Cracking trading cards packs and web analytics

Editor's Notes

• #3: Billy Beane and Peter Brand have revolutionised baseball but this deck is not about Moneyball. I enjoy reading a lot and I am always interested in linking the writer to specific families or schools of thought. And speaking to other fellow digital and web analytics people often reminds me of my youth when we were buying packs of stickers showing the soccer players playing in the World Cup. In the US, you might bought Upper deck MLB, or Topps NFL Stadium Club, Hockey, Basketball cards. So let’s imagine what a card set of the most influential people in web and digital analytics would look like. Some may have died centuries ago, lived in various countries so, expect obscure names in that set but people that every web and digital analyst should know.
• #4: Let’s crack open a pack and fill empty slots in our album! Who are you gonna get? Daniel Kahneman? Robert Cialdini? Ignaz Semmelweis? Dan Ariely? In this session, we will only have a look at a few people and see how their story is surprisingly relevant, and a challenge to deeply held beliefs in our field. I will leave that connection a mystery for now, at least until I have told you the story of these very interesting people.
• #5: Phineas Gage is not holding a baseball bat or a hockey stick but a metal rod. He was working as a foreman in Vermont in the 19th century, clearing rocks with his team for a future railway. The method used at the time was to drill a hole in a rock, pour some gunpowder in it, push it with a metal rod. Unfortunately for Gage, the rod caused a spark and the explosion caused the rod to break through his skull from below his cheekbone and out at the top of his skull. Blood and brain matter was found as far as 30 yards from the spot where the incident took place. He survived the immediate aftermath of the accident, able to walk and speak again. But soon, his exposed brain became infected, and Gage fell into a comatose state from which he recovered, too. 3 months later, he seemed to be normal, but his personality was altered forever. He became vulgar, emotionally shallow. To his relatives, friends and previous colleagues, Gage was no longer Gage. He lost his job as a foreman, worked odd jobs and even as a stagecoach for 7 years in Chile, of all places. Then he started suffering from epilepsy, moved back to San Francisco with his mother and died at age 36, almost 12 years after his accident. He had regained most of his social skills by then. The human brain is made of the limbic system, a very old part of the brain which deals with most automatic functions of the body such as breathing. It is also where emotions are formed. Around this old brain, you find the neocortex, a younger brain dealing with evolved and sophisticated functions. A key part of the neocortex is the prefrontal cortex, just behind the skull bone of our foreheads. This part of the brain regulates emotions and is heavily involved in decision-making. In most people the prefrontal cortex and the limbic system operate together which results in balanced behaviour, not too emotional, not too logical. Gage’s accident caused to lose that connection between both parts of the brain, his limbic system was off the hook, his frontal lobes unable to regulate it.
• #6: Antonio Damasio is a Portuguese-American neuroscientist. As a student, he became fascinated with the story of Phineas Gage. But Gage’s accident was so rare, and it could no longer happen with today’s work practices. Who would be today’s close equivalent to Phineas Gage? Damasio treated a patient, whom he named Elliot, who had to undergo brain surgery to have his frontal lobes removed. The prefrontal lobes are key parts of the prefrontal cortex involved in the regulation of emotions. After the operation, Elliot was able to live almost normally. At some tasks he became even more efficient. But he soon lost his job because he became lost in the details and became unable to deliver work on time. Elliot also stopped feeling any joy, sadness or anger from watching or reading the news. Gage was a lot like that as well. But more interestingly, Elliot was simply unable to make any simple decisions such as whether Wednesday or Friday would suit him better for his next appointment. All options seemed equal to him. Damasio found that the frontal lobes not only played an essential part in decision-making but also that without emotions, that decision-making process becomes extremely difficult. A healthy person acquires over their lives a collection of associations between any sensory input such as images, sounds, tastes and body reactions. These are what Damasio calls “somatic markers”. People build these associations and exhibit body reactions to sensory input such as sweating a little in a stressful situation. The body compares the current proposition with past experiences. If a preponderance of these past experiences were good ones then the person will be likely to make a favourable decision. Some might see their body as merely a vehicle for their brain, but the brain informs the body. The body has a memory and the brain is a servant to the body.
