Anchoring - SEM Bias Series Part 1

Posted on November 22, 2010

Anchoring is a cognitive bias that describes the common human tendency to rely too heavily, or “anchor,” on one trait or piece of information when making decisions

I took over a new AdWords account. The click through rates were horrendous. I split tested some adverts and improved the click through rates until I was much happier with them

Further split testing did not yield big improvement but I was generally pretty happy with how things were going. Later I handed the account over to a colleague a pretty much forgot about it. 6 months down the line she had doubled the click through rate.

I made two mistakes here, both to do with anchoring:

  1. I was happy with the improvement I made in the CTR because it was so much better than what I saw initially. My brain “anchored” on the initial CTR and everything else was judged against that. This meant that I did not really consider the possibility that the CTR could be doubled again.
  2. The adverts that my colleague came up with were completely different to the ones I thought of. I think I was capable of coming up with that idea, but I had “anchored” on my own ad texts as being a template for what would work. I didn’t try anything that was really different because I thought I had cracked it.

Fixing this is tricky. I came up with the following rule (which I do not follow all the time, unfortunately):

Always test at least two adverts. Every fifth advert you test should be completely off the wall. Try and do something that no one else is doing right now.

The problem with this rule is that, whilst it is easy to come up with small tweaks to adverts, it is hard to come up with creative new ideas on a regular basis. It is also tough to force myself to write new ad copy when I know it is probably not going to make more than a 0.01% difference.


This post is number 1 in a series. Read the 0th introductory post on cognitive bias and search engine marketing and see the complete list of biases