It's common and even expected within the online advertising marketplace to employ an IP blacklist to traffic—a list of IPs that are considered high risk for generating fake impressions and clicks. Companies are given the lists and told to steer clear to prevent ad fraud. But what is the reasoning behind the list? And shouldn't a user have a bit of insight about why an IP made the list, especially if it means blocking potentially valuable impressions?
This is a common complaint among supply-side ad tech companies where the volume of bids directly affects revenue. These companies need to take steps to prevent ad fraud—their reputations and demand partnerships depend on it—but at the same time, blindly blocking traffic based on black-box methods is unnerving and potentially costly. Transparency and a bit of insight around a blacklist can help alleviate this anxiety.
With this in mind, Fraudlogix recently added reason codes to its IP Block List solution and found that sharing some of the insight that we use to flag an IP can be helpful for those companies wanting a "why" before they block. We divided our reasoning, which is based on machine-learning algorithms that look at over 40 different traffic anomalies, into six buckets:
- Non-human Scripts: A browser or server has made calls without declaring itself as a bot and has instead declared a valid user agent where there was no real human user.
- Bot Generated: A device has been modified to call HTML or made ad requests that aren't under the control of a user and was made without the device owner's consent.
- Basic Fraud: Bots or people that continuously visited a domain to count multiple impressions, typically under the same IP address.
- Spoofed: Bots or people that forged their browsers' user agent string.
- Bad IP Reputation: IPs that have a historical reputation of being associated with fake, non-human traffic and malicious or compromised devices.
- Data Center IP: IPs from a data center where no real traffic comes from them.
The added transparency in blocking decisions can help companies confidently filter their traffic for ad fraud and evaluate their publishers and supply partners.