• #7: Pavlov ran a famous experiment involving a dog, sausages and a bell. The dog sees and smells sausages, it begins to salivate. Pavlov started ringing a bell each time he brought sausages. Then simply ringing that bell makes the dog salivate after a while, even without the sausages. When I was a teen, a hurricane tore through the county, something that happens less often than once a century. The hurricane ripped away the whole roof of the house of a high school friend of mine. They survived the ordeal but each time the rain hit the classroom window with a little more wind than usual, she cried. It’s all about something called “priming”. When pressed about why some made a specific decision, they will probably rationalise it, but it was driven by subconscious processes, emotions.
• #8: Antoine Bechara, Steven Anderson, Antonio Damasio and Hanna Damasio worked together at the University of Iowa where they developed a psychological experiment which is called the Iowa Gambling Task. They recruited participants who were told to maximise their gains by picking the top card off of 4 different decks of cards many times. With each card they would either gain or lose money. But the 4 decks had different odds of success. There were 2 good decks that gave rewards more often than penalties and 2 bad decks. After how many cards would the participants know which decks are which? For healthy subjects the number was 40 to 50 cards. But the participants were also connected to a machine very similar to a lie detector. After as few as 10 cards, the study measured a stress reaction when the participants hovered their hand over one of the bad decks. The body knew before their brains did. Not all participants were healthy, some suffered from brain lesions similar to Elliot’s or in other areas of the cortex. They would win some, lose some, but never figure out which were good or the bad decks were. They also never expressed any stress reactions to the bad decks, all 4 decks were equal to them. They were unable to strategise, fail to understand the future consequences of their actions, and played as if at random. Without these somatic markers, their brains were blind. All decision-making starts with body responses to our stimuli, feelings, which are the building blocks for emotions which eventually drive action, and decision-making. This resembles very much the situation where the lowest-paid employees of a company, the ones at the coalface know more about the economic reality of the company than a middle-manager or the C-suites do. It is as if the brain is not as closely connected to reality as we would like to think. This experiment has had its share of critics, but the main takeaway for me is how without emotions, people can’t make decisions.
• #11: Web and digital analytics aspires to be scientific and emotions in the decision process may seem undesirable. Elliot felt no emotions after having his frontal lobes removed, which should have turned him into a hero of rationality. But we have seen that neuroscience is challenging that assumption and suggests that a lack of emotions hinders the decision-making process. The brain procedure that Elliot underwent bears a name that should send a chill down the spine to everybody. It’s called a lobotomy. During the darkest times of Soviet Russia, the regime considered that anybody dissenting against the party line must be mentally ill because socialism is good for the workers and everybody at the bottom of the social hierarchy. The tyranny of the Tsar, his family and the aristocracy are over. Dissenters were often given lobotomies as a “cure”. They definitely stopped caring about their plight.
• #12: Emotions in the decision-making process are good and desirable even. Learn to instil emotions into your analyses, not too much, not too little. Try to find a sweetspot, what Swedes call “lagom”. Many claim that data storytelling aims to achieve just that. I believe that making your stakeholders and clients understand your analysis is a fool’s errand. This will be the topic of a future deck. There is an alternative to eliciting emotions via data story telling, it’s called the Ikea Effect, which also goes hand-in-hand with the Endowment Effect. This will be the subject another deck. Keep your eyes peeled for plenty more trading cards for your collection! Aristotle said something very similar. Persuasion requires logos, i.e. logic, facts and data, but also pathos, i.e. emotional appeal.
• #13: I wished I had the exact quote and reference. Julia de Funes, a French contemporary philosopher, referred to Nietzsche in an interview, where he posited hat behind every decision there is an emotion. I believe that deep down, everybody makes decisions at an emotional level. But we are all distributed between two extremes with people who have no problem admitting that they do just that at one end, and people who feel a lot less secure about that at the other end. They are people who feel they may need to defend their decisions with a rational explanation. They will need facts, and data to support a decision they have made earlier, an emotional decision like everybody else.
• #14: Trying to be on top and staying on the cutting edge of data science is a cause for a many analysts, data scientists and people working in data. But analysts face a last-mile problem where their analyses fail short of getting buy-in from their stakeholders and clients. Trying harder by bringing more facts, more data, more sophisticated analyses is tantamount to doubling down, produces the same result which leads to frustration. Some companies are more data literate than others but I doubt that this approach will have a greater impact elsewhere. Analysts would benefit from gaining people skills, persuasion and influence skills in particular. They might find that a somewhat basic analysis, sparse data and facts balanced with people skills has more impact.