Ad Networks Are For Idiots -- And Here's The Math To Prove It

Ad networks are a tax on lazy publishers.

They are a cancer that slowly eats away at you from the inside, doing severe damage even though you feel fine. They are a cancer that has spread to nearly every publisher, and threaten to do irreversible damage to our industry.

Yet our response is lame and predictable.

We sit at countless industry conferences, self-righteous, nodding our heads in collective agreement about the perils of ad networks. We make the same boring, emotional arguments to feel better about ourselves. We have high hopes and good intentions, like the first day of a Weight Watchers program.

Then the next day we eat cake.

We are slaves to the short-term need to make the quarter; addicts who take the revenue boost despite the pain we will endure later. What is the cost of this reckless lifestyle?

Put another way: Would you risk your entire business for a chance to make one percent? Ultimately, that is exactly what you are doing. Here is a quantitative analysis of the bet you are making:



Competing for Scraps:
First, it is important to understand that the revenue distribution among publishers is an extremely non-linear curve. The top 10 publishers in the U.S. capture more than 70% of every online advertising dollar, and the top 50 publishers capture a whopping 91% of all online advertising dollars.

As for the other 9%? There are estimated to be more than 250,000 publishers using Google AdSense alone.

The long tail is very long, very crowded, and very poor.

It Only Makes Sense if You Can't Count:
Ad networks now generate a whopping 27-cent average CPM, while premium publishers rake in a $20.17 average CPM through direct sales.

Publishers earn 74.7 times more money per page when they sell ads direct.

The problem is that the average premium publisher only sells 30% of its inventory direct. This leaves every publisher with a critical decision: What to do with the other 70%?

This is where people start to use bad math.

If a publisher sells the entire 70% of remnant inventory through ad networks, that is equivalent to selling 0.93% more inventory direct. Less than 1%!

Figure out how to sell an incremental 0.93% as premium by innovating, not by betting the farm.

Dumber than Buying Lottery Tickets:
You are collectively paying more than $2 billion a year to ad networks to create a competitor to your own sales force, devalue your content and brand, and ultimately drive down your own CPMs.

When you buy a lottery ticket, the average value of that ticket is 20% to 50% of the cost. Lotteries are smart enough to realize that if they don't give you some of the money back, you will stop playing quickly.

Yet ad networks take a product worth $20.17, and give you back 27 cents.

After you include fees and unsold inventory, they often take more than 50%. You would be outraged if the government demanded more than 50% of your money in income tax, or a charity wasted more than 50% of your donation on administrative fees, so why do you accept this from an ad network?

If you are a very small publisher that cannot afford a single salesperson, then ad networks may be your only option. For all the rest of you, you are just being dumb.

The Right Answer:
Publishers need the courage to take the short-term revenue hit and turn off ad networks. It is the only chance you have to keep the cancer from spreading.

Search was just a hobby until Overture cracked the code on how to monetize it, and Google spent the next decade refining it. It is time for publishers to spend our collective efforts to crack the code on display advertising and start innovating on ad formats, reporting, measurements, and sales channels.

I will be at AdTech in a few weeks. I will be the guy wearing the "I Hate Ad Networks" pin and painfully listening to another boring argument about why ad networks are bad.

Want a matching pin? Let me know in the comments.


Hear Ye! Hear Ye! King Content Requests Your Input For OMMA Publish!

If content is king, then what is his Majesty's next move in this troubling environment? Is this the time to expand territory and grab land, experiment with new technologies for extended reach? Or should you trim sails, keep the powder dry and stay afloat in a storm? At the upcoming OMMA Publish, June 17, we will engage the best tactics and strategies for publishers in a challenging year. We are soliciting panel and speaker proposals around these topics now. If you're interested in speaking or leading a panel, please fill out the Speaker Proposal form here (only those who fill out the form will have proposals considered), or email Steve Smith at with more general suggestions.

56 comments about "Ad Networks Are For Idiots -- And Here's The Math To Prove It".
Check to receive email when comments are posted.
  1. Jason Krebs from Tenor/Google, April 9, 2009 at 1:24 p.m.

    Amen brother !!

    If a publisher doesn't know the name of every advertiser on its site, how many impressions that adv. is buying, and the exact price they are paying to be there, said publisher deserves what they are getting.


  2. Larry Downes from Emmis Communications, Indianapolis, April 9, 2009 at 1:35 p.m.


    #1 - I want one of those Pins!!

    #2 - I've been preaching this same message to my radio brethern for 3-4 years. Not just for display and search advertising on our websites but for streaming and video pre-roll ads as well. I have yet to figure out why anyone would trade pennies for dollars with ads that ultimately devalue your advertising products as well as your content! You nailed the reason--they're being lazy or dumb. I'm not sure which is worse.

  3. Jeff Skaggs from CraveOnline, April 9, 2009 at 1:37 p.m.

    "Yet ad networks take a product worth $20.17, and give you back 27 cents."

    It's funny how you're "quantitative analysis" assumes that all networks are the same and all provide the exact same payout to publishers. "Bad Math," eh?

    Nice try. Publishers that "get it" know how to work with the right network partners to effectively monetize their unsold inventory, develop innovative site integrations and sponsorships, and still position their internal sales teams for success.

    Those that don't will continue to sit on their unsold inventory and reminisce about the days when advertisers cared more about targeting SITES than targeting AUDIENCES.

    Maybe if your article wasn't so obviously myopic and one-sided it would carry some weight. Unfortunately, though, it just comes off as desperate.


