blackout window for app retargeting campaigns

How to Establish a Blackout Window for App Retargeting Campaigns

By David Bennet | February 6, 2019

Double paying is a growing issue in the mobile advertising space. We addressed this topic in a previous blog post on mobile app retargeting, as well as in the third issue of the Mobile Hero comic, “Miss Attribution and the Blackout Window.”

To avoid double paying, we encourage advertisers to implement a blackout window — the time between when a user installs your app and when you start re-engaging them. In this post I will recommend blackout window periods to provide ample protection of your user acquisition campaigns and avoid double paying for your users’ actions.

While I wish I could say “if your app is a game, your blackout window should be X number of days,” or “if your app is a lifestyle app, your blackout window should be Y number of days,” given that each app is so different there is no hard and fast rule. Instead, I will walk you through a process for establishing a blackout window that is ideal for your unique app.

Key Terms

Let’s start by defining a few key terms:

  • Attribution Delay: The amount of time between when a user installs your app and completes an event for the first time.
  • Median Attribution Delay: The median amount of time that it takes a user to complete an event after install.
  • TP90 Attribution Delay: The minimum amount of time that it takes for 90% of users to complete an event after install.

Setting the Scene and Data Gathering

The easiest way to explain how to use your data to determine the ideal blackout window is to share an example.  Let’s say we have a weather app that falls in the Utility category and monetizes through in-app purchases. When running UA campaigns, we are optimizing for a target CPA (cost per action) on an in-app purchase event. Looking at our data for the weather app, we determined that the media attribution delay from install to in-app purchase was 3 days, and the TP90 attribution delay was 6 days.

Setting Up the Right Solution

Now that we determined our attribution delays, we can implement a blackout window for our re-engagement campaign to ensure that we do not target users during the time that most new users are likely to convert without retargeting. Using Median Attribution Delay (3 days) as a blackout window will allow you to protect some, but not most users driven by your campaigns. Instead, rely on TP90 attribution delay (6 days) which covers 90% of the users and gives you a better picture of how long it takes users to complete events.

More Ways to Avoid Double Paying

Implementing a blackout window isn’t the only way to protect your user acquisition campaigns. It is, however, the simplest way to avoid double paying for events when running a re-engagement campaign in parallel. When launching a re-engagement campaign, make sure you setup your campaign in a way that protects your user acquisition efforts and keeps you from double paying.

Learn how Miss Attribution, leader of, maliciously steals UA campaign dollars by retargeting too early. The Mobile Heroes are called in to investigate in Issue #3 of the Mobile Hero comics, “Miss Attribution and the Blackout Window.”