Julia Angwin’s recent Wall St. Journal article (about ads purchased through networks showing up alongside racy content) underscores the point that there could be better controls in place around network buys. One observation in particular is a point we bring up a lot around here: in order to gain access to more media, ad networks often buy and sell from each other. So, while an advertiser thinks it’s buying on Network A alone, part of its order may be placed on unknown Network B, and its ads may end up on some objectionable sites as a result.

Loss of control over where ads serve is a very real problem when networks trade inventory. This is why we’re putting so much work into our MediaGuard product. Advertisers, publishers and ad networks using the Yield Manager platform need to feel confident that content and ads within the network are being classified properly.

But while loss of control over content is the most visible problem of inter-trading between ad networks, it’s not the only one.

Why Networks Trade with Each Other
The inter-trading of ads between networks is a means for networks to fill gaps in their supply and demand. Why else would a network sell inventory to a competitor, while another buys inventory from a competitor? This practice is a way for ad networks to create a proxy for liquidity in their closed (and relatively illiquid) markets.

Networks see a compelling reason to remain closed. Think of them like ticket brokers. The great thing about being a ticket broker is that you know how much the buyer is willing to pay for a certain ticket, and you know how much the seller is willing to sell for. Neither the seller nor the buyer has information about the other (otherwise, what would they need the broker for?). The broker can sell the ad to the buyer for maximum price, and from the seller for minimum price, keeping an undisclosed (and often huge) margin.

Now what happens when your friend (who’s also a ticket broker) has the buyer as a client, and you have the seller? Well, you cut a deal and split the margin.

The Resulting Friction
Ad networks operate on different platforms, with different pricing structures, counting methodologies and other disparities. Ask enough network ad operations people and you’ll hear a horror story about an ad being brokered back and forth between two networks, each buying and selling the ad to/from each other several times in a nearly endless loop. In short, when networks sell to each other, complications abound. Controlling content of ads and sites becomes impossible. Buyers and sellers see discrepancies in reporting. Reconciling payments becomes a formidable task. When we talk about these dynamics here at Right Media, we talk about friction.

Solving the Problem in an Open Exchange
This sort of friction is eliminated when networks, publishers and advertisers share a common platform (with a single counting methodology and set of pricing structures). Today there are over 50 ad networks using Yield Manager, interconnected to all other buyers and sellers on the platform, that have open, immediate access to media and trade it seamlessly. If, at a given moment, a network lacks needed inventory, it naturally acquires more in the open media exchange (what we’re currently calling the “Right Media Exchange”). Likewise, if it runs out of paying ads, buyers are available in the exchange.

The point is that networks, like any brokers, will continue to have relationships with each other. But many are realizing that they owe it to their advertiser and publisher constituents to eliminate as much friction and uncertainty from those transactions as possible. With Yield Manager’s community linking capabilities and ever-expanding Media Guard features, we are striving to eliminate as much friction as we can.

As the exchange expands and the benefits of an open, transparent market become more obvious, there should be less reason for any of its participants to look externally to increase scale. With that, problems like ads showing up where they shouldn’t and multi-platform conflict should be greatly diminished.

In the financial markets, there are multiple layers of regulation. Banks have their own compliance departments; the NYSE has its own regulations; and the SEC provides yet another set of rules to protect consumers and financial institutions from fraud and other unsportsmanlike conduct.

Online advertising has yet to evolve a similar set of standards that protect consumers, advertisers, publishers, and networks from each other. Some steps toward this have happened. The government has implemented very specific laws to protect against egregious violations of consumer trust (spam, spyware, etc). More directly applicable to the industry, the IAB, NAI, and other standards bodies have begun the process of standardizing measurement, nomenclature, terms and conditions, and other key principles of industry-wide commerce.

However, these external regulatory bodies aren’t sufficient either in finance or in online media. Often, their rulings lag the market by years, and their arbitration process is too slow to react to the day-to-day issues that companies encounter. So what else is being done? Currently, most ad networks have their own standards and policies. However, these standards are self-imposed, and more importantly, are often subjective. For instance, the Right Media network has a policy that if we make a trafficking error, we don’t spread the cost of the error on to our publishers. Now, that’s a great policy – but how can publishers be sure that we implement it? Obviously, this is a subtle issue, but it’s something that can dramatically affect revenue.

