Click to install time (CTIT)
Click to install time, or CTIT, measures the time elapsed from the moment a user clicks on an ad to when the user installs and opens the respective app. This metric is often used by marketers to detect mobile ad fraud such as click spamming and click injections.
What is CTIT?
With mobile app fraud and data manipulation becoming a common occurrence in today’s mobile marketing scene, click to install time (CTIT) metrics help marketers detect suspicious behaviors and prevent future damage. CTIT data is compiled across multiple different campaigns and apps to identify any alarming activity by publishers, sub-publishers, or their affiliate ad networks, specifically picking out anomalies that stand out from an app’s installation pattern.
Note that in a mobile marketing context, “install” generally refers to the first time an app is opened. If a user downloads an app but leaves it unopened, this does not count as an install, and the click to install time will only be applied once they open the app.
How CTIT can help with detecting ad frauds
Detecting click spamming (click flooding)
While most click to install times are normally distributed, if there is a high rate of an abnormally long CTIT, this may be a strong indicator of click spamming. A long CTIT signals that there was an excessively long period of inactivity between the first click to its eventual install. This is because most click spams take time for potential, if any, installations to take place, and the inability to detect these illegitimate clicks results in marketers having to pay fraudsters.
Detecting click injections
If there is an abnormally short CTIT, this may indicate that click injections have taken place within your marketing campaign. Typically, in most CTIT distributions, the amount of time it takes for users to click on an ad and install the app is very similar. However, installs that have occurred due to click injections normally display an excessively short CTIT, and this anomaly is a major signal for a fraud attempt where sub-publishers are taking advantage of the last-click attribution model used by MMPs. Failure to detect injected clicks can result in marketers having to deal with distorted data and giving attribution to fraudulent sub-publishers.
Ultimately, by looking out for anomalies and excessive deviations from your compiled CTIT data, you can reliably point out fraudulent activities and block these servers accordingly.