Kirby Winfield,
CRO, Mpire
Hello, Oldtimers. It's a pleasure to "meet" you. I am attempting to write a decent introduction to myself and Mpire/AdXpose from somewhere in the midst of New York Advertising Week madness...as a recovering marketer and current revenue head, I am torn between the desire to make this intro a creative opus and the reality that I have yet another agency pitch in 30 minutes.
Enough about me. Mpire Corporation is the market-leading advertising optimization technology company and maker of AdXposeTM. Founded in 2005, Mpire Corporation is backed by Draper Fisher Jurvetson and Ignition Partners.
We strive to make the advertising experience better for publishers and advertisers. This has not always been an easy task. We initially ran a long tail ad network. As rates were depressed in Q1 2009, we began feeling pressure to fulfill our ad buys via daisy-chaining, or downstream buys from other networks. We did so via exchanges and on networks we felt had high standards, and provided attractive site lists, but performance was all over the map, and we ran into the typical content and fraud issues.
So we developed technologies to combat these issues. We started receiving alerts when clicks happened without corresponding engagement (fraud), when blacklisted sites and networks showed up on our buys, and when bad content appeared in adjacency to our ads. We used this data to start winnowing the bad traffic and requesting credits, make-goods, and chargebacks. And we commercialized the service.
Very coincidentally, the OT admins asked whether I would like to write for the Spotlight around the time we have targeted for a press release and white paper around some of the data we are seeing, and how to use it to make exchange and network buys safe.
The buys referenced in the white paper were delivered in July on more than 20 million impressions to ads from 53 advertisers. These impressions were filled by the initial nine ad networks through downstream daisy-chaining on at least 45 additional ad networks on more than 100,000 sites.
The study revealed that more than half of ad impressions and 95% of clicks in online ad buys were fraudulent. Nearly all the fraudulent traffic was hidden behind numerous layers of nested I-Frames -- ad units pulling ad content from other sources that can hide URLs and in-view data. The test shows significant impression fraud and URL padding in standard run-of-network (RON) online ad buys.
There is a short thread currently hitting the list around this data, and some great questions have been posed, off-list, live, and otherwise. I thought I could attempt to answer a few of them here.
95% fraud? That seems high.
If we were buying on site lists on premium networks, or direct, this would be shocking. We were shocked ourselves when we started tracking performance on our own downstream networks - we spent months trying to shoot holes in the data because it seemed egregious to the point of disbelief. But we have simply come to learn that buying via non-vetted linked partners on major exchanges opens advertisers up to a universe of bad actors and unqualified publishers. When we open the floodgates, and get daisy-chained blindly, we should not be surprised at the numbers. At the end of the day, we as advertisers/buyers have been fooling ourselves in thinking we can purchase real traffic at a $0.10 CPM.
Isn't this self serving for Mpire to release such data?
We're just reporting the news. To the extent that the news is what drove us to get into the business we are in, we are certainly going to publicize it. However, we brought in a third party (Radar Research) to examine the methodology and approve the results. This is simply a window into what our clients see when they use AdXpose.
Who should be implementing solutions to combat these issues?
According to networks, it's the clients' responsibility. According to agencies, it's the networks' responsibility. According to exchange buyers, it's the service providers' responsibility. The fact of the matter is that fraud and bad content have become accepted parts of the online ad ecosystem: Removing them completely is like cutting off an appendage, and no one wants to hold the saw.
There are press releases, lip service, and human resources hurled at the cleanup task, and yet very few players with meaningful reach and budget have implemented solutions to date. Google's 2.0 exchange signals a shift here; buyers are already assuming safety therein, and expecting other exchanges to follow suit. (Disclosure: Mpire currently partners with OpenX to provide transparency on that exchange.)
Maybe the most interesting way to answer the question of who ought to lead the charge on stamping out fraud, waste and abuse in our industry with another question:
Who has the most to lose?