Here’s where our open media exchange comes into play. The exchange (we’re currently considering the name “Right Media Exchange” as a replacement for “Yield Manager Marketplace”) plays the same role as the NYSE does for the financial markets. Not only does it provide the underlying transactional infrastructure for media, it will provide a common, trusted, enforceable set of rules and regulations that all exchange participants adhere to. This doesn’t mean that every member of the exchange has to adopt every policy. However, and this is important, it means that if an exchange member says that they adhere to a certain policy, the exchange will regulate and make sure that they actually do.

Who is this good for? It’s great for advertisers and agencies, because they can be certain that if they ask their exchange-member networks and publishers to run only on particular types of sites, they’ll know for a fact that this is the case. It’s great for publishers, because they know that if they ask for advertiser exclusions or protection against errors and adjustments, they’ll know for a fact that that’s what they’ll get. And it’s great for networks, because they have a partner looking out for them to help them implement their policies.

Another really cool thing about the exchange providing oversight is that we’ll be able to provide audited reach and traffic numbers for all exchange members, with far better accuracy than outside measurement companies like Comscore. I’m not sure why Doubleclick and Atlas don’t provide this functionality for their customers, but it’s something we’ll certainly do (on an opt-in basis).

One crucial part of regulating the exchange is to get our members involved in the process of defining standards, policies, and auditing processes. We’ve started this process already, with our Media Guard program, and we’ll be expanding it as our regulatory efforts continue.

So the bottom line: exchange oversight is necessary to build industry trust, and that trust will translate directly into dollars for the exchange and its members. This is an exciting step for all of us, and we look forward to rolling out our regulatory bodies in the upcoming months.

What exactly constitutes excessive cleavage? What is the difference between a woman reclining to relax and a woman in a supine, sexually suggestive position? I bet few people have ever discussed that at any business meeting, but here at Right Media these issues have been a hot topic (no pun intended). As we build Media Guard – Creative Review and create processes and guidelines, these interesting questions require us to consider community standards, the opinions of our clients and partners, and our own experience.

It’s never easy to define what others will see as obscene or objectionable. Not only does it vary country by country, but it also varies state by state. Often as our work progresses I am reminded of the frustrated Supreme Court Justice Potter Stewart’s 1964 definition of obscene material, “I know it when I see it.� Unfortunately we cannot be as free-wheeling as Stewart suggests – we need a fair and clearly defined process. Additionally, we are not going to make the final decision about what is an annoying creative or what is acceptable to a publisher. Instead we are going to categorize creatives in as detailed a manner as possible, and then allow the publishers to select what types of creatives they want to allow on the website. Of course, the reality is that no matter how hard we try to make a formal set of classifications, we are always going to be subject to human judgment – what exactly is excessive cleavage? That’s where we, the Media Guard team of auditors, come in. Our job is to properly and uniformly classify the creatives, building a set of examples and “definitions� to make sure others can understand what is meant by each category and we take the responsibility very seriously.

Of course we don’t spend all day discussing the difference between an animated bikini and a real one. We also consider how granular our classifications should be, so that the categories will be useful to publishers as a method to accept or ban creatives. For example should we classify creatives with looping audio separately from creatives with a short noise? (yes.) Or should we have a category for creatives whose content is different from their landing page and does it matter what the content of the landing page is in those instances? (yes and yes.)

In determining these categories we bring not only our practical business knowledge to the table, we also bring our extensive experience as internet users. We too have tried to close a creative while surfing only to have it spawn a dozen pop-ups that also need to be closed. We don’t like it any more than the average internet user and know that publishers might want to exclude ads like that from appearing on their sites. That’s why we have a category for it in our audit list, and classify a creative as such if we discover that it uses such tactics.

As our Media Guard project progresses and we talk to our partners and colleagues we realize what an awesome responsibility we are undertaking. Mike Walrath called our team the “guardians of the free world�. I think he was only half-kidding.

