About bidding in universal app campaigns
In universal app campaigns, you’re able to optimise your targeting and bidding for specific campaign objectives – whether that's getting more app users or more in-app actions from your new users. This article describes the different optimisation and bidding options available, and how to set those bids based on your objectives.
This article also explains how the Android bid simulator works. This is the graph that you may see when you create your Android campaign, which can help you to set target cost-per-install bids based on projected performance for apps like yours. To see the bid simulator, you’ll need to link your AdWords and Play accounts.
Before you begin
If this is the first time that you've used a universal app campaign, it's best to begin by reading About universal app campaigns. When you're ready to create your campaign, you can follow the instructions in Create a universal app campaign. In order to optimise your campaign for in-app actions, you'll have to set up in-app actions as conversion events in your app. Find out more about mobile app conversion tracking
How it works
When you create your universal app campaign, you can choose to optimise your campaign based on which type of user is most valuable to you. However, before you can optimise for in-app actions, you'll need to set up those actions as conversion events within AdWords, or using Firebase.
The simplest way to track all of your app conversions is to use Google Analytics for Firebase for your app conversions. Once you've linked Firebase and AdWords, you'll be able to bid for in-app events that you've set up as conversions. Read more about tracking mobile app conversions with Firebase
Campaign optimisation settings
Universal app campaigns can focus your ads on getting the most valuable users for your app. There are two settings, based on the type of user that you'd like to acquire. The second option is only available if you've set up in-app actions as conversion events in AdWords.
- App installs. If your goal is to get as many new users as possible, then select this option. This will optimise your targeting and bids to generate more installs from your ads.
- In-app actions. This setting focuses your ads on the people who are most likely to perform the actions in the app that you select. Once you've chosen this option, you'll be able to select in-app actions from those that you've set up previously. AdWords will find users who are most likely to perform the actions that you select.
Bid strategy setting
For universal app campaigns, you'll set a bid for app installs. This is called the target cost-per-install – how much you're willing to pay for a new user of your app. In other parts of AdWords, this is referred to as "Target CPA", or target cost-per-acquisition. It works the same way, but an "acquisition" in your universal app campaign is an install or first-open of your app.
Bear in mind
This isn't your bid for an in-app action, it’s always for an install or first-open.
Using the bid simulator
You may see bid projections for app installs in a graph underneath the bid value box. These estimated weekly app install volumes show the potential performance of your app.
In the charts, you'll see a curve showing the bid value on the X-axis, which is labelled "Total cost per install". In the chart, you'll see the potential performance based on that CPI bid.
- To see the install volume associated with a certain level of performance (for example, 700 installs per week), you can hover over the dot that corresponds to that number on the Y-axis.
- To set your bid amount to match that level of projected performance, click the dot in the chart or the "Apply Target CPI" link in the box that appears. Both actions will update your CPI bid value in the "Bid strategy" box above.
If you can't see this chart, make sure that you've linked your AdWords and Play accounts and check back in your campaign settings when your campaign has more traffic. With more data, the system may be able to run a projection for your campaign.
Confidence intervals for estimates
In your chart, you'll see a line with dots for potential bid values and a shaded region. This shaded region is called the "confidence interval" for the bid projection. "Confidence interval" is a statistical term that refers to a level of certainty behind a prediction or estimate. In this case, it's calculated as the percent likelihood that a prediction is true. For data points inside the shaded region, the predictions are stronger – more likely to be accurate.