    - Proud Employee of an Ad Network that doesn't work with Idiots and is not a Cancer

  4. Jim Fox from Mad River, April 9, 2009 at 1:39 p.m.


    Spot on! Finally someone speaks up. Problem is unless the entire online publishing community turns off the ad networks or forms some sort of OPEC of online inventory the industry is screwed.

  5. Arlo Laitin from Cardlytics, Inc, April 9, 2009 at 1:49 p.m.

    I agree with Jeff that not all networks are created equal, but the point is well taken. What publishers need to do is find key network and aggregator partners that complement their direct sales efforts, and bring value to their process. Unfortunately, most publishers act against their long-term interests by taking the quick buck, instead of taking the long view. Not only are most networks causing channel conflict, but they are dropping cookies on your users, and selling the data that should rightfully yours. Not to mention telling agencies they can "cherry pick" specific sites, and undermining their publisher "partners".

  6. Chad Little from, April 9, 2009 at 1:54 p.m.

    So let me understand the proposed solution.
    Each publisher should get their own sales rep.


    How about we do a detailed/difficult/intense analysis on that idea....

    Let's assume that takes place and some fraction of the publishers out there do it.
    (hang on computing here....carry the 1....move the 4....done!)

    Looks like we end up with thousands of sales reps calling on every advertiser and agency out there.

    no scale issues there....sounds perfect.

    wonder why it hasn't happened yet......

  7. Sarah Romer, April 9, 2009 at 1:55 p.m.

    Hi David,

    Your argument is on point in regards to the large ad networks that are selling ad space for CPM's lower than $5 and then splitting profits with the publishers.

    You have neglected to mention that there are some small, niche ad networks that are not selling inventory for $0.27 CPM's and actually delivering revenues to publishers that mirror what a publisher could get if the ads were sold directly. is not one of the top 50 or even 250 publishers, but we have a loyal audience and strong brand in our region. We have been successful selling online ads in our local market, but have trouble attracting national advertisers because of our size and reach. We have turned to a few small luxury ad networks to help out by offering our site as part of a collection of affluent adults. The value these networks provide to advertisers is efficiency, not rock bottom pricing. As a results, we have had several well known luxury brands running on our site as part of a larger buy that we never would have had if we went after them on our own. The CPM's are respectable and the value of associating with these luxury brands helps us raise our profile locally.

    If publishers are selective, they can find value in ad network partners. Instead of disregarding all ad networks, let's band together and only work with the networks who respect our brands and the value of audience and our ad space.

    Best Regards,

  8. David Koretz from Adventive, Inc., April 9, 2009 at 1:55 p.m.

    @Jeff, Those two numbers are directly from the IAB representing averages across the industry for the most recent quarter.

    Obviously, as with any average there are some that are both above and below the number, but the average is correct.

    I love being told that we don't "get it".

    1999 called and wants their phrase back.


  9. David Cooperstein from Figurr, April 9, 2009 at 1:56 p.m.

    Let me take a different approach. How about "not all ad networks are selling at 27 cents" and that data is flawed for publsihers who know how to manage their inventory and their relationships. Well, keep in mind that the 27 cents is based on the Pubmatic report that only deals with inventory sold directly to and between ad networks, not the high value rev share inventory that many ad networks fill with high value campaigns.

    We, as an ad network but more so as a publisher network, represent the publisher and ask the advertisers to pay for quality content. We are in the business of selling the value of content, not the volume of impressions. That is how our publishers benefit, and how our advertisers find the audience they want.

    Managing revenue (as other commenters have stated) is not something that can be done passively and expect high CPMs. Can we get $20. Not all the time. Can we can $5-10 CPMs? Yes, and have been able to even during these budget crunching times. We allow publsihers to focus on what they do best, which is to create great content.

    They have full control of what they take, the price they can take them at, and can see the ads they will run. I won't speak for them directly, but I hope our publishers will chime in on how they view the good and bad of ad networks - us and others.

  10. David Koretz from Adventive, Inc., April 9, 2009 at 2:16 p.m.

    @ Chad, that is EXACTLY what I'm proposing.

    And let's look at those scale issues with more accuracy.

    Large brand advertisers (top 5,000) need to spend in large enough increments to be efficient. They simply can't afford to make small buys. These advertisers are destined to buy on a few large sites that give them reach, solid brand alignment, and credibility.

    There are 5.7 Million businesses in the US alone though that can buy local advertising (for local pubs), industry-specific advertising (for vertical pubs), etc.

    There are millions of salespeople in the US selling a litany of products. That's how most products are bought.

    For some reason people forget that just because you bring the content to the Internet, doesn't mean you eliminate the need for people to sell to people.


  11. Craig Swerdloff from LeadSpend, Inc., April 9, 2009 at 2:20 p.m.

    Let's talk about relevance. As exemplified so beautifully by Media Post, take a look at the ad placement along this post and in the newsletter; relevance in advertising is critical. Successful ad Networks deliver relevant advertising where the publisher themselves cannot.

    Thanks for the good laugh. Although I am sure Burst's marketing department did not think sponsoring this absurdity was a laughing matter.

  12. Marti Funk from Sportgenic (Formerly Active Athlete), April 9, 2009 at 2:33 p.m.

    Quite an interesting argument, but the post does leave out a number of evaluative elements. One of the reasons that publishers work with networks is to help sell large percentages of inventory that isn't being monetized at the $20+ CPM. If you were to take that $20 CPM and amortize it across all the unsold impressions, your average CPMs begin to tell a different story. Go further to assess sales staffing, technology and marketing strategy/execution that networks provide on publisher behalf and the ROI balance begins to flip.

