This year, US advertisers will spend close to $240 billion on digital advertising in an effort to reach and engage consumers who are new to their brand, as well as re-engage existing customers. The budget size speaks to the important role that digital advertising plays in growing businesses.
Unfortunately, the sizable pot of money also attracts a host of nefarious actors who seek to bilk digital advertisers and publishers alike. Ad fraud will siphon some $80 billion from legitimate players – that’s $1.00 out of every $3.00 allotted to this critical business-building activity.
There is no easy solution to fight ad fraud. Ensuring that ads are seen by real users in brand-safe environments requires multiple strategies and industry-wide cooperation. Fortunately, a tool the ad industry already embraces for targeting purposes can also be added to the industry’s anti-fraud arsenal: location data that is derived from IP addresses.
How IP Address & Intelligence Data Can Spot Bots and Fraudulent Traffic
Let’s start with the basics, what exactly are IP addresses and intelligence data? IP stands for Internet protocol, which is a set of rules that govern the format of all data that are sent via the Internet. An IP address is a unique string of numbers that can identify an Internet-connected device.
There is a lot of intelligence that surrounds IP address data, including the precise geolocation data (city, state, and ZIP code), which is incredibly useful when it comes to validating ad clicks and app installations, as we’ll see below.
What’s more, this data also includes other critical context – or intelligence data, such as whether an IP address is connected to a VPN, proxy, or darknet. Today, a host of entities, including mobile measurement and attribution companies, leverage this insight to detect fraud on behalf of their clients. Let’s look at how they’re using it.
One of the most important ways that IP intelligence data (or IP data) can help the digital advertising sector fight ad fraud is to detect fraudulent clicks and app installs, thereby ensuring that budgets are spent on real impressions seen by real humans.
Here’s how: Location data can help verify that an ad was shown to the intended audience. For instance, IP intelligence data can identify where ads are viewed, and determine if they are seen in a region of the world that makes sense for the campaign. If not, it may be evidence that the click or app installation came from a click farm. Additionally, IP intelligence data can be used to identify proxy data, which in some instances is masked IP data used by fraudsters.
Let’s see it in action.
Click & App Install Fraud Detection
Fake app installs will cost marketers an additional $20 billion, according to AppsFlyer, a leading mobile marketing analytics, and attribution platform.
IP data, when combined with other forensics, can help security teams and fraud-detection companies assess if an ad click or app install is legitimate or fraudulent. For instance, IP data can be leveraged to identify when a suspicious number of clicks come from a specific radius or timeframe, clear signs that they are stemming from a click farm. Once the suspicious clicks or installs are investigated, the ad measurement company can share that information in order to stop that click farm from committing crimes against other advertisers.
IP data can also identify mobile proxy farms by assessing which mobile IP addresses are legitimate, as well as identify mobile IP addresses that never move (an unlikely scenario as real people carry their mobile phones with them as they go about their days). A mobile device that remains stationary is likely to be evidence of a mobile proxy farm.
Another strategy is to compare the entrance and exit nodes of an IP address in order to identify instances in which bot traffic is blended in with residential traffic. Bot traffic typically enters from one location, say Russia, and exits via another, typically in the region where a campaign is targeted.
Finally, IP data can identify a group of interesting IPs that appear in a campaign log, but can’t be connected to a logical source. In such cases, the media agency or brand can escalate the traffic to their fraud prevention provider to investigate.
IP data by itself won’t protect the digital ad tech industry from ad fraud, but it will provide important context around traffic, and help to distinguish between legitimate and illegitimate traffic. By collecting and sharing this insight, the industry can put a serious dent in ad fraud.