Welcome to the bloggy home of Noah Brier. I'm the co-founder of Percolate and general internet tinkerer. This site is about media, culture, technology, and randomness. It's been around since 2004 (I'm pretty sure). Feel free to get in touch. Get in touch.

You can subscribe to this site via RSS (the humanity!) or .

Random Thoughts on Online Advertising

It might be selective, but it seems to me like online display advertising is a hot topic at the moment. I’ve got a whole big post in my head about the whole thing, but instead of writing it now I’m just going to throw a few interesting points from other folks your way.

First off, Brian Morrissey writes: “Advertisers, I think, are questioning the entire notion of buying bits of real estate on the periphery of content. It’s just not that enticing – and with good reason. Despite all the studies showing banner ads increase search conversions and do some to lift brand metrics, consumers don’t seem to care.”

I think Brian is right on here. The economy is a good excuse for what’s happening, but it’s not the root cause. Who wants to live on the periphery of content when you can make yourself part of it? Back in November I wrote, “Maybe the answer is that advertisers need more variations on their creative. What I mean is, I think part of the banner blindness problem (and this is all speculation without any data behind it so take it with a grain of salt) is that we’re all trained to recognize when something doesn’t belong and, in the case of the web, to ignore it. Banners tend to be a different color, font and they move all around, add in the fact that they sit along the edges and they’re just too easy to quickly spot and dismiss. But once in awhile someone like Apple comes along and does some fancy custom unit where they pay attention to everything including getting the NYTimes.com typeface right. That kind of stuff must make more of an impact than your run of the mill banner, no matter how cool it might be. Right?”

So that’s one thing agencies can do, but Brian also ends the post with a good point from Harvard Business about publishers: They “need to think more like marketers and, like it or not, mesh advertising with their content.” Precisely.

Okay, onto the next point, this one from Terry Heaton, “There’s no incentive to change. When your life is based on broadcast and print CPMs, the only ad model you see is, well, CPMs.” This is on both sides of the coin (both publisher and agency). Things are still all about scale (even more than ever now that media agencies are increasingly moving away from models based on taking a cut of spend). I’ve said it once and I’ll say it again, the web works best when it’s not used as a scale medium. Sure, it works sometimes (Barbarian Group is responsible for one of the more famous successes), but the more consistent and long term solution is to build great experiences for a very specific group of people, plain and simple (and scalable, actually, just not scalable in the same way a television buy is). [As a side note, scale is a problematic word since it’s thrown around as a hard number (does it scale?) but has very different meanings to everyone. For whatever that’s worth.]

Speaking of scale, there’s a flip side to this situation. In January Fred Wilson wrote about super cheap CPMs. His prediction for 2009 was that “display advertising will get so cheap and the tools to target it will get so good that it will be shown that it can outperform search.” It’s an interesting one. When the media space gets cheap enough, it doesn’t really matter what the click-through rate or anything else is. That’s why spam works after all (some incredibly small fraction of those spammed has to actually act on it for the ROI to work out). Of course this kind of flies in the face of everything I said before, but hey, who’s counting? The issue of course with super low CPMs is that it’s hard to make a lot of money off them if you’re a publisher (well that and really crappy banner ads hurt your brand as a publisher). But anyway, it’s a different way of thinking about online advertising. (And honestly, the biggest thing that needs to happen is we need to stop thinking of it as advertising altogether. It’s so different than print and television. But again, I am not here to talk about semantics.)

And yes, I understand that there is more to online display than click-throughs. But again, the branding stuff requires people noticing your creative, which gets me back to my earlier point about doing stuff that looks more like it belongs on the page.

I think that’s pretty much it for now. Sorry about the random nature of this post, just had a bunch of different stuff I wanted to get out there.

For good measure here are two more things to read on the topic: One from Adweek.com and the other my friend Clay explainig the similarities between birth control and online advertising.

February 5, 2009


  • Alan Wolk says:

    Noah- as I noted on @bmorrissey, I think the essential issue with banner blindness is how we’re looking at them. Banners are the roadside billboards of the internet. We see them when we’re on our way someplace else or doing something more important. Using you and I as a decidedly non-scientific sampling, how many times have you been on your way to dinner, seen a billboard for something and turned the car around to head to the mall instead?

    I figured as much.

    So why do we think that people will stop checking the sports scores, the weather, the flight schedule to Miami, etc. to click on a banner?

    If we start thinking about them the same way we think about billboards and bus shelter posters, we may find they’re not as useless as we’ve assumed. I often learn things from banners- about products and sales and whatnot– that I’ll file away for future reference.

    From a creative POV that would involve giving me all the relevant information in the banner rather than trying to lure me into clicking, which should actually make the creative better/more entertaining.

    And acknowledging that “yes, you might want to fly Delta, but we realize you have no desire (or time) to navigate away from ESPN right now” would go a long way to making the units more likable.

  • amber says:

    wow – i’m really glad to see that someone agrees with me (comment above) on the fact that people maybe look at display advertising in the wrong way. Display ads are essentially the online equivalent of OOH or print ads – sure, there are much better ways to talk to people, but these vehicles perform a lowest-common denominator function: blips on consumers’ mental radar. It’s obviously important to build relationships with consumers, but the thing is that probably only a small percentage of a brand’s consumers actually want to have that relationship. Does that mean you should write off everyone else?

