In 2004, I was recruited to Microsoft, where among other things I was put in charge of coming up with a new plan for the overall advertising experience for MSN and, soon after, Windows Live. I spent about eight months digging into the advertising experience as it then existed and tried to rationalize how advertising should work on a major site like MSN and across a variety of user experiences.
In an early meeting with a group of folks from the sales team, Gayle Troberman made a fateful suggestion: "You really need some kind of framework for assessing what kind of ad fits in what kind of experience." This was a key suggestion because it forced me to assemble a cross-disciplinary team and create a shared language that drove numerous long-term decisions.
The first-order considerations were driven by "user modality," which is defined as the behavior and related mindset that a user is engaged in during specific activities. We needed to determine which advertising experiences were acceptable in each type of modality that existed across the myriad experiences on our properties. By carefully considering modality, we were able to create a set of guidelines for what advertising should be enabled in each type of environment.
To illustrate the point, let me give a few key examples of what we put together:
- Users who are reading email are open to advertising experiences that are relevant and non-invasive, but that are not explicitly targeted to that user based on the content of the mail -- which just is creepy.
- Users who are writing email are not open to advertising experiences.
- Users who have sent email are open to a broader ad experience with a larger format ad.
- Users who are reviewing a home page or section front are open to a large format ad.
- Users who are reading an article are open to non-invasive ads that can be large format as long as they don't encroach on the reading experience.
The guidelines I created with that team quickly became the overall framework used by Microsoft to drive advertising experiences across all content experiences across MSN, Windows Live, and even in a variety of emerging media experiences. My "day job" at the time was managing product planning for emerging media, which at that time included video, over-the-top television, mobile, video games, software applications, and new device formats (e-readers, tablets and other device prototypes, Zune, etc.).
Some key principles that I came up with include the following:
- Ensure that ad clutter is kept to a minimum. It's better to have one very large-format ad on a page than five small-format ads.
- Ensure that ads have enough white space around them.
- Give the user the ability to give feedback about ads (both positive and negative) -- such as rating ads.
- Be transparent about behavioral targeting of ads, including how an ad was targeted to them and what profile information we stored about users. Enable users to correct and enhance their targeting profiles. (This was the most controversial of my recommendations and was discussed at length.)
- Enable every ad unit to become "rich media enabled" with specific templatized enhancements, such as a store locator, a pop-up video unit, RFI, and others.
Like many efforts I've been engaged in over the years, this one met with a mixture of success and failure. It took almost five years before we enacted most of the privacy and targeting features I recommended. And none of the rich media templates ever saw the real world. But the user modality guidelines were a huge hit -- maybe in a sense these were too successful. Sometimes the creation of a set of clearly defined "rules" empowers folks who are embedded more deeply in an organization to say "no" to next efforts very quickly. This is often the case with any standards effort, whether at the industry level or within a specific organization.
I experienced this one day when I was trying to roll out a new set of ad formats for software applications. I sat down with the product manager in charge of the effort, and when I started walking him through the prototypes, he quickly stopped me with a clear set of concerns: "Uhm... look -- these ad formats clearly don't fit the ad experience framework we use here. So I'm just going to have to say 'no.'"
Of course, once he learned that I had written those guidelines, the conversation was reopened. But this is an important lesson. Core principles always need to be flexible enough to allow testing the edges and borders of experiences. Once a new content experience is rolled out, an ad experience needs to be tried out with it. Sometimes that new experience doesn't fit in the guidelines you've created.
"Electronic news in the internet" image via Shutterstock.