I’ve seen confirmation bias kicking in many times, especially when dealing with black box problems. SEO for instance is a well developed topic and you can find many established practices nevertheless some areas of SEO are still empiric. The fact that Google does not disclose the entire algorithm behind the ranking and changes that algorithm on a regular basis creates grey area of action that from one side makes many SEO Gurus extremely profitable and on the other end takes people into thinking that there is some magic involved.
This establishes the ground for irrational practices based on Confirmation Bias. We tend to believe that a strategy we are developing is the right one and then we tend to believe that the results we get are directly connected to our strategy, if positive, depending on something else, if negative. Then we tend to ignore the negatives and just focus on the positives. False success is reached! Hurray!
The sunk cost fallacy
The sunk cost fallacy is the irrational tendency to focus on something we already lost and cannot be recovered rather then focus on the future opportunities. It’s deeply coded in our brain because in the evolutionary journey it helped our genes to survive the adversities. Unfortunately it easily becomes a burden in a very unpredictable and fast changing environment like the online business.
We are victims of the sunk cost fallacy every time we decide to stick to a loosing strategy just because we got used to it or we already spend large part of our budget into it. When we care more about what we lost in the past rather than on what we can gain in the future we are throwing away opportunities.
The anchoring effect
Rather than making a decision based on pure value for investment, we tend to compare the options and use that comparison as the decisive element. A very good example of this bias is the following:
Dan Ariely conducted an experiment based on a real subscription model by The Economist.
There were three different options:
- web only subscription: 59$
- paper only subscription : 125$
- web + paper subscription: 125$
16% of the people selected option 1, 0% option 2 and 84% option 3.
As soon as option 2 was ditched, and only option 1 and option 3 were available, the discribution changed and the vast majority opted for option 1.
These are not the only cognitive biases we have to pay attention to. Wikipedia has a huge list of biases that affect decision making and a little exploration of that list is a great investment. Knowing that our decisions can be biased and learning how to avoid such effects can dramatically improve the quality of our strategies.
Image by XKCD