    I appreciate Sarah Romer's (from Washingotnian) comment, and agree that publishers have great relationships with advertisers that networks won't be able to replicate. As a vertical network, we support and encourage such relationships and have created a mutually beneficial balance. Did you know that media buyers receive over 300 media inquiries a week on average? Aggregated, relevant, quality media is much for palatable for many, and helps ensure presence with national brands that can't accommodate every independent partner who believes they have the "perfect solution" and 100,000 uniques. Truth be told, there's a lot more competition coming.

    Keep in mind the ability to leverage advanced targeting, technology, scale and relationships with large advertisers does become cost prohibitive for mid- and long- tail publishers. It's not an either/or, it's a balance. Due diligence, service, technology, transparency and mutually beneficial terms is what publishers should be evaluating when (not if) considering working with networks.

    Not all networks are created equal and many have sketchy business practices that discredit the network model. I get it, but the industry and models are maturing. (ShortTail, Jumpstart, RealGirlsMedia, Sportgenic, etc). It's up to networks and publishers to work together to change that perception and create solutions that work for both.

  13. Lionel Sherel, April 9, 2009 at 2:45 p.m.

    Would you trust the opinion of someone that wears jeans like that?

  14. David Steinberger from Gomper, April 9, 2009 at 2:47 p.m.

    Too much emotion and not enough logic. It boils down to economics 101. Buyers only have limited demand for your supply of ad inventory at the prices you set. If you are not selling all of your inventory, than your product is not providing enough value to warrant the price you charge. Sorry. This wasn’t a problem 20 years ago when there were only a handful of media companies who could control supply. But now there are hundreds of thousands of media companies trying to aggregate consumers and supplying advertising…demand will never keep up. Not when you’re delivering 2 clicks for every 1000 impressions.

    Economically speaking, publishers have 3 choices:
    1) Reduce supply, creating scarcity, raising prices (fewer ads per page industry wide…good luck)
    2) Lower prices, increasing demand
    3) Attempt to perfectly price discriminate, essentially riding the demand curve down, selling the same thing at different prices to different buyers (the ad network model)

    None of these options work for publishers because…without a cartel-like hold on the supply of ad impressions, the media business model is REALLY hard to pull off. I’m sorry…I know this won’t be a popular view here but please tell me where I’m wrong. The future of the ad economy belongs to consumers…something I explain on my blog

  15. David Koretz from Adventive, Inc., April 9, 2009 at 2:47 p.m.


    You have never been to an 80's party?!?

    Live a little.


  16. Martin Edic from WTSsocial, April 9, 2009 at 2:54 p.m.

    Ha ha DK, great linkbait!

    And how about when you add in banner blindness, the networks don't benefit their advertisers either. This model never worked well and is never going to work well for either publisher or advertiser. We need an option that does something like dynamically creating relevant inserts whose content adds value to the publication's readers- semantically relevant product content. I'll let you know when I figure that one out...actually I have an interesting idea.

  17. Lionel Sherel, April 9, 2009 at 2:55 p.m.

    So what concert was this, Rick Astley?

  18. David Cooperstein from Figurr, April 9, 2009 at 2:57 p.m.


    Refresh your page - we were not pleased.


  19. David Honig, April 9, 2009 at 3:05 p.m.

    Dave, perhaps you should do some homework first before publishing .This article is very much web 1.0. and perhaps should of been published April 9th 1999 when technology was not as strong and your argument may of held more water.

    1st - Publishers need the courage to take the short-term revenue hit and turn off ad networks? It is the only chance you have to keep the cancer from spreading?
    ( Yeah. this will happen. Fact here is most publishers have an incredibly hard time getting into the deals that matter because their is so many hours in a day and media buyers just dont have the bandwidth. This is the truth, like it or not. )

    brands are looking more horizontally and not within the walled gardens of a publisher. Buying publisher direct gives you insights against that publisher, they cant tell you much more than what they index high against. Again, web 1.0

    Agency holding companies are developing their own networks where they can use their real time data insights and dynamically use them on where that ad should be placed.

    And finally, agree or not. Content is not as important as it once was. People are. Content is a good point of entry but brands need a scalable way to amplify and some ad networks that leverage good technology is a good bet for them. I am not even going to get into Social Data and leveraging ad networks.

    i would imagine you would recommend all publishers to sever relationships with Google. ( that may of been a better argument )

    You really need a better, well thought out argument that speaks horizontally leveraging new innovations to bring the budgets the industry expects from brands.

    Well done at getting my attention!


  20. Frank Addante from the Rubicon Project, April 9, 2009 at 5 p.m.

    The fact that publishers let some ad networks take advantage of them is not the ad networks' fault, it is the publishers'. Publishers don't understand how to manage channels -- and they let their channel resellers (ad networks) sell without boundaries. The publishers haven't been trained on how to properly manage channels like traditional businesses (software, hardware, tech, etc.)

    Some networks even work -against- the interests of the publishers when they arbitrage. They buy low, sell high. They try to buy as low as possible. This trend started because publishers started forcing ad networks to guarantee inventory. Ad networks eventually got smart and said if they are going to take on all the risk, than they shouldn't share the upside with the publisher. Hence, arbitrage was born.

    Further, not all inventory is created equal. Publishers sell what they can, the rest they don't calculate in their rCPM. Some of that inventory many be international as an example - they may getting $0 for it. Ad networks selling it for more is not "lower" than direct. The value of that international is lower to begin with, but they are able to monetize (as a channel) where publishers cannot. We could say the same for the home page vs. inner pages, buttons vs. leaderboards, above/below the fold, etc. Ad networks are judged by total inventory sent divided by total revenue generated for it. If publishers calculated overall CPM the same way (including unsold) than their rCPM would be MUCH lower. It is not apples to apples.

