The Market Forces Killing Display Advertising


Display advertising is at a crossroads. Its core foundation, the publisher, is crumbling. The latest technology advances are little more than band-aids
that invite scrutiny. The channel that once seemed destined to define digital
is latching onto as many operating principles from Search as it can. It might be time to make a deal with the devil.

From my perspective the causality is two fold:

Market forces:
Optimization conflict on the buy side.

Technology forces:
Relevance is rarely achieved in a user-controlled medium by anyone except the

So WTF is wrong with this market and can it be fixed?

A major failing of display media is that market participants
are in conflict with one another each optimizing for selfish interests — advertisers
trying to get the most for the least, networks trying to allocate impressions
efficiently, exchanges with bottomless “shares outstanding,” data hawkers
opening trench coats lined with user profiles, while yield optimizers are doing
their best Drew Rosenhaus impersonation.

In the midst of massive fragmentation like this there is
always opportunity. In the case of display the opportunity is for performance
marketing. If you’ve gone one click into from the homepage to “premium”
content you’ve seen the opportunity. It’s the original business of the web. Arbitration. In many ways despite the inevitable future FTC inquiries into the use of Oprah & Barbara
Walter’s likeness it has a certain model of purity. It's vigorish that you control based on your performance. Game on.

This is only getting bigger. The “brand”
dollar world that display thought it was building is DOA. Sure digital spends will increase but that brand money is going into
building digital assets, not ads. For the “move the needle” spenders online
isn’t going to eat into TV anytime soon while radio and print budgets seem gone forever.

There is another opportunity. Fix the underlying problem
with online advertising. Build advertising as web services. This requires new technology and creation of new market
platforms. So be it. The channel is only sustainable when publishers and advertisers capture an
equitable amount of performance value from the audience and content. This is the only type of market you can have when media is attributable. It is also precisely why Search is so efficient.

As it stands now this value delta between advertisers performance
and publishers revenue is huge and growing. Is the true value what someone will
pay for it or is the value in how it will perform? Of course the answer is both
however this doesn’t work unless there is a meeting close to the middle.

Here’s a back-of-the-napkin equation (understanding some readers will pick at it) to see the media value

6,000,000 Impressions @ a brazenly high $1 CPM =  $6,000 (Publishers gets portion of this while ad networks
getting 45-60% margins

.10 CTR on creative

7% conversion rate (if you want 10% give me a call 😉

= 480 leads @ $40 value per = $19,200 Gross or $13,200 NET (assuming they only sell the lead once).

So I’ll make the case that publishers are getting about 10-20%
of the total value of their audience & content. Even if I’m off by a factor this is not a sustainable model. Worse off, there are not other ad dollars coming in if performance media can’t
keep performing — display’s current catch-22.

I don’t see the new or current buying and technology systems
making things more equitable. In fact from what I have seen already exchanges
are going to drive down media costs even further and become a new haven for
performance advertising at the expense of pubs.

The solution falls on the back of publishers, as it should,
to create new platforms with new revenue models. Pubs need to solve the problem
by removing the fragmentation, leveraging their internal sales teams and owning
the technology. Instead of deals with the devil pubs must recognize that
they alone have the most valuable and constant parts of the media equation, traffic
& data. They need to own it.

11 thoughts on “The Market Forces Killing Display Advertising

  1. PubMatic is a startup that helps publishers increase the yield of remnant online ad inventory; think of it as A/B testing and behavioral targeting for publishers, with an eye to increase yield of inventory sold through ad networks & exchanges. A few days ago they put out a report that covers ad pricing trends for the remnant inventory they sell; as such, this data is specific to network/exchange-sold inventory, not inventory sold directly by publishers.
    They’re seeing a rapid & consistent growth in ad pricing since a Jan ’09 bottom. One possible explanation for the data is that behavioral data, testing and applying the search auction model can & is being applied to increase publisher yield.
    A summary of the report is here:


  2. Hey Chris- thanks for stopping by.
    I’m familiar with that report and it’s great those pubs that are getting better CPM & certainly your reasoning may be valid.
    You also may have seen the NYT digital (,, et al) announced huge revenue losses last quarter so I’m not sure where we net out.
    What I do know is that for display the costs seem to be cheaper on the exchanges and the performance seems to be the same.


