NEED

Prevent Digital Ad Fraud

The digital advertising marketplace is under sustained attack from increasingly sophisticated bots that divert, steal, and defraud billions of dollars a year.

Clear Analytics.
When bots go to your website to generate pageviews without clicks, the CTR for your entire site is lowered. The result is optimizers divert their ad dollars to sites with higher CTRs, operated by the bad guys.

eDNA provides digital publishers with an accurate understanding of what’s happening on their websites, and makes it easy to identify and police human, good bot and bad bot traffic.

Many anti-fraud platforms use a single pixel embedded in an ad in order to track non-human traffic. The single pixel approach is great for auditing an advertising program after the fact, yet fails to block bots before they hit your site. It’s an inherently reactive approach to the problem.

eDNA takes action on the fraudsters before the page loads, then interactively inspects the client to ensure an unpredictable experience for the bad guys and their bots.

Outsmart Sophisticated Bots.
The single pixel approach relies on data across a vast network of sites yet there is little or no domain-specific behavioral modeling or heuristics based on requested URLs, click path, or speed of navigation. eDNA leverages domain-specific behavioral modeling and machine learning to identify the fraudsters. This is not possible with a single pixel firing from creative.

eDNA can even detect sophisticated bots that fake human attributes like mouse movements and page scrolling.
Analyzers

In order to build up a behavioral profile eDNA uses Machine Learning to continuously monitor many risk attributes, or what we call 'fingerprints'. These include:

  • Unknown devices
  • High risk IP addresses
  • Suspicious login locations
  • Tor usage
  • Whitelist and blacklist countries
  • Velocity checking
  • Geo-location anomalies
  • Project Honey Pot
  • AlienVault Open Threat Exchange

Continuous RISK ANALYSIS

Using the the eDNA API real-time requests for each users credibility is made. Each request is given a real-time risk score based on our machine learning techniques which covers 250+ technical and behavioural attributes. The score -  is returned within 200 milliseconds – if the score returned is 1 the request being made is blacklisted. Where a score less than 1 is returned for the given request,  the requester can either be challenged or be fed bad data. Where a score of 0 is returned the request is considered not to be suspicious and will be able to access your website & applications as usual.
Customized Security

              Workflows