    Lastly, publishers also usually send the wrong inventory to the wrong networks - resulting in an enormous amount of media waste.

    My company, the Rubicon Project, exists because this has become very complex (and fragmented) and most people do not understand how to put proper channel management programs in place to fully maximize the benefits (and therefore, revenue) from leveraging channels (ad networks).

    Let's fix the publisher problem, let's realign ad networks as true resellers (partners that are fully aligned with the publishers) and the whole ecosystem will thrive.

  21. Mark McLaughlin, April 9, 2009 at 5:16 p.m.

    When a senior executive at a mega-media agency who owns the relationship with a Fortune 500 client wants to defend the status quo (TV), he looks for message threads like this to show the client that our industry still has no idea what it wants to be when it grows up.

  22. Steve Levy from Hype Circle Ad Collective, April 9, 2009 at 5:43 p.m.

    Most definitely agree that the large ad networks are hurting the publisher with now crazily low CPMs. They are also hurting the clients as their dollars are just buying ineffective "shotgun" marketing.

    We specialize in sitting in the middle and crafting campaigns for our clients targeting an audience that is relevant and then engaging that audience.

    Our publishing partners benefit as we are able to justify premium CPMs ($10-$20) and we bring advertising that does not alienate their audience--no mortgage ads on a gaming site. So its a win win for both the publisher and the advertiser.

  23. Gene De Libero from Digital Mindshare LLC, April 9, 2009 at 5:48 p.m.

    The ad network math can be difficult for some, but one thing's certain - with ever increasing pressures to produce results, publishers are looking for new ways to monetize their unsold inventory.

    Sure, they could stop creating an endless supply of inventory and maybe, if they really have their acts together, they could actually hire salespeople (not order takers, but real salespeople) who can deliver the revenue they seek.

    But if publishers choose to use ad network technology, it's like any other tool - in order to make it work and get quality results, you need to understand the tool and have an implementation and use strategy that appreciates the ad network for the innovation it is.

    And like all innovations, ad networks should be handled not with a cookie cutter approach, but with a unique strategic and tactical POV and plan that properly serves the business.


  24. Andy Horvitz from Two Plank Consulting, April 9, 2009 at 6:25 p.m.


    Billions of dollars, hundreds of networks, serving thousands of publishers and I doubt anyone would mistake your "quantitative analysis" for anything other than grandstanding. The numbers are stark. That doesn't make them meaningful.

    The simplistic and myopic view you've offered is perfect for the "idiots" you reference. Furthering your "quantitative analysis", the numbers clearly demonstrate that every publisher ought to hire not one, but 3.3 salespeople, instantly ensuring that they sell 100% of their inventory at $20.17!

    Our publishers aren't idiots. We work with them as partners to deliver value for their inventory; value they can't find on their own. If we didn't, they'd fire us. As in any business, publishers who understand and closely monitor inventory management win with networks.

  25. Robb Lewis from Visa, April 9, 2009 at 6:28 p.m.

    Who is this targeted to? If it's one of the top 50 properties that already capture 91% of all online ad revenue then I can see your argument...why sell the same inventory at 1-2% (or even 10%) of it's price and risk devaluing the inventory across the board. But sites below that cutoff, say the next 1000 sites for arguments sake, have to ask the question if they can afford an in-house person or should they look to a network provider. As others have said in these comments, not all ad nets are the same. IDG tech, Netshelter and federated media are somewhat of a hybrid. They sell an audience direct to advertisers but they do have a few of the top sites they promote. So Advertisers can see some of the premium sites their ads will be shown. At any rate, companies should do the ROI of having an in-house ad sales vs using a rep firm or pure ad net. In many cases they will find an ad network or ad rep firm is a better ROI. So you just can't make a blanket statement that ad nets are bad. Maybe you can get away with it if you specifically want to say bad for the top 50 sites that should be selling all their own inventory. @robblewis

  26. Ramsey Mcgrory from Yahoo!, April 9, 2009 at 6:37 p.m.

    This article reinforces common misperceptions by people not close enough to the market to understand the difference between the rational intermediation where value (human, data, aggregation) is added to the transaction versus the somewhat irrational, short term arbitrage due to the friction(shortage of skills, technology shortcomings, media fragmentation) in the marketplace. The former is what the different market segments (buyer, seller, intermediary - small, medium, large of each) demand based on their core business focus. The latter will slowly go away as profits shrink.

    This article is attention grabbing, but not for the right reasons.

  27. Jeff Hirsch from CPXi, April 9, 2009 at 7:07 p.m.

    I find it interesting that blame is being placed on networks for rates dropping. Anyone observe anything about the massive proliferation of supply? Put that next to an economy that pressures demand, and you have a more realistic answer to pricing issues.

    Networks provide publishers with a viable outsourcing mechanism for monetizing their inventory that there were not able to sell based on their salesforce productivity, their value proposition, their customer base, their vertical, their audience, etc. It is ludicrous to think that and impression, which is no longer monetizable a moment after it exists, should be left underutilized. Networks can call on different clients, sell outside of site verticals, aggregate audiences, optimize against wide swaths of inventory. These are things a site can not always do, meaning that there are valid business reasons for both approaches. A systematic look at yield and appropriate sales channels is not an web advertising issue - it is relevant in most industies.

    I find it interesting that while we are speaking there are dozens of publishers creating their own networks. Perhaps they are telling us that the question is not whether there should be networks selling, but whether or not there should be any other way to sell!


