Why 97%?


We’ve gotten a huge amount of feedback on our announcement yesterday that we will provide our customers a 97% fraud free guarantee starting in January. The reaction has been extremely positive, but one question has come up consistently, no matter whom we talk to. Why 97%?

Why not 99%? Why not 100%?

The simple fact is that the bad guys are really good at what they do and getting to 100% is just not possible right now.

Even running on a custom white list of premium publishers will not get you to 0% fraud. According to Integral Ad Science, that kind of campaign can yield a 2% fraud rate. DoubleVerify has found rates of 2-7% on similar campaigns.

It’s similar to the challenges faced by online banking as it grew in the late 90s and early 2000s. Phishing emails begat keystroke loggers begat zero-day browser exploits. It took the banking and security industry a decade to work out strong defenses and good consumer protections.

Even today, a system we regard as pretty mature experiences a steady trickle of fraud and occassionally massive losses. It is, unfortunately, the price we pay for the convenience of digital finance.

That same arms race is now happening to Programmatic Marketing. This approach to digital advertising has grown so big so fast, and the volume of money is so great, that it has attracted the attention of the cream of the black hat engineering crop.

What do I mean? Some of the best technical talent in the world is trained on defrauding this industry.

We interviewed Aaron Doades, Vice President at DoubleVerify in preparation for this announcement. We asked him about this:

Question: You’re telling me that the people who are writing code for ad fraud could get jobs at some of the top technical companies in the United States or the world?

Answer: Absolutely.

Here’s Aaron explaining more.