’s algorithm to collect enough data while providing a return on your investment. But what if the customers you want to reach change throughout the year?
Google gives this example:
“Say it’s been a couple of months and you’ve added a social feature and new levels to your game. You want to get this update to as many new users as you can. You adjust the target CPI and budget UP for the campaign that’s focused on “Install Volume.” At the same time, you adjust the budget DOWN for the campaign that’s focused on “In-app actions.” You’re signaling to AdWords yet another change in direction.”
By wisely altering your bids and budgets, you have control over the results of your campaigns that couple with the automation/optimization that Google brings to the table with machine learning.
Reminder: machine learning requires data to learn in order to optimize for performance goals. When you set up your campaign budgets and bids, make sure you give the campaign ample time to run before making changes – Google suggest allowing time to gain at least 100 conversions.
3. Utilize the Right Creative Mix
While you can’t create individual ads as a whole in UACs, you can provide creative assets that Google will combine and rotate – these combinations will be used to optimize toward the best-performing asset groupings in order to maximize the conversions you get at your Target CPI.
You’ll provide 4 “text ideas” that need to be standalone and should be able to show in independent order. Testing different “text ideas” is one of the ways you have control over campaign variables.
You also are able to provide additional ad assets, such as images and video. In order to show in all placements, be sure to add images of various orientation and sizes, and videos of varying duration and size.
You can utilize the Creative Asset Report to evaluate the performance of your assets. They will be listed by performance grouping of “Best,” “Good,” “Low,” and “Learning.” Replace “Low” performing assets with new test assets, and leave “Learning” assets be while Google acquires enough data to give them a ranking.
Giving Up (some) Control
While many of us are resistant to the change and loss of some of the “control” we were used to with mobile app install campaigns, Universal App Campaigns show promise for great performance with a more hands-off management requirement. However, you still have control over the goals you choose, the bids and budgets you utilize, and the creative you provide for Google to test and optimize around for data-driven results that are simply beyond manual capabilities.
Time will tell how it works for you, but since Google has been moving toward more and more automated processes that use its machine-learning algorithms rather than the “manual touch” of the past, it’s best to stay ahead of the curve and learn how we can make the best decisions to help Google learn how to drive great performance.