  28. Chris Vriavas from CBS Interactive, April 9, 2009 at 7:19 p.m.

    It would seem to me that if the big boys can only sell 30% of their inventory for $20 CPMs, there is not enough demand for more $20 CPM ads. I don't think the challenge is that they need to increase their sales force; I'm sure they'd hire more people if that would help them get more $20 CPM advertisers. I think the flaw in this logic is that there aren't more advertisers willing to spend that much for ad placements.

    And perhaps the reason there aren't more of these advertisers willing to pay $20 CPMs is because that figure is, well, unrealistic and not cost-effective for most placements on most website and the overwhelming majority of advertisers.

    I'm not sure that has anything to do with direct vs. ad networks; that's a whole debate that I feel has little to do with your main math argument.

  29. David Koretz from Adventive, Inc., April 9, 2009 at 10:36 p.m.

    Ben Wilson,

    We handle email for over 750,000 small businesses and have a great track record with 99.94% up-time.

    That said, we do occasionally make mistakes. If we made one in your case, I would like to make it right.

    I will personally reach out to you tomorrow to understand what the issue was.


  30. David Koretz from Adventive, Inc., April 9, 2009 at 10:41 p.m.


    Thank you for a great and lively (if sometimes personal) debate!

    For all the people that responded from ad networks, you are right that ad networks are not to blame. Publishers CHOOSE to use them. They created their own mess.

    The simple fact is ad networks DO compete against a publishers in-house sales force. That is not a healthy situation.

    That said, the broader problem is NOT ad networks themselves. It is, to the point raised by many of you, the issue of a lack of demand for the other 70% of inventory that doesn't get sold as premium.

    As I have argued before (and will again), we need to innovate as an industry on delivering better ad formats, better ROI and better measurements. Until we do that, this problem will continue.

    Thanks again,


  31. Joe Kyriakoza from Oracle Data Cloud, April 10, 2009 at 12:22 a.m.

    While I agree that publishers have become extremely dependent on network, something that I have not seen covered is the dependency that agencies now have on ad networks. In many cases, digital agencies (or the digital media teams within larger agencies) are not staffed to manage the hundreds of monthly calls and meeting requests they receive from the media publisher community. Ad networks become an outlet to achieve their targeting goals. A couple of key points to add:

    - There's an abundance of individual 'premium' publisher options out there for advertisers in every category and lifestyle. All of those properties want to get their hands on Fortune 500 ad dollars, which means already strapped agencies are bogged down with meeting requests and calls, and ultimately in the RFP process, have to sift through dozens of proposals.

    2. As a result of the above, the media agencies develop a dependency on ad networks in some cases as a mechanism to do the job of media planning - what they have been hired by the marketer to do - for them

    So the reason for the proliferation of ad networks comes back to your first adjective - lazy - this goes for both publishers and agencies.

  32. David Holland from, April 10, 2009 at 5:31 a.m.

    David - congrats on opening a lively debate. As a veteran web publisher of multiple niche sites over 12 years in OZ and UK - my experience of forging multiple direct brand sponsorships has been the most profitable for me and the brands. Sometimes we have done via an Ad Network, sometimes via the brand's agency, sometimes direct. Either way we have never used run of Network, and our site visitors know the difference and reward our efforts with their loyalty. We never use cookies and we never sell email lists. Our brands measure our success by the results we deliver. Call me old fashioned if you like but I think you and I are singing from the same ancient hymnal.
    BTW - will you Twitter? Tweet @daveishere

  33. Rob Rasko, April 10, 2009 at 8:02 a.m.

    There are many publishers that derive value from their ad network relationships and there are some that do not, this is clear today but does not need to be our future.

    Clearly many advertisers are happy with the networks since they are sending more and more dollars their way, either through agencies or directly. Clients are constantly speaking to networks advising them on how to better position themselves to help advertisers and the networks do their best to assist.

    My advice to the writer would be to follow the lead of the advertisers and develop a real relationship with a network and find out how a smart ad network can really help your business, our goals are aligned, our mission is to drive CPMs up and not down, which may be contrary to some public opinion.

    Ad networks are here for the long term and we can blame or we can work together as and industry to improve our services and flourish.

    I wrote an article about what we can do in 2009 as group found here, feel free to comment.

  34. Jason Hinkin from Movoto , April 10, 2009 at 12:18 p.m.

    I want a button!

  35. Joe Fredericks, April 10, 2009 at 1:10 p.m.


    You should run for head of the OPA.


  36. Kevin Klein from Adventive, April 10, 2009 at 2:34 p.m.

    I had the opportunity to attend Ad Age's Digital conference this week, at which two industry leaders had the chance to square off on the subject during a panel titled "Can the Web Support Quality Content?". Very enlightening perspective that demonstrates where at least one ad network executive believes his value is.

    Kevin Klein//President//Adventive, LLC//Also Wearing The Pin

  37. Jonas Halpren from Federated Media, April 10, 2009 at 2:45 p.m.


    Seems like most folks who agree with your article are not from publishers, so it makes sense that they jump on the band wagon. Also, your reply to @chad's comment makes very little sense. Most small publishers cannot afford a sales force, even at $20 cpms -period. That is reality. Just doesn't scale.

    I also agree with others that your math is flawed. In you example a site selling 30% @ $20 is averaging a $6 CPM on all 100% of inventory.

    I am not defending all Ad Nets. There are very good networks and very bad one's. A good publisher will manage these relationships closely.

    The one thing I do agree in your article, is that publishers need to innovate its product offering to mover beyond the banner. We do that here at FM, developing integrated brand campaigns for our advertisers and publisher partners.

