Commentary
Display Advertising 2.0: Brand Safety is Just the Tip of the Iceberg
- by Steve O'Brien , June 14, 2010
In his annual Internet Investment Guide published in January, esteemed J.P Morgan analyst Imran Khan predicted that investment in display advertising will surge by 10.5 percent in 2010. It seems Mr. Khan's predictions are ringing true. According to ComScore, in the first three months of 2010, U.S. Internet users viewed 1.1 trillion display ads or 15 percent more than they did a year ago.
A big part of the growth can be attributed to display's second-generation transformation into something closely resembling the more automated search advertising medium. Display advertising providers, such as DoubleClick, now offer self-service distribution options and streamlined ways to reach the audiences of your choice. Demand Side Platforms (DPSs), ad networks and ad exchanges are also making it easier for brands to quickly gather the exact eyeballs they want online. Or so it seems.
As the number of middlemen between the advertiser and publishers increases, it's also becoming more difficult for advertisers to know exactly where their ads ran and who saw them. Furthermore, the growing opaqueness of display advertising 2.0 brings with it a greater possibility that some may game the system. Advertisers are well aware of this problem and it has caused many to think twice before pouring more of their online marketing dollars into display.
This hesitancy of brands to invest more in the new world of display advertising has spawned a new genre of technologies and innovations to help ensure advertisers get what the pay for. One of the most popular is brand safety technology, which helps ensure that display ads don't appear next to questionable web site content.
But as advertisers begin to examine the initial returns on new display investments, they're realizing that brand safety is only the tip of the iceberg when it comes to ensuring display advertising results. A brand also wants to know things such as:
- How many people saw my ad?
- How many times were they in the target demographic or geography?
- Was the ad above or below the fold?
- Were competitors displayed more prominently?
These offerings track actual impressions, including which ones were real and which were invalid (out of geo, robotic programs, etc.) Others are providing "block lists" consisting of IP address ranges that publishers can reference to determine whether to serve an ad, and advertisers can use to determine if they should bid on impressions or not. Together ad verification technology and brand safety offerings can help to give brands the confidence they need to continue to grow their investments in display advertising just as others have done for the search marketing and traditional TV and print media markets. But until these services become more widely embraced by the media inventory brokers and providers themselves, the biggest opportunity for the display market may lie just out of reach.
Great article. Another thought to consider is how both agencies AND publishers are giving up control in this new black box environment. Agencies (unless they use their own DSP) have no clue how they are delivering performance for their clients when they employ audience targeting. Publishers essentially give up 70% of their audience to another company for an average of 70 cents, and allow exchanges to monetize it at $4.00 CPMs. Until some more transparency enters the equation, both agencies and publishers will be fighting the new technology middlemen for power, control, and relevance in the new "ecosystem." Very thought-provoking...please contribute more articles!
I heard an interesting stat from DoubleVerify that 30% of banner impressions were being "mis-represented" (whether it was picked up by affiliates or black-boxed - just not where it was supposed to be). That alone should be a red flag to publishers - because the new DSP technology does actually give you real-time transparency into the when and where, and if you have someone grabbing your traffic - it's going to bring you down. Here are some interesting things for you to look for with DSP's as well that you're "Mother may have never told you about":
1) Make sure the impressions are coming from the exchanges. Some DSP's who were former networks still claim "exclusive" inventory and put their network inventory into your bids. You can control it if you know about it!
2) When you drill into your data, there are a number of sites that will show up as ad server tags. They've found a way to not display who they are - so cut them out.
3) DSP's charge a different fee for managed versus self-directed - feel free to ask what % they charge. Afterall, it is transparent :)
If display advertising continues to chase search, premium publishers are going to continue to erode their own revenues. There's no question that machine based buying, enhanced by brand safety technologies and data, greatly improves direct response advertising. But if the industry wants to really attract significant volumes of TV advertising dollars, we need to create memorable consumer experiences - and that takes partnership between advertisers and premium content creators. Read more at http://www.adify.com/can-machines-build-brands/
Invisible website hijacking, if not properly handled, will kill online ad campaigns. Savvy advertisers use services to eliminate unwanted traffic.
The technique (Invisible website hijacking) consists of injecting obfuscated (encrypted) Javascript code in web pages, to further and recursively spread a virus, and steal information from unsuspecting visitors, and create click fraud.
The web site that has the nefarious code on its web page is typically a victim. From a non tech-savvy user perspective, the code is invisible, you could also say it is "dormant" just like a human virus.
Typically, the virus spreads by sniffing databases of FTP login/passwords. Regularly changing your login/password (if you are a webmaster), not sharing login/password with third parties, and checking the size and last update timestamp of your web pages (in particular, index.html or index.htm) will help you notice that "someone" (a bot) changed your web pages, typically by adding obscure Javascript code at the very bottom of your page (you might have to scroll down A LOT to notice the Trojan code on your web page, when you open it with Notepad).
In one instance, the web page in question was a redirect, making the redirect URL look bad, but the original page looked good (according to the browser) despite the fact that the virus was on the original page. Indeed, this virus could be used to blacklist any "clean" (non-infected) web domain, via this redirect mechanism. Even Google.com itself could be indirectly hit, assuming that the virus has infected a large number of domains and that in all instances the hijacked web page was turned into a Google.com redirect.