How to deal with advertising fraud in mobile?

Photo: Chartboost
Signs of fraud in advertising, according to Sefert, are as follows:

• the terms of placement of the advertisement are violated (for example, they ordered advertising in America, and the installations come from the Philippines);
• real traffic is not delivered;
• installations are carried out not by real people, but by bots;
• advertising orders are transferred to extraneous grids;
• click spam;
• there is no binding of installations to devices and/or geographical location;
• there are direct redirects from web pages.

Fraud causes direct damage to developers – it turns out that they pay for traffic that does not exist. Or for useless traffic.

There are two ways to resist mobile fraud – turn to technology or minimize the damage.

With technological solutions, everything is clear – they imply a detailed analysis of installations and traffic using appropriate programs. However, Sefert believes that this option is not available to many studios (especially small ones). He recommends that such developers try to minimize the damage. To do this, it is necessary to establish a clear framework at the stage of advertising placement that will determine desirable and undesirable traffic. And don’t pay for the unwanted one. The “right” traffic is the one that gives the retention of the first day (Day One Retention) less than 25%, provides less than 80% of downloads/installations, and so on. In addition, it is necessary to be able to immediately stop the campaign if there is a suspicion that the traffic is fraudulent.

If all the signs of fraudulent advertising are present, Sefert recommends not to be shy and demand back the money invested. To do this, the developer needs to have evidence of fraud and an advertising contract, where the already mentioned clear framework is established. Plus, it will take perseverance.

Source: Mobile Dev Memo