  38. Alan Schanzer from Pandora, April 10, 2009 at 3:16 p.m.

    Seems David is enjoying the sport of piling onto ad networks without being in the game. There are many, many valuable relationships between ad networks and publishers and ad networks and marketers. Usually this is the result of hard work and effective partnering. Publishers should hold the value of their inventory and should also have the freedom to choose how and how many ways they want to monetize their content. They should expect good value in return, protection for the valuable audience data they have and good quality brands running on the site of that is what they want.

    Its a big eco system with room for everyone who wants to participate in a legitimate, regulated way. More time should be spent on working together v working against each other.

    I spent 11 years at a global interactive agency and spend 100's of millions of dollars with ad networks/portals. And then I joined one. Is far as I know, I have not been accused of being an idiot (not to my face at least). I learned early on that sitting behind my computer sending out my personal POV's was not the best way to feel the market for what it really is. Get in the game. We'd love to have you.

  39. Andy Horvitz from Two Plank Consulting, April 10, 2009 at 4:51 p.m.

    @ Kevin Klein of Adventive, a Blue Tie company

    I'd love to hear more about Adventive and how stirring the pot like this serves its interests.

  40. Ted Rheingold, April 10, 2009 at 6:30 p.m.

    Great Post. Far over due.

    The massive increase in supply and the long-term decline in banner attention mean that bulk CPMs only have one direction to go in.

    In early 2005 Dogster, Inc (publishers of and was a 3 person company and we realized then that unless we served 100s of millions of pages a month we'd never pay the 3 founder salaries from network revenue, and certainly never throw off the cash that it's popularity indicated was realistic. So we made a strategic move to committing to inside/direct ad sales and would consider network inventory fills a zero-growth revenue stream.

    For our business this has made all the difference and I strongly urge anyone seeking to grow their business out of the house if they do not see a path to serving 100s of millions of pages a month.

    Advertising agencies are out there looking for new properties that will help them reach customers via novel and engaging internet experiences. The sooner you commit to direct sales the sooner you will never worry about bulk CPM rates again.

    Don't wait for the economy to recover to build strong relationships with the agencies. Do anything you can for them now and in a year they'll call YOU directly.

  41. Paula Lynn from Who Else Unlimited, April 10, 2009 at 6:34 p.m.

    You are all so funny sometimes and so struck with yourselves. You do forget just because something is published, doesn't mean it has value. And publishers hiring their own sales staff. HA!. have you ever run the numbers of what it costs to pay a salesperson a decent living? How much they would have to sell? CPM's would have to run in the hundreds of dollars. If you doubt it, go try it yourself for the next year.

  42. Marissa Louie, April 10, 2009 at 7:36 p.m.

    This article makes a flawed assumption: that publishers of all sizes have access to the direct advertising model, and that they should therefore skip the ad networks.

    Smaller and mid-size publishers do not have access to the right ad networks. That's where AD-Village comes into play: we provide publishers with access to ad networks.

    Publishers of all sizes have unsold inventory, and the ad network is the tried and true standby for fulfilling unsold inventory.

    We also champion the ad networks: we let them select which publishers are recommended their ads. A key solution to the overdistribution problem with ads is providing full transparency into which publishers the ad networks are distributing to. We have always provided full disclosure.

    One more problem we solve with regards to your sentiments on ads "wrecking" the publisher's website branding: our publishers have 100% control over every ad they place on their websites. They are held responsible for every ad they place.

    Some of the previous commenters on this post are working with AD-Village. We help solve the ad network puzzle for publishers by recommending (and not forcing) which ads to place. We do this with full knowledge of a publisher's traffic level and content.

    I'd like to meet with you to hear out more of your thoughts.

    Marissa Louie, CEO of AD-Village
    510 375 1941

  43. Jonas Halpren from Federated Media, April 10, 2009 at 8 p.m.

    @ted Rhiengold

    Nice comment, so how come I am seeing ad from networks on Dogster? First view no less. What am I missing, is not a network.

    So you don't like ad nets but you use them. Am I getting that right?

    Everyone agrees that you can earn higher CPMs by selling direct. It just doesn't make sense for many small pubs to hire sales people.

  44. Jonas Halpren from Federated Media, April 10, 2009 at 8:06 p.m.


    Do publishers really have a hard time accessing networks? They may have size/quality criteria, but typically most pubs can join at least 5 or 6 if they are 1/2 decent with some traffic.

  45. Ari Rosenberg from Performance Pricing Holdings, LLC, April 11, 2009 at 2:10 p.m.

    The current definition of what is premium and what is not caters to the argument supporting the use of ad networks by publishers who also have a direct sales team.

    My contention is that all page views online publishers produce, have premium value and publishers need to uphold this value even when the inventory goes unsold.

    Let’s not forget our simplest advantage --- in all other media, ads are priced and paid for based on the opportunity for an ad to be seen (for example, not everyone turns every page in a magazine or sits through an entire pod of ads watching TV, but the price for those ads are based on that occurring) -- only in online advertising, are (CPM) ads paid for based on the guarantee the consumer is sitting closer to the ad than your last gimme.

    Premium inventory in distress is different than non-premium inventory, and how publishers handle this stress is what David claims to be in poor judgment if they use ad networks along side their own sales team, and I agree.

    Treating a page view as premium, even when unsold, maintains and grows the value internal teams see and feel, in the product they are being paid to produce and sell.

    What those on the ad network side of David’s argument don’t consider heavily, and the feeling he contends premium publishers who use ad networks ignore, is the slow yet steady erosion of the value of what they sell by outsourcing sales at lower prices than they charge directly. You don’t need a “quan” analysis to know this is occurring.

