Commentary

The Mobile CPM Train Wreck

Before the start of the 2011 Academy Awards show hosted by James Franco and Anne Hatheway, Franco was tweeting live video clips of himself from backstage.  Right before the show was set to begin, Franco turned to the camera and said, “This could be great, or this can be really reallybad.”

It was an epic train wreck, of course -- one that still haunts the actor today.

At just about that same time period, executives at premium publishers were sitting around conference room tables reviewing site traffic numbers, and were likely surprised at the steady rise in the number of site visits driven my mobile devices. They must have thought at the time, “This could be great, or this could be really really bad.”

The evolution of mobile has been an epic train wreck for non-social media publishers, but unlike Franco’s performance, this can be fixed. First let’s dissect the problem:

  1. Mobile site visits do not generally add to a publishers’ overall audience; they just change the percentage breakout between mobile and desktop.
  1. The number of mobile visits continues to rise month over month. This means the percent of desktop page views is going down and the percent of mobile page views is going up. 
  1. Mobile ad impressions are priced significantly lower than desktop impressions -- and/or they get blended with desktop CPMs -- which has the same effect of lowering prices.
  1. So a premium publisher’s fastest-growing segment of users is leading to significantly lower ad prices (that’s the really really bad part).

advertisement

advertisement

When you read USAToday.com on your Samsung Galaxy, is the reporting any less accurate than if you read it on your laptop? If you check this coming weekend’s temperatures on Weather.com from your iPhone, are they any different than if you checked on your PC at work?  When you update your Facebook status or tweet from your phone, is the experience any different from what you have using your MacBook Air?

Consumers don’t assign lesser value to a site visit from a mobile device, so why are the ads less valuable? Because buyers say so.

Buyers point to smaller screen sizes, less targeting capabilities, and lower time spent per visit as some reasons the value is lower. While these points are all true, what’s also true is that clients need to be where consumers are, and consumers are on their phones. 

Buying and selling media is a tug of war with words, and buyers say shit to get lower prices.  It’s their job.  The words “your site is not working” will be in enshrined in the media buying hall of fame. These words buyers use, along with a supply-demand imbalance, have knocked the snot out of premium prices.

The solution is to fight back and raise CPMs, and make these prices universal across any screen.  The key is to create a rate card with “equivalent” ad units, so for example, a 728x90 is an $18 CPM on desktops, and the “equivalent” unit on mobile is a 300x50 that costs $18 bucks, too.

I am pretty sure Facebook and Twitter are pricing their mobile and desktop ad offerings at the same rate, so this solution may appear obvious to some. But why this is the right time for premium pubishers to raise CPMs may not be so obvious.

A digital media plan includes a list of direct buys on sites the client feels good about running on, and the rest is spent with networks and/or exchanges.  Buyers present the plan to their clients at an eCPM of “X” and makes subsequent quantifiable promises.

93% of programmatic spending is RTB on open exchanges.  That’s because this is the cheapest.  As more spending takes place on that end of the plan, the eCPM of digital media plans will drop considerably. For example, a plan last quarter that came in at an eCPM of $6 bucks shifts a higher percentage of dollars into the open exchanges the next quarter, and the eCPM could drop to $4 bucks.  Here is the opportunity: a plan can absorb the increase in CPM’s from the lower percentage of overall spending with premium publishers, and a buyer can still come in flat or even under last quarter’s eCPM.

I get that talking about raising prices in a column about selling media is much easier than doing so with a 2015 quota staring back at you.  The risk is less revenue initially, and more unsold inventory, but the current strategy of increasing sell-through at the expense of premium CPMs is not working.  Premium publishers would be better off selling ads to fewer clients who are willing to pay more, and then using unsold inventory as promotional impressions for the native programs that come with these advertisers' buys. 

Media plans are balanced between math and emotion, and clients will pay higher CPMs to feel better about running on the sites that matter most to them.  The trick for premium publishers is to figure out which advertisers see their site in that way, and then have the guts to play the CPM chicken game with their buyers.

3 comments about "The Mobile CPM Train Wreck".
Check to receive email when comments are posted.
  1. Bob Gordon from The Auto Channel, October 30, 2014 at 2:23 p.m.

    As a co-founder of a 20 year old advertiser supported website, I have been screaming that the internet should have been priced as direct mail on a per user basis and not as TV.
    The reason this did not happen is that tech guys were put in charge of the busisness side of the web and we see the results. Imaginge if chief engineers were put in charge of TV when it became commercialized.

    Buyers are doing their job, but it takes sales management with balls to say no!

  2. Ari Rosenberg from Performance Pricing Holdings, LLC, October 30, 2014 at 3:14 p.m.

    Well said Bob!

  3. Leslie Laredo from The Laredo Group, November 7, 2014 at 1:44 p.m.

    For many, especially those with little or no sales training, media selling became an “exercise in discounting.” Primarily because sellers didn't understand nor could communicate the value of why their digital ad programs would perform and how to measure success. Sellers from traditional media were not going to sell digital as better and in fact, the CPM train wreck started when digital was “value add” in many proposals. In today’s sales discussions, there are few mentions of how view-through metrics would demonstrate performance and little mention or questions addressing how their clients are using attribution measurements. As we know, for years search kept getting all the credit for performance via last-click attribution. Sellers were ignoring one of the most important features or aspects of advertising…ads start the awareness/consideration process and create triggers or stimulus for interest in products. Google’s Zero Moment of Truth research highlights the importance of the stimulus, but why are media sellers not espousing the role of their ads play in the journey to purchase. Once the interest is established via ads, re-targeting, social connections, CRM and other DR activities can move the consumer through the journey to purchase. Most of the sales presentations were about the site’s/network’s UVs and PVs (which is not what advertisers buy - they buy ad impressions, and PVs/UVs don’t tell the story about how and why the digital advertising program works) and then the pitches reveal “what they have to sell.” The train wreck for CPMs continued when media companies failed to train their sellers how to sell the value of advertising and how to not let the CTR become the metric of success. All advertising, including digital, has to account for a “take action later” scenario. Even though a click is the simplest and easiest “action” most people go to publishers’ websites to read contact and not to click on ads and it is why view-throughs is more important than CTRs. BTW, my definition of a click is “an impulse reaction to a creative” which means clicks don’t necessarily convert (there are several significant studies showing clickers are not converters). So let’s stop the CPM train wreck and start training reps (a plug for Ari and Laredo Group) to champion their “higher” CPMs and discuss the values (implicit and explicit values), the how and why their ad programs are designed to deliver results.

Next story loading loading..