Query length is about human behavior. The desire to game the system is fundamental to human behavior. Split A/B testing is about human behavior, about saying, "Forget about our opinions; what do people actually do?"
And our choice of search engine is most definitely about human behavior.
This may seem obvious to you, but it is evidently not obvious to everyone. It's not obvious to the person who commented on my column last week saying there was no "pain" associated with switching search engines. The pain of changing a habit is one of the worst pains there is. Habits can easily trump cash -- otherwise, Microsoft would have had it nailed once it offered Cashback. Habits can cause us to spend more than we should, use outdated legacy systems, or stay in unfulfilling relationships.
If you are a student of human behavior, then you already know that we are far more driven to avoid loss than we are to seek gain. In her 2004 TED talk, cognitive researcher Nancy Etcoff described the "negativity bias" -- effectively, the imbalance between these two drivers. We detect sweet at 1 part per 200, but bitter at 1 part per 2 million. If we were governed only by pleasure-seeking, says Etcoff, we could not survive.
We are eternally preparing for famine and Judgment Day, where our survival will depend on how many cans of food we managed to stockpile. If you want me to give up my can of Google, you can't just be offering me a can of something a little bit better. What if I give up my Google and your Bing is no good? I've just condemned my family to starvation.
If you are a student of human behavior, then you know that we are both rational and predictably irrational. The economics of our actions aren't always dollars-and-cents tradeoffs, but they are almost always benefit-loss tradeoffs -- and the calculation isn't 1:1. Thanks to the negativity bias, the losses are given greater weight. In addition, we tend to downgrade potential benefits because of the risk that they won't eventuate. Sure, on that search Bing's results were more appropriate, but on this one Google's are better. What if I switch and it turns out Google's are better 70% of the time?
The negativity bias is why I believe that Google will be unseated not by a new and better search engine (which will require us to change our habit) but by a new environment that renders the old habit obsolete. The Yellow Pages were replaced by our migration to the Internet, not by the Orange Pages.
If you are a student of human behavior, then you know that we give brands a mental context. Brands are effective because they act as a behavioral shortcut -- push this type of lever, get this type of reward. Go to Starbucks, get a certain type of experience. Go to The Gap, get a different type of experience. Brands do not automatically extend beyond their original context; if they did, Chrome would have more than 2% market share. Or Microsoft would already have won the search war.
I'm a long-time Windows and Office user. I'm a long-time Google user. Both companies have changed our lives, in my opinion for the better. So my purpose here is not to be a hater. It's to remind, reinforce, and reiterate that to understand what drives search, we have to understand human behavior.
Because at the end of the day, search is, and always will be, about people.