    The bit above where brian morrissey notes that “Despite all the studies showing banner ads increase search conversions and do some to lift brand metrics, consumers don’t seem to care.” is interesting, because I don’t actively care about print or OOH ads either (or TV for that matter), but that doesn’t stop the brand whose ads I’d seen all over the subway from popping into my head when I’m at the shelf.

    And to the point about publishers providing more ways for brands to support and be integrated with their content – I just lamented the death of Domino Magazine on my blog, and if they had just paid attention to their website and found better ways to monetize it (beyond two standard units and “subscribe now!” pop-unders), they’d probably still have a business, because people LOVED them. Their fans knew they could go online and get their content for free , from design bloggers who posted scans, and from the Domino site, and Domino should have recognized this, and found ways for brands to subsidize their online business model.

    sorry – longest comment ever, but I’ve been thinking about this a lot lately too. The post that went along with that birth control venn diagram thing made sense, but I think people who are unfamiliar with how paid online media works took it as a standalone thing and ran with it, which was weird.

  • Adam Singer says:

    Advertising *IS* content (and content is advertising).

    Read this whole series if you are in advertising:

    Mike’s not even in advertising and he really gets it

  • Matt Mantey says:

    I’ve been reading the same posts and it’s been eating at me. I posted a bit about it this morning.


    12 years in, and it persists that clicks are tied to banner effectiveness. We’ve moved so far beyond that, but yet get dragged back into the same discussion even with people that understand online advertising.

    With cookie windows you can see so much further than the unrealistic diversionary click that Alan correctly point out.

    But, it’s hard damn work to wade through that data every month, day, hour to make it really useful.

    I don’t see how banners are anything like billboards or print ads. Banners are embedded in the medium where direct and indirect actions can take place, be tracked and attributed. Results and actions are not just anecdotal and recorded in diaries.

  • stephanie gerson says:

    i Love how it’s a revelation that advertising must increasingly be useful/interesting. soon we’re gonna be saying the actual products/services advertised must be useful/interesting! wowsers.

  • Michelle says:

    Does anyone know the industry standard for page views or cost per impression in the U.S.??

    I can’t find a number anywhere!

    Email me if you know.


  • Matt says:

    @Noah: it’s awesome to hear “great experiences for a specific group.” This suggests that there is some type of targeting methodology. The problem, however, is that no one, including Google, has solved the relevancy problem with content advertising. I’m curious to see how the balance of scale and relevancy plays out, especially from a CPA perspective.

    @Alan: interesting perspective on banners. Should banners change immediate behavior or provide bits of knowledge for future actions? This suggests down-playing a banner’s call-to-action and creating a functional “widget,” with which many companies are experimenting.

  • michael maurillo says:

    Since your post drove me to bmorrissey’s post and i ended up commenting there, i’m restating here…

    I spent a large part of my career on the agency side selling banner creative to clients. Really, we’re all to blame for this. The agencies that sold (sell) banner strategies to their clients b/c they know they will buy them and it’s an easy deliverable to mark-up AND an easy deliverable to template. The brand managers that are afraid of any other type of digital tactic b/c it is the closest thing to what they know best – tv & print ads (even though we know that’s not true). And the properties that sell the ad space b/c they haven’t come up with another way to monetize their business.

    I do want to mention Pandora.com, though, as a good example of keeping it simple. I continue to be impressed with their ad model. Not just how well they integrate something as basic as a display ad into their experience, but how they have found relevant and contextual ways to integrate their advertisers into the experience without getting in the way. To be honest, what they are doing is not earth shattering. They have just made a commitment (from what i can tell) to keep it simple. Both in the creative integration and the actual quantity of ads (i’ve never seen more than 1 advertiser display ad at a time). As a result, i always find myself actively aware of the advertisers on their site. And to be honest, when the page reloads with a new ad (if you use pandora, you know the experience i’m referring to), i actually pay attention to see what the next ad is going to be. It’s very odd. I’m almost can’t explain it.

    i know Hulu is known for “obsessing over every pixel,” but i really think that pandora is a better example of that mantra.

  • Noah Brier says:

    @Matt: Thanks for the comment matt. i generally agree, but am not sure about this part “Banners are embedded in the medium where direct and indirect actions can take place, be tracked and attributed. Results and actions are not just anecdotal and recorded in diaries.” How can you measure indirect actions? (Anything that doesn’t involve actually interacting with the banner.)

  • craig says:

    Thanks, Noah. This was a great post and I agree it’s a timely one in our industry.

    The Harvard Biz article is spot on with the mentality publishes must adopt and practice rigorously to keep their footing and not loose readership (or advertisers) interests in what they consume.

  • Matt Mantey says:

    Noah – hope to clarify my earlier comments and your question. Direct, referring to the action (s) that take place immediately after the banner is displayed – a click, a Google search, direct url entry, etc. Indirect, referring to all of the possible actions a banner viewer (whether they knowingly saw, saw and read, saw read and considered the ad) could do within the cookie window. We’ve run campaigns where the view-based data has informed big campaign messaging adjustments. Like many others, we can’t always afford to dedicate the resources to chew that data.

  • Ruthanne Games says:

    Once you start making money online it is the best feeling in the world!

  • Leave a Comment

    Your email address will not be published. Don't sweat it.