Editor's Note: The following post was first published last November, but is still relevant today:
Publishers’ yield optimization strategies run the complexity gamut. From leveraging multiple yield optimizers in tandem to none whatsoever, this variance is largely driven by the evolving marketplace of the digital world. Still, all publishers are focused on the same end goal: generating as much revenue as possible from their available inventory. Within such a complex marketplace, however, this is no easy feat, particularly when it comes to unsold inventory.
Today’s technology has allowed publishers to address this issue, but often at the expense of cost per mille (CPM). Thus comes the challenge of addressing unsold inventory while maintaining quality pricing.
There are many current yield optimization strategies, but they tend to focus almost exclusively on either direct sales or technology solutions. Instead, most publishers really need something in the middle, where technology is augmented with human understanding. A yield optimization strategy that can meet the needs of both STR and CPM should factor in such variables as sales channel composition (how much is direct-sold?), the topology of inventory (how many sites and ad spaces available and how diverse their characteristics?), and ultimately the resources available to tweak the knobs (an optimization orchestra or a one-man band?).
For example, publishers who sell 70%-80% of their inventory directly tend to prefer to protect rates, typically pricing unsold inventory nearly as high as their primary stock. For such publishers, programmatic solutions that guarantee 100% STR are less than ideal. Instead, they need a solution that can do more than just sell impressions at floor prices. This means taking into account these publishers’ mix of direct sales efforts and partnered relationships, doing an audit of their block list, and augmenting the quality of data on their impressions in order to simulate the bid landscape. Only then can one credibly price inventory that addresses both STR and CPM, and even then that is just an average price. The very best solutions will adjust pricing in real time based on the unique dimensions available on each impression.
On the other hand, publishers who sell a lower
portion of their inventory directly and hence have a larger pool of impressions to monetize programmatically might gravitate to a 100% fill solution, especially if they are strapped for resources. But
this is often done at the expense of eCPM, decreasing the overall value of a publisher’s site. While a 100% fill solution is often an easy way for publishers to offset a large amount of
inventory quickly, it becomes more of a remnant monetization solution. Without taking into account bespoke factors, there are lost opportunities for maximizing yield optimization. The use of
technology must be complimented with analysis to establish the best pricing model for the available inventory. Only then can publishers really obtain their maximum revenue potential.
Are you an Online Publishing Insider? We're looking for folks who can provide insightful commentary about issues/trends in online publishing, and/or relevant tips on how to become a better publisher. Potential regular contributors and one-offs both welcome. Send credentials and one or two ideas (along with writing samples if possible) to firstname.lastname@example.org. Thanks!