    Here is a possible compromise that can make this publisher ad network relationship work to everyone’s advantage and provide some blanket clarity to buyers and clients; A premium publisher engages an Ad network(s) to only sell unsold ad impressions below the fold.

    This topic always stirs things up huh :)

  46. Scott Rafer from Lookery, April 11, 2009 at 2:49 p.m.

    @Marissa It's not a flawed assumption for professional publishers. @tedr 's points are pretty easy to support. Well managed direct sales forces make more money in every business. However, direct sales forces are expensive to build and difficult to manage. Few startups can execute against the opportunity, but it _is_ a superior path.

    Ted touched on it, but what should be explored further are the practical tradeoffs of getting direct sales going. It's the inability/unwillingness to make those tradeoffs that give the ad networks a healthy place in the world. If Koretz chooses to be a link-bait whore by calling that inability/unwillingness "an idiot tax," so be it. It's clearly effective, and I've done worse.

    Publishers who don't understand their audiences die fast. Publishers who don't understand how to sell marketing campaigns need business partners to help them over that hump. Those business partners might be rep firms, networks, optimizers, etc., etc. The existence of potential partners isn't the problem.

    The actual problems are two-fold. First, the automated marketing systems (networks, etc.) impact the audiences in a way that publishers can not control or usually even see. Publishers who have slaved lovingly over creating just the right audience experience have their efforts diluted to one extent or another -- not maliciously -- but in ways we all need to build services to correct. Specifically, the ads are (generally) unknown until the campaigns are already running and may not match how the publisher manages their audiences. Plus, certain low-quality campaigns can make the site look bad, cheapening the prices that the network/site can charge later. No network always sells the super-premium bestest stuff. There are intermittent attempts to start purely transparent networks (the old IndustryBrains and now Burst and InterClick), but so far that business plan has not provided the differentiation necessary to break out in a huge way. The industry's move towards targeting ads better will help, plus someone someday will find an ad transparency mechanism that scales non-linearly.

    Second, publishers and their network partners often seek to put ads on too much of the inventory -- 100% coverage is counter-productive. The optimizers (admeld, pubmatic, rubicon) are helping correct for this by encouraging publishers to include house ads in their portfolios, but a lot more work needs to be done in that world. Any page unlikely to make money for the publisher with a network ad should be a house ad instead.

    As Ted alluded to, getting direct sales up and running was a long and painful process. Having watched most of it as an advisor to Dogster, it only worked because all three people running the company have a keen sense for both audience needs and commercial realities. That level of talent is not broadly available and must generally be replaced by ad networks or financing dollars when absent.

  47. John Ramey, April 11, 2009 at 6:19 p.m.

    Many of you are taking one extreme or the other. There is a middle ground here for direct sales without building an in-house sales staff.

    Remnant networks |-----X-----| Sales staff

    In the middle there are new tools that bring the benefits of direct sales, personal relationships, self service buying, etc without the overhead or expense of building a large sales dept.

    As the frictions of conducting direct sales get easier, the "floor" of buyers and sellers who conduct direct sales are dropping. While the big buyers are still going to purchase from the big sellers, I think you'll see a new middle class of direct sales.

  48. Russell Glass from Bizo, Inc., April 11, 2009 at 10:40 p.m.

    As mentioned a few times here, all ad networks are not created equal. Bizo's publishers earn an AVERAGE of $2.75 net to them. Using your math, the publisher who asked us to fill 70% of their inventory would be earning around 25% of their revenues from us. With zero cost of sales or marketing to support that revenue, that's probably approaching 50% of their margins. Publishers are actually a lot smarter than you give them credit for, and know above all how to manage margins.

    Not all of them can make this claim, but the fact is that there are many high quality ad networks out there such as Collective Media, Interclick, Turn and others that pay well for good inventory.

  49. David Steinberger from Gomper, April 14, 2009 at 4:58 p.m.

    15 years ago this article would have read "the internet is for idiots...why publishers don't get that they're cannibalizing their print business"


  50. Michael Tennant from Vice, April 15, 2009 at 7:05 p.m.

    Simply lumping all ad networks into a bucket of crappy ad shilling, cpm killing, monsters, is no longer completely accurate. In fact, there are Ad Networks like the one I run (AdVICE Network), that would argue similar points about competitors like Google AdSence and CPX.

    As GM of AdVICE i've taken the care to hand selecting publishers that complement the Publishing offerings my company already had ( and We treat all of our partners content as premium content and never sell a display ad below $5 or a video ad below $20. Advertisers realize that by working with AdVICE and the partners we've hand selected - that they are getting what the pay for - unprecedented access to the segment of the longtail they are targeting.

    We reached an threshold where our sales infastructurer allowed us to monetize traffic beyond our own - and work with our partner publishers as though they were owned properties. We offer infrastructure, direct sales, creative production, accounting, credit checking and consulting. It's easy to say - hirer your own direct sales, but there is a lot of overhead that goes in to direct sales. And even if you do go that rout, who's to say who you hirer will be good. Or that you will have the time necessary to build demand while paying those added expenses. If the recommendation to capital strapped publishers trying to stay afloat in a down economy is spend indiscriminately for the long term return. Um...i have to question your credibility.

    The fact is that the Ad Network Industry has proliferated. There is enough competition that Publishers can seek better terms and attain relationships that allow them to approve potential advertiser before proposals have been submitted. Vertical Ad networks exist that act like direct sales teams. Email me and i'll tell you all about one.

