Adomik’s VPs of Revenue and Client Success, Benjamin Lanfry and Remi Quilliet, had the pleasure of taking part in Google’s “Publisher Talent Academy” on March 22nd, 2018 in Paris. Benjamin was part of a panel with Google and Mediarithmics, where he presented strategies for inventory valuation and yield management to attending publishers.

We’ve shared three general rules for publishers to use when evaluating and improving their yield management: :

  1. Work with “Real” data
  2. Set up a “global” monetization management tool
  3. Adapt your decision-making process and measure its impact

In this article, we’ll share our vision of the big picture, and in the coming weeks, we’ll detail the 3 strategies we recommended  to maximize yield.

Valuing Inventory – today’s picture

Over last 10 years monetization practices have drastically changed. In fact, publishers’ monetization and yield strategies moved from a “hierarchical” system to a multichannel and multiproduct universe. Publishers now deal with:

  • “Advertising channel’s: Direct Guaranteed (MG), RTA, First Look, Private / Preferred Deals, Open Bidding, Header Bidding, Exchange Bidding, Devices, User Data, etc.
  • AND, in addition: affiliation, content discovery, text links, data sales to third-party publishers.

In short, this is what happened:
Screen Shot 2018-05-02 at 15.10.43

That said, we still see some publishers thinking, organizing their team and guiding their decisions using the old “hierarchical” framework. They focus on isolated, local aspects, trying to maximize gains on one specific channel, without considering the collateral impacts, which ends up harming revenue.
To illustrate this, let’s look at three examples of sub-optimal yield management setups that we typically see in this new multichannel universe – all the data you’ll see below is real.

Example #1: Programmatic prices inconsistency

Screen Shot 2018-04-27 at 13.48.44
Source: Adomik Report – Real data from a UK publisher – March 2018 
Here, first look prices are completely inconsistent with the perceived value of the inventory in Open Auction (x20!)

Screen Shot 2018-04-27 at 13.57.22

Source: Adomik Report – Real data from an international publisher – focus on one country – March 2018
And, in the example above, looking at March 12th the price scale seems reversed: first look is sold cheaper than PMP, which is itself sold cheaper than Open Auction.

In both situation, the pricing strategy does not take into account the “bigger picture”, which eventually arms revenue.

Example #2: Poor SSP/backfill integration

Let’s look at the following situation

  • Primary SSP floors are under $0.9
  • Backfill RPM is ~$0.3,
  • Backfill RPM is stable by content categories
  • … but floors are defined by format and traffic sources

We see a misalignment between the criteria used to set backfill and primary SSP floors – we’ve even seen situation where the primary SSP floors were higher than the backfill RPM! Eventually, these biases result in revenue losses for the publisher.

Example #3: undervalued inventory in preferred deals

Screen Shot 2018-05-02 at 17.28.51
Source: Adomik Manage
 – Real data from a Dutch publisher – February / March 2018
In a preferred deal, the impression is sent to a buyer that is not necessarily the highest bidder in the auction. This means that if the deal is improperly priced, the very same impression could have been sold at a higher price, generating more revenue. This is what you can see in the above example: the deal featured on the chart generates a 7% downlift vs. what this publisher would have made by sending this impression to the Open Auction. In addition to this downlift in $$, this deal probably entailed operational costs that are not even accounted for here. Conclusion: less revenue, more costs, not ideal.
To avoid such – frequent – situation, as a publisher, it is critical to think twice before pricing preferred deals. When discussing with your buyers, you should consider both

  • Your own metrics: what is the CPM and competition level you see on your comparable inventory
  • If possible, market competitive data  will also tell you a lot about how you should price PMPs.

These examples clearly illustrate that in this new ecosystem, valuing inventory has become very complex. That said, doing it right is absolutely critical if you want to maximize revenue. Luckily, in the coming weeks, we’ll detail three specific strategies that publishers can adopt to achieve this goal – Stay tuned!

Next Post > 1st Way: Work with Real Data

Tags :

inventory evaluation prices management programmatic yield management publishers Yield Management