Mele’s Musings

The Click-Through Rate Must Die

I’ll never forget one of the first conversations I had with a publisher in my early days as an Account Manager.  We called one of the publishers - and to this day I can’t remember who it was - to tell them that we were going to optimize out of their site due to poor performance.  “Sorry,” I said, “but we are just not hitting our goals, and we need to issue you an out.”  “Wait,” the person on the othe end of the phone pleaded, “before you do that, let us do some things to improve performance.”

Feeling generous, I decided to hear their ideas.  “Ok, see, we can improve your click-through rate a ton if you just put some pictures of chicks in bikinis on your ads.”  The silence on my end of the phone must have made them nervous, because the voice on the other end quickly added, “Look, man, if you send us the ads, we can put the girls on them ourselves.  We’ll do it for free.  Seriously.  It always works.”  I politely told them that we weren’t interested, and that increasing click-through rates was not our goal.  Not to mention that girls in bikinis was not exactly pertinent to the advertising we were running.  “Ok, man, suit yourself.  But we’ll seriously do it for free.  I guarantee tons of clicks.”

That was over 10 years ago.  Nothing could have made it clearer to me that the click-through rate was a seriously flawed metric.

Link to image

I find it shocking in 2010 to still be writing about this subject.  In fact, I wrote about the very same thing almost exactly a year ago, and I am flabbergasted that it is still an issue.  But, just last week, eMarketer sent out an article entitled “Is the Click Still King?” and my jaw almost hit the floor.

I guess I’m just not used to it anymore because measuring beyond the click just isn’t that hard.  It’s kind of a basic thing in fact.  The major ad serving systems measure beyond the click, and setting up ads and tags or beacons are pretty standard.

Not that the click is a total waste, mind you.  Clicks are obviously the measure of choice for search.  And, for rich media they can be an indicator of deeper interest, particularly if you measure clicks within the ad unit.

But seriously, who really clicks on ads?  When was the last time you clicked on an ad that wasn’t search?

It can’t be utter laziness that keeps us from measuring beyond the click, can it?  Actually, the eMarketer article points out a few things that I have seen in my time.

First, we have shown in study after study that people do take action after viewing ads without clicking, that conversion rates and awareness increase after ad exposure, and that viewing ads actually lead to more search clicks and more search conversions.  In fact, measuring conversion activities that are post-click only is guaranteed to lead you to make the wrong optimization decisions.  We have seen it time after time.  The challenge?!?  Doing this kind of study is not easy.  You have to have serious analytics jockeys on staff to set up a test correctly and to make sure that the results are measured accurately.  Measuring off of clicks feels “safe,” but we know that you must measure beyond the click to understand the true value of your digital advertising.

The second is that figuring out what to measure can be hard.  Clicks are easy to understand and easy to measure.  Other activities, particularly if they are not based on a purchase or a sign-up are harder to measure and harder still to evaluate.  Again, you need serious analytics and data jocks on staff to be able to connect ad exposures to meaningful value-driving activities, and harder still to connect online activities to offline activities that actually matter - things like total sales (online and offline), customer satisfaction, in-store traffic, etc.  And, for clients, you need to be able to connect different areas of your business together - something that can be very challenging in large organizations.

But, it’s not impossible, and it’s no excuse to still be using click-through rate.  In fact, I would argue that in most cases (excepting search and some rich media) judging by click-through rate may be the exact wrong metric.  If “chicks in bikins” get people to click, then it should be hardly surprising to find out that the things that people click on may be titillating or intriguing in some way, but may have absolutely no bearing on value-driving activities.

So, if you are one of those still using click-through rate as a measure of success, you have to stop.  There’s nothing else to say.

Tweet This Post  Buzz This Post  Delicious This Post  Digg This Post  Reddit This Post  Stumble This Post