A lot has been written lately about the effect that networks can have on publisher ad sales (for better or worse). In the end, it is a publisher's decision whether or not to use ad networks, so I will leave that issue alone for now.
However, I would like to begin a discussion on ways that publishers can increase total direct sales and reduce their reliance on third-party sales.
One way to look at this problem is to divide ad inventory into two categories: 1) Highly demanded (e.g. sold out or nearly sold out); 2) leftover (AKA remnant, network or house).
Highly demanded inventory is the easiest inventory to sell and often commands a high price. Examples include home pages and targeted sections. Since this category of inventory is the easiest to sell, a potentially lucrative strategy is to over sell and then use traffic acquisition strategies to help fulfill campaigns. This strategy can have a large impact on total revenues given the higher CPMs these products can attain.
Leftover inventory is the hardest to sell, but this doesn't mean it is not valuable.
Leftover inventory can be classified as either unsold inventory that was forecasted as available and not sold, or un-forecasted inventory resulting from spikes in traffic or other increases in inventory capacity that were not planned for and thus never sold. The latter case is the perfect place for publishers to use spot market ad networks to backfill sales.
Besides the overall slump in today's advertising dollars, there are several other possible reasons for leftover inventory that was forecasted and never sold. These reasons, which have not been discussed much, include a self-fulfilling sales prophecy, lack of buyer direct access to inventory and a dearth of timely and actionable information about inventory.
Self-fulfilling sales prophecy. Over the years, I have noticed that the sales process itself can often be a self-fulfilling prophecy. Sales reps usually sell the same inventory they sold before, and unsold products, which are not top-of-mind, are not easily or normally included in RFP responses.
Publishers can clearly benefit from using their data to help to provide new and different options to advertisers. By making these products more prominent in the sales process, you will be surprised at how you can increase sales and reduce the leftovers.
Lack of buyer's direct access. The online ad sales process is still very complicated and manual. Publishers that find ways to automate the buying process can further assist in the sale of leftovers by making it possible to efficiently service smaller buys of unsold products and facilitate testing of new, "never tried" products to help advertisers find a "diamond in the rough."
Dearth of timely information. It is often hard for publishers to determine which products in inventory are getting ready to "perish" (i.e. go unsold), let alone come up with a pricing strategy for them. Thus, there is little chance that a sales team can sell these products before they go to networks or house campaigns.
Publishers need to be able to easily view which products should be actively promoted, which is a hard problem to solve because it changes daily! And when exactly do publishers consider it "too late to sell" their leftovers? One week in advance? Two weeks? One month? (Please post your opinions, as we would like to learn the consensus view on this important issue).
We will be drilling into each of the above issues in future posts. Stay tuned.