Digital advertising has transformed dramatically as programmatic advertising has become increasingly ubiquitous. This method of media buying — in which advertisers use automated technologies to purchase ad space rather than...
Digital advertising has transformed dramatically as programmatic advertising has become increasingly ubiquitous. This method of media buying — in which advertisers use automated technologies to purchase ad space rather than manual methods — eliminates effort, reduces costs, and increases reach for advertisers.
But programmatic advertising is not without its risks, and can even seriously damage a brand’s credibility and reputation.
To learn more about these challenges — and how to smartly mitigate them through brand safety best practices — we spoke with Eric Wittlake, 6sense’s Senior Director of Category and Product Marketing,
Eric is a former Gartner and TOPO analyst who has long specialized in demand generation, digital, content, and media. In addition to helping companies create and grow new service offerings, he’s led new mobile, search, and lead-gen initiatives. He’s passionate about helping brands safely navigate the waters of digital advertising.
Eric: Ultimately, it’s ensuring that an organization’s digital ads are running in online environments that create value for the brand. For instance, most companies don’t want their ads published on adult websites, or software piracy sites. There are plenty of dark corners of the web, and brands don’t want to be associated with those places — and they probably don’t want their dollars supporting them, either.
Another element of brand safety is protecting the company from fraudulent ad networks or websites. Some websites use automated agents or bots to visit the site and “view” the ads, which artificially inflates content consumption and increases total ad spend.
So instead of your $10,000 paying for 1 million impressions to humans instead is paying for 800,000 impressions to humans and 200,000 impressions to non-humans. This is usually registered as non-human traffic or IVT — invalid traffic. Those aren’t people.
So the question facing organizations is: How do I make certain this is an investment that’s reaching a real audience, and in an environment that’s appropriate for me to reach them?
It is. Which is why it’s so important to determine those controls and how to apply them. Because there’s a lot of stuff out there that you don’t want your brand to be alongside.
There have been examples of major brands unwittingly running pre-roll ads before Jihadi and ISIS videos. That’s an instance where the advertiser approved a major site for distribution, but the ad was placed on a page the brand certainly didn’t want to be on.
It has, but the programmatic environment that’s emerged in recent years has made it worse. As inventory moves through programmatic channels, it disintermediates the relationship between the seller and the buyer — and it makes it a lot easier for someone with fraudulent intent to step into the middle of that.
Back in the “old world” before programmatic, when you had to fill out an insertion order to make an individual site-buy and send it off to somebody, you never would’ve bothered with running ads on these lower-tier sites [where fraud is more prevalent]. But programmatically, you’re now buying inventory from them.
So how do you put controls in place when there’s no opportunity for a human review cycle? And then there’s the flip side of this: You also probably don’t want to give up all of that inventory if you have a way to validate that it’s actually good.
Many companies today use some form of brand safety and fraud detection programmatically as a piece of their investment. They use third-party tools that are nearly standard across the industry. DoubleVerify and Integral Ad Science, or IAS, are two of the top ones.
But over the last five years or so, we’ve also seen companies saying, “I want to actually apply that and also only run on a distinct list of sites.” They want a relatively scaled site list, maintain a pretty high level of efficiency and scalability, and also have that control in place that ensures that they’re not alongside undesirable content.
And at the individual impression level, they’re examining the data and also saying, “Yep, that looks like it’s a real person, behaving like a real person, has that fingerprint and that footprint. I’m confident I’m reaching people in the appropriate context for my brand.”
There is, and different companies have different thresholds — different levels of what they’re comfortable with. Some companies might be willing to run on a much broader range of sites that reduces their cost and increases their coverage. Direct response-focused marketers generally still take a more scaled approach.
But there are other companies that say, “Hey, maybe I’m going to narrow down my coverage to a few thousand sites, and that’s all. Period. Done.” They’ll still put more robust controls in place because there’s still a lot of stuff on the internet that they still don’t want to be associated with.
Generally, every company has to find a balance that’s right for them. That’s one of the challenges here at 6sense as a platform. How do we serve a range of clients that have different levels of sensitivities to ensure that the controls are in place — but that the flexibility is also in place for their price sensitivity and their brand sensitivity?
In three key ways. The first is the relationships that we have with the suppliers we work with. We have requirements in those relationships, so that these suppliers simply won’t offer us certain kinds of inventory. We conduct a lot of diligence to confirm that they’re telling the truth. We’re not afraid to cut ties with suppliers who aren’t on the level with us.
Also, we’ll cut suppliers that aren’t giving us the level of quality that we want, or who’re meeting a minimum threshold but aren’t incremental to the value of what we’re able to deliver.
The second way 6sense solves for this is with our own AI and data science. We’re continuously saying, “How do we identify fraudulent traffic and remove it?” We use our AI to identify that. We use it to refine the inventory we’re willing to buy on behalf of our customers. That’s something that’s distinct to 6sense.
And finally, we use those third-party solutions I mentioned earlier. We work with the leading providers for brand safety, for fraud, invalid traffic identification, to actually bake those solutions into every single campaign that runs on the 6sense platform.
That means our customers have a foundation from a safety perspective. If you want to run narrow to just a few thousand sites or want to run really broadly, we always have a multifaceted safety net in place on every single campaign that we run.