• by Jon Espejo , Op-Ed Contributor, April 22, 2016

These dirty words are thorns in programmatic’s side, and—according to some—the reason to stay away from the opportunity. But that’s a lazy argument, thrown around by people who see the problems but don’t take a nuanced approach to dealing with them.

The reality is that non-human traffic (NHT) is not only detectable in the programmatic marketplace—it’s mostly avoidable.

To be clear, it’s true that non-human traffic is prevalent in the programmatic marketplace. Those who vastly underestimate the impact of NHT are just as guilty of false narratives as those who say it’s cataclysmic. According to Integral Ad Science’s Q4 2015 Media Quality Report, 11% of network and exchange inventory is fraudulent. By comparison, publisher direct fraud rates are 3%.

But what if I told you there was a way to get NHT rates on exchange inventory down to 2%—or even as low as 1%—by utilizing today’s technologies? It’s entirely possible to get lower fraud rates buying on the programmatic exchanges than by buying the old-fashioned way.

Here are five tips marketers can employ to reduce their NHT rates from over 10% down to 2% (or lower):

Deploy third-party advanced measurement technologies. You don’t have to fight ad fraud alone. Vendors such as comScore, AreYouAHuman, WhiteOps, Integral Ad Science and DoubleVerify can empower you or your media buyer to detect and mitigate ad fraud.

Take control over your own campaign data. You or your media buyer should be able to make decisions on an impression-by-impression basis. Don't just rely on domain-level reports or creative summaries to fight ad fraud. Many advanced measurement technologies offer log-level analytics, so you can create tools that will minimize ad fraud based on past impression activity.

Measure your direct buys in addition to your RTB-enabled buys. Just because you trust a certain publisher doesn’t mean that site is free from fraud. That’s because fraudulent activity doesn’t necessarily begin and end with inventory. Bots are user-based, and it’s actually more difficult to optimize direct buys since you or your media buyer do not have the luxury of optimizing around various signals (without an ad blocker). As a result, it’s possible to do a great job of not serving ads to a particular bot with your RTB-enabled buys—but still serve them to the bot through your direct publisher relationships. You must measure both channels to account for this.

Pay attention to inventory trends. Supply-side platforms (SSPs), exchanges, ad networks, and other inventory aggregators are constantly bringing on new supply. Their ideas for “brand safety” may not meet your needs. For example, a publisher may be fraud-free most of the time, but fraud-filled during the holiday retail season or at the end of a month. Monitoring these inventory trends is extremely important to maintain the quality of your ad buys to avoid various types of ad fraud (bots, pixel stuffing, etc.).

Optimize regularly. Fighting fraud is not just done on a company-wide level. Fighting fraud on a campaign-by-campaign basis is no different than optimizing to a key performance indicator (KPI): You need to do it regularly in order to stay on top of potential fraudulent behavior—or get ahead of it.

Marketers who follow these five steps can reduce their fraud rates by 80% or more, even when operating primarily in an exchange-based ecosystem.

The war against fraud is an ongoing battle, but it’s one the industry is well-equipped to fight.