By Jim McCarthy Nov 17, 2010 1 comment

You Can’t Fault ’em For Not Being Optimistic…

Via Thomas, I said this story on e-marketer about the click-through rate on banners. Here’s the key tidbit:


“One of online marketers’ simplest metrics to keep track of, the clickthrough rate, has been in decline for years…however, that decline appears to have stopped.”

Hmmm.  Let’s take a look at the data.  This is the graph that was presented in the article referenced by e-marketer.  Pay particular attention to the red line:

So from the July 2007 quarter to today, the “annual average CTR” went from .12% (dismal) to .09% (dismal’s sub-basement).

The contention made in the article linked from e-marketer is that CTR has “plateaued” at .09% over the last two years, so maybe now everything’s fine and CTR will always and reliably be at this incredibly low number.

I’m not so sure, and let’s look at the red line to see why.  The red line is a quarterly overall CTR, and here are the last eight quarters of that number:

.09%

.07%

.09%

.09%

.09%

.08%

.15%

.08%

Yes,  the average has remained .09, but only on the strength of one mysteriously strong quarter (.15%) earlier this year.    Remove that and the average is well on its way to something below the vaunted “plateau” of .09%.

Two things: get it right , people.  A plateau is when something grows up to a certain point and then stops growing.  Calling this a plateau is like saying a plane that crashes has merely achieved a new cruising altitude of zero.

Second thing, what is that .15% quarter?  There’s got to be some unusual event or explanation for it.  If there’s something in there that could potentially happen again, I’d say consider it in your future thinking.  If not, you can’t.  Perhaps Media Mind, the firm that did the research, will contact me with what they think explains the extraordinary result.  It doesn’t appear to be seasonal, and since it’s a big universe of data, I can’t imagine it’s simply random variation.

So while it’s great to have a sunny outlook on life, as a rule, I don’t think the facts of the situation with banner ads are being accurately depicted here.  Once numbers get small like these, it’s easy to say that .08 isn’t that much different from .09 and therefore it’s stable.  I will say that for the first time in the last twelve months or so I’ve seen and heard of ads clicking through at .05% for the first time in a decade of doing this.  Anecdotally, I’d guess the downward drift continues, but we’re now at such a fine scale, it hardly seems to be moving.

The larger point is this: banners (and not just banners, but most of what Seth would call “interruption marketing”) are an unresponsive medium.  This kind of advertising, generally speaking, no longer works.

Share and Enjoy

    1 Comment

    • John Loken

      Interruption marketing is becoming increasingly ineffective on the so-called Semantic Web, and the eMarketer analysis illustrates that we’re really just rearranging the deck chairs on the Titanic (or rearranging something smaller, like drink coasters or moustache combs – hey it *was* 1912). I’d argue that if there is any clickthrough rate “plateau” to speak of in 2010, it is due to a) Facebook’s codification of “Likes” enabling advertisers to target ads based on these interests, thus increasing CTR, and b) the emerging importance of Facebook as an ad platform. This chart from the Wall Street Journal shows an increase in FB’s marketshare in Q3, possibly correlating with that spike in click through rates: http://online.wsj.com/article/SB10001424052748703665904575600482851430358.html