The recent article in The Wall Street Journal by Julia Angwin, “Web Ads Appear On Racy Sites Despite Checks,� caused a lot of conversation at Right Media this week. A lot of that discussion centered on Media Guard, which will allow advertisers to keep their content off of sites that do not meet their content requirements. This functionality will be part of the second phase of the Media Guard development, Media Guard – Site Review (MG-SR).

Building out MG-SR and the processes around it will be a daunting task, and one that we are already thinking about. How will we be able to classify the content of all the sites that are in the Yield Manager Marketplace? As an example, how do we classify all of the user-generated content on MySpace, which can be changed and updated at anytime? Do we just classify all user-generated content as such, with no additional classification and effectively make those pages caveat emptor (which we assume will lead most conservative advertisers to choose not to run on pages thus classified)? Will this take so much content essentially non-classified that it ruins the goal of MG-SR? Of particular interest, a recent article in The Economist, “Murdoch’s Space�, discusses some of the issues that MySpace is addressing to be sure it is attractive to advertisers.

I thought that the WSJ article dealt with some much bigger issues than pure content classification, such as network re-selling and advertisers being disintermediated in the buying process, which are beyond the scope of Media Guard. But by tackling these issues of content classification, with MG-SR, we should be able to solve some of the major end problems (ads on inappropriate sites) that advertisers face.

Of course there is a lot more to say about this topic and how we tackle these problems that advertisers face and develop the best solution for them, but for now we are currently focused on getting MG- CR into a live mode - an exciting time! We have several “auditors� who have started and are getting close to the first major lock-down of the classifications for creatives. Then we will move into a beta test with some live ads to get comfortable with the flow and timing. Then after moving into live mode for all creatives, we can really start to focus on creating MG-SR and solving some the problems that advertisers face.

Creative Tester is the automated testing component of Media Guard. The purpose of Creative Tester is two-fold: first, to determine whether a creative is auditable (see the prior post on auditableness); and second, to flag and detect any changes to the initial “behavior” of the creative that was observed. If the latter happens, and the creatives content changes in any way, the creative will immediately be removed from “Media Guard Classified” status until a human can re-review the creative.

After Creative Tester tests a creative and assigns the Technical Attribute tags to the creative, the creative will either go to a human review to assign the Content & Offer Attributes, or the creative will be marked as “unauditable” and will not be reviewed by a human. If a creative does not go through a human audit it will be categorized as non-Media Guard Classified content. A creative will be marked as “unauditable” by creative tester for several reasons, such as the if the creative runs Active X or has any viruses. Additionally, creatives that are 3rd party tags are given special consideration.

Determining auditableness of creatives that are 3rd party tags:

• Is it a 3rd party tag?
• If it is a 3rd party tag, does it always load the same creative?
• What are the creatives being loaded?
• Does the creative always go to the same landing page?
• What are the landing pages the creative sends the user to?

The above tests are used determine the difference between a rotating 3rd party tag, which is not auditable, and a legitimate advertiser 3rd party tag. Creative tester will analyze the 3rd party tag behavior and record anything and everything it finds. If we determine that the creative consistently loads the same offer (say, 4 creatives for NetFlix) that go to the same site (netflix.com) we will pass the creative to our human audit phase. Of course, if we ever determine a change in the creatives downloaded or click-through URLs we will immediately flag this creative for re-review.

Other ‘Auditableness’ Characteristics:

Since we are continually loading creatives to make sure they are auditable we decided to also include some tests for malicious behavior that sometimes sneaks through 3rd party tags. Specifically we will be testing for:

• Strange popup behavior
• Does the creative have an exit pop?
• Is it a banner that is trying to launch a pop?
• Does it load multiple pops or other JS warning windows?
• Does the creative try to initiate an Active-X install?
• Do any of the files downloaded test positive for viruses?
• Are there any fatal JavaScript errors?

We are still determing exactly how we will deal with some types of “behavior” found in Creative Tester and which creatives will not be submitted to a human review. Our primary objective is to keep out very malicious content or content that can change significantly and thus is impossible to classify.