  51. Carol Williams from Media Dynamics Inc, May 7, 2009 at 8:40 a.m.

    It seems to me that ad networks are a great way for Internet publishers to gain incremental advertising revenues, particularly from "non-endemic" marketers. It all boils down to sales planning and ad inventory control. Nobody is forcing a publisher to accept a $.27 CPM from an ad network for ads that normally go for $12 CPMs. Remember that an ad network is, in effect, packaging its components into a single deal. If the average CPM to the advertiser is $.27, that doesn't mean that some publishers in the network's package cant get $6 CPMs while others earn $.10. In the former case, unloading a block of unsold inventory at half your normal CPM may not be such a terrible idea.

  52. Scott Alperin from Betanews, Inc., May 22, 2009 at 1:51 p.m.

    Congrats on generating a lot of reaction. There is an advantage to writing something that is utterly wrong.

    To effectively monetize a mid-sized or larger site, you need a mix of highly targeted super-premium ads, premium ROS or run-of-channel ads sold at good rates, and broadly targeted low-value ads to fill in the rest.

    If you are selling everything at your top rates (ha!), then you probably need to raise those rates.

    There are only a handful of publishers who are big enough to have a salesforce that can cover all of their bases (or come close to it). That means having sales teams aimed at all kinds of different market segments AND being in a premium segment where there is high demand for the audience.

    Most sites need to join with others to have the critical mass to sustain a sales force. And most who do have them are still to small on their own to cover anything but their most core advertisers. The rest have to be covered by ad networks.

  53. Scott Alperin from Betanews, Inc., May 22, 2009 at 2:10 p.m.


    From the Ad Age story about the study:

    "The four most aggressive publishers used ad networks for an average of 21.2% of their inventory and commanded on average a CPM of $10.36 for the inventory they sold directly. The three others used networks for an average of 7.6% of their inventory and commanded on average a CPM of $20.17.

    Of course, there's a definite danger in reading too much into this data. The study only looked at seven publishers -- not enough to make wide-ranging extrapolations. And that has some IAB members questioning the decision to put the information out there. "

    The truth is, you'd need to look at publishers with almost exactly the same content and audiences and equal quality sales forces, but with different ways of allocating inventory, to get decent data for this.

    In addition, publishers lie about rates, even to themselves. You need to look at eCPM per Page View to actually see the real rates as well as inventory utilization. The Ad Network results were probably for all inventory filled, while the in-house numbers almost certainly left out bonuses/value-added.

  54. David Koretz from Adventive, Inc., May 28, 2009 at 5:15 p.m.

    @Scott.... your conclusions are dubious at best and dangerous at worst.

    The idea that publishers should just assume they have to sell run of network and "low-value ads" is really dumb.

    Google only displays ads when the ads have a shot at being useful, and publishers should consider doing the same.

    Filling the screen with crap, when you KNOW it's crap is only going to devalue your product.

    Further, this unsophisticated notion that publishers are "too small for a sales force" defies all logic.

    Publishers are like most other businesses. They need salespeople to drive demand, and message the differentiation between their product and others.

  55. R.J. Lewis from e-Healthcare Solutions, LLC, May 29, 2009 at 3:04 p.m.

    Let's be careful not to group "Ad Networks" together with "Rep Firms". I think there is even a distinction to be made between Ad Networks that are Pure-Plays (only represent the sites of other publishers) and Hybrids (also have their own owned & operated sites).

    Rep Firms, act much as a publisher would themselves - promoting and selling the publishers brand and maintaining price integrity.

    Furthermore, "pure plays" are different from hybrids, in that they have more of the publisher's best interest a heart. Hybrids have a inherent conflict of interest in their model, as it behooves them economically to feed their own properties first, saving only the scraps for their partners. I've been shouting this for years, and only now (in a down economy) is it become more apparent. Even the likes of Google (the mother of all hybrids) is showing this weakness.

    From a recent IBD article by Peter Barlas on the AOL spin off:

    "Google might be less concerned about AOL for another reason. These days, Google generates a greater percentage of its revenue from its own Web sites than from its partner Web properties.

    In the first quarter, Google said its Web sites produced revenue of $3.70 billion, up 9% from the year-earlier quarter. But revenue from its partner sites fell 3% to $1.64 billion."

    Image that. A Hybrid network feeding its own sites first at the expense of "partners".

  56. Jim Burnette from, May 31, 2009 at 8:01 p.m.

    After re-reading most of the feedback on this very touchy subject, I think we are all missing one very significant metric: the Top 50 Publishers control 91% of the online dollars in the market. That obviously leaves seeds off the roll for the rest of the VAST online marketplace. Check out many of the publishers in the 51-200 range and see if they have been selling premium inventory & sponsorships. Not many at all. I sold for 2 ad networks and communication with your Publishers is paramount. I sold many sponsorships and premium inventory at CPMs in the $9.00-$15.00 range. There is a difference in selling REMNANT vs. PREMIUM inventory. The LARGE publishers use ad networks to sell remnant inventory. The Mid-Small publishers use ad networks to sell their Premium inventory. You must make that distinction.

    Then you must look at the agencies. They are understaffed, underpaid and they have minimal time to meet with salespeople. The bottom line for the Holding companies is way more important than the process of purchasing their clients online media plans. There are literally hundreds of emails & phone calls that buyers/planners receive each month. I would never call the agencies LAZY, but purchasing the Top 50 sites is a lot easier than sifting through the long tail.

    So in the end, we work hard for the 9% that is still available to the rest of the long tail. Ad Networks will always be a part of the long tail, like it or not. If the Ad Networks can provide desirable content, full transparency and scale for a client's target demo, then let them fight it out for the seeds that are left in the market.

    Jim Burnette:

Next story loading loading..