ROS inventory is typically sections, pages, and positions that lack sufficient commercial context to drive premium rates. It's usually better than remnant (which is what is left over after the premium and ROS slots are sold) -- but not by much. It is priced at the very bottom of the rate card, frequently "forced" in a package with premium buys, to help publishers better utilize their inventory and to extend the reach of tightly-targeted campaigns. On the back-end, its performance can be all over the map, given the fact that the placements are all across the site and the audience composition is frequently much lower on reach and much higher on frequency than most buyers would like. This is why frequency caps are often requested on ROS campaigns.

When it comes to post-campaign reporting, ROS campaigns continue to underwhelm most online media buyers. There is rarely anything more on the reports than the flighting dates, the impression volume delivered, and the click-thru rate. A buyer will almost never get a report with impressions and clicks broken down by section or page. The sections and pages that it was delivered on are generally not ones that buyers want, so reporting that would either remind them that their ads were not delivered in the best context. Or, if the opposite were true, and some of the best sections and pages were used in the ROS, communicating that to the buyer could undermine the rate card for that section, since the ROS inventory tends to price at a fraction of the pricing for premium placements.

So, now that you know that ROS will not be our industry's poster child during Advertising Week in New York, why should you spend very much time thinking or caring about it? Isn't it just one of those techniques used to make online advertising work better, much like broadcast has spot TV sales and DRTV (Direct Response Television)? Actually, with the advent of behavioral targeting, we are starting to see a whole new kind of ROS, one with more and more significance.

In a world where ads are targeted to segments of people, not sections or pages, R - O - S can take on an entirely new meaning. Behavioral targeting is all about segments (such as in-market car buyers, young parents, gadget geeks, etc.). Thus, the new ROS is "Run-of-Segments." Rather than scattering ads around different sections, the ads are scattered around different segments of users.

Does it matter? Isn't untargeted inventory still untargeted inventory, no matter what you call it and whether it is measured by the person or by the page?

It does matter -- for two very important reasons.

First, in the world of behavioral targeting, where analytics is a critical driver of the entire process, the post-campaign measurement is sometimes as important as the targeting itself. For example, while campaign reports in the traditional ROS world have very little to offer, in behaviorally targeted campaigns reports on ROS can offer a lot. They can tell advertisers and their agencies how ALL of the segments received and responded to ads. Instead of just reporting total impressions and total clicks, a mobile telecommunication advertiser, for example, could learn about how its family-shared-minutes plan offer resonated differently among parents, young boys, golfers, home-owners, office workers, and small-business decision-makers. This process can be very powerful, working to forge insights into tools to help shape and optimize campaigns. ROS can also give advertisers indicators of what target audiences they should be talking to and where they need to improve their creative messaging.

Second, behavioral targeting is so new, and there is still a limited amount of demand for particular segments, so publishers have no reason not to share segment-by-segment performance data with buyers. Everyone is still in the learning stages with this type of targeting, so most everyone is happy to share information if it can grow the market. Providing detailed back-end reports for untargeted campaigns -- the New ROS -- is a great way to introduce the capability to agencies and their advertisers, and it puts a lot less pressure on the publishers when it comes to inventory management.

So, do I think that the "New ROS" will be the poster child this week in New York? No. We've got a long, long way to go before an emerging technique in online advertising becomes the buzz at a large, traditional advertising event. It will be enough for online advertising itself to be the poster child.

This commentary is insightful. I recommend it to others.