The anatomy of mobile ad fraud: Smart Bots
Smart Bots are a new type of Bot capable of performing in-app activities for up to 30 days after install. This type of fraud is hard to detect because Smart Bots look very similar to real users, with a personal IP, device ID, etc. In 2018 Smart Bots are used more than classic Bots and therefore Scalarr considers them a part of the New Face of Mobile Fraud category.
1. In 2018 Smart Bots were responsible for 21,5% of all fraud cases.
2. Smart Bots were the fastest growing and most dynamic fraud category in 2018, which uses constant reverse engineering.
3. Last year Smart Bots were found in social networks, which have historically been considered a trusted source.
How Smart Bots work
The basic mechanics are similar to classic Bots, but these Smart Bots are much better prepared for the "attack":
1. Fraudsters analyze all advertiser post-install events and KPIs so they can prepare the Smart Bots to be able to avoid detection by many anti-fraud solutions.
2. These Smart Bots are configured to emulate financial events, and sometimes even real payments.
3. They are engineered to use specific app/game SDK versions directly from the store.
4. The timing of their actions are set so that a realistic TTI distribution is seen. TTI (Time To Install) is the measurement of time between clicking on the ad (impression) and the actual install time.
5. Finally their traffic volume is divided into various smaller sub-publishers to emulate multiple GEO locations and avoid detection of too much traffic from one source.
How to deal with Smart Bots
Unlike classic Bots, Smart Bots cannot be detected by manual analysis of abnormal post-install activity. This new type of fraud is very dynamic with constant reverse engineering. It requires a detailed analysis of post-install events using Big Data techniques.
Read Scalarr’s Ultimate Guide to Mobile Fraud Types to understand more about dealing with Smart Bots in all specific cases.
The general principle of “mixes” grounds on the conscious use of several various types of fraud to get over the known protection measures of fraud ...
Fraud is an adaptive crime, so it needs special methods of intelligent data analysis to detect and prevent it