  3. Jonathan:
    Good post. We’re at a point where ad networks need to choose if they are primarily buyside or sellside focused. There’s a lot of value to be created on both ends of the equation.
    Publishers willing to take risk on their inventory will naturally be able to make more money than taking CPMs that are offered to them by networks. If they data they have or they can get that they can apply across their inventory truly is valuable, then they should be able to use it to generate more revenue.
    Publishers historically have been able/content to sit back, take orders and see the checks roll in, and (sometimes not so secretly) complain about many of the same arbitrageurs coming in, buying a lot of their inventory at cheap prices when they couldn’t fill it at the unreasonably-high CPMs they were able to sell to brands. The top end has been getting squeezed, and the margins at the bottom mean not much left for the pub.
    So in theory, pubs getting arb dollars makes a lot of sense (and is one of the things I was working on at Root Markets a few years ago– long story there). One of the problems, however, is that just as the average advertiser doesn’t know (and probably shouldn’t have to for the most part) how to navigate the new ad exchanges and marketplaces popping up, so publishers do not for the most part know how to navigate the CPA/offer-driven DR marketspace.
    Right now optimizers like Rubicon or Pubmatic are mostly shuffling the deckchairs, shuffling between ad networks, but there really hasn’t yet been a paradigm shift in the risk-participation model of publishers, using their own data to drive higher revenues. They won’t get their themselves, they will need better infrastructure and access to new kinds of ad markets not-yet-built to do it… but I’m very excited about how the display space will change in the next couple of years. That’s why we’re in it. Much left to do. Bring it on!


  4. Jonathan, whilst I agree largely with alot of the great points you make – what you say here is extremely significant:
    “What I do know is that for display the costs seem to be cheaper on the exchanges and the performance seems to be the same”
    Does this not then signify that the pricing of such inventory is currently over valued? Much in line with Rob’s point – if it is valuable enough, it will make money surely?


  5. Paul,
    My argument is that the true value of the digital media is based on advertiser performance.
    It’s hard to say it’s overvalued when is many cases it’s not being purchased at all. In theory I believe that exchanges can “correct” the value of media however on an exchange the insane amounts of impression liquidity is a market force greater than performance.


  6. Jonathan, I couldn’t disagree more with your post if it jumped off my monitor and started strangling me.
    The unspoken and oft ignored factor that is killing display is marketer incompetence.
    …Let that sink in for a minute.
    Banners are poorly sold, wrongly executed, misjudged, terribly developed/designed, and woefully reported on.
    I’ve done an extensive piece on this for more of my thoughts.


  7. Hi Daniel,
    While that may be true for some I know many marketers that are killing it using display advertising getting boatloads of leads and huge ROI. I also know Ad Networks that are making a killing — so it’s hard to say that marketer incompetence is to blame. There are many that are doing quite well in this current environment.
    Your analysis too seems to point to “market” incompetence. You can’t blame every facet of the ecosystem and then point blame to those that try and work within its confines. That’s seems to be missing the larger point, especially when as mentioned there will always be smart marketers taking advantage of the fragmentations and inefficiencies to make a large profit.


  8. Just a note to Chris if you are still following this thread JP Morgan just announced no recovery in sight for display and 12% decline in spend rates in Q2 09. All bad news for Pubs & BT providers.


  9. Completely agree we are in desperate need as publishers of a better system.
    I commented on Fred’s post today (under your comment) that we are going to see a complete overhaul of the Internet distribution system. The fair value of influence and sales can only be measured if we can completely track usage/behavior.
    To truly measure each piece of the distribution chain all the processes that lead up to product sales and service/subscription purchases needs to be made totally transparent.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s