Cognitive Biases — A Product Manager’s Achilles’ Heel
We humans believe that we are perfectly rational beings yet end up making a lot of irrational decisions in our daily life. How many of you plan to go to the gym but end up watching Netflix in the comforts of your bedroom?
In the 1970s Daniel Kahneman and Amos Traversky brought a paradigm shift in the way behavioural economics was looked at. Their work showed that people’s decisions are sometimes flawed because they constantly deviate from rationality. In his book Thinking Fast and Slow, Kahneman attributes this irrationality to the two modes of thinking “System 1” and “System 2”.
System 1 : This is the automatic, fast mode of thinking that is mostly intuitive and comes from your subconscious. This system kicks in when you have an existing pattern in your brain that directs you what to do. For example changing gears in your car when you have been driving for many years or typing fast without looking at your keyboard (muscle memory).
System 2: This is the effortful, slow mode of thinking that is driven by our consciousness and is controlled in nature. For example — making retirement planning or buying your engagement ring.
We think that we are using System 2 all the time but in reality we engage System 1 most of the time because it saves a lot of energy compared to System 2. System 1 looks out for existing patterns, information, associations available in our brain and creates a plausible story which might be influenced by our biases. When system 1 is unable to find an answer it calls for help through System 2.
As a product manager it’s crucial that we are aware of these biases. It is impossible to completely get rid of the biases as many of them are hardwired in our brain and driven by our subconscious mind but we can try and reduce their impact on our daily decision making. There are more than 100 cognitive biases but I would like to highlight some of them. In Part-1 of this series I am going to analyse 3 of the most important ones.
Confirmation Bias
This is the human tendency to favour information that reinforces our existing beliefs. Any opposing argument or information that contradicts one’s belief is considered as wrong and eventually dismissed. One recent example is that of the 2016 US presidential election where in Fox News kept feeding news that perpetuated specific stereotypes and political stance to viewers who partake in similar ideology. Coming back to the world of Product Management, we as product managers tend to believe that being experts in a domain makes us infallible and hence, we try to prove that our hypothesis is correct instead of testing it. I was part of a product team where an A/B test was stopped even before it reached statistical significance because the Product Manager’s preferred design had high conversions than that of the UX designer. Another word for Confirmation bias can be cherry picking. When a new feature or product is launched, Product teams look at VOC comments or feedbacks that confirms their initial hypothesis upon which the product was built. In this case even the metrics dashboard becomes the Product Manager’s enemy because they constantly refer to those metrics that shows the product’s success. Consider this example that the Product team has just launched a product and they see an increase in sales volume while conversion remains low. The product team concludes that the new product is a success. However couple of months down the line they see sales volume going back to the prior numbers. It could be due to seasonality but the point is that the Product Manager was looking at incomplete or wrong metrics to substantiate their biased beliefs. While usability testing is one of the ways to reduce confirmation bias but it has its own share of problems. Quite often I have seen users being guided through the session with questions that we want them to answer favourably. For example we ask users whether they like A or they like B? We are forcing users to make a choice without even knowing if they want it or not. It is like asking someone if he or she likes Americano or Espresso when the user might not be a coffee drinker at all.
Ikea Effect (When labor leads to love)
In 2011 Michael I. Norton, Daniel Mochon and Dan Ariely through a series of experiments found that people tend to place higher value on products that they have created. In those experiments customers were asked to assemble IKEA furniture, build lego sets and create origami pieces. It showed that labour lead to increased valuation of a product when the outcome was a success. This cognitive bias came to be known as the “IKEA Effect”. As a product manager when you give credit to a team or enable them to be more involved then it subsequently creates more engagement and leads to a better outcome. However, there is also a negative effect of this cognitive bias. People start to justify their efforts even when the outcome is undesirable. People believe that they have invested considerable amount of effort in building the product and hence the justification is rational. Product managers find it difficult to accept that their product does not meet the business requirements. Effort justification makes it difficult to be agile and change strategy when the product doesn’t work. Ikea effect is part of a larger bias known as the Endowment effect, which is the tendency of humans to overvalue things when we already own them. Many of us might have used products developed within the organization that are obsolete and do not serve the end goal. Yet the Organization keeps putting money and effort to help build new features and enhance the product. People fall prey to endowment effect due to the theory of Loss Aversion, mentioned later in the article. Ikea effect and Endowment effect are biases, which builds resistance in teams too. Teams love the way they code or test or follow processes and it becomes difficult to convince them to change something that they have been doing or owning for a long time. While it is difficult to emotionally disassociate yourself from the product or the process, we can look at the intrinsic value of the product and not just the effort behind that. Also, motivation to improve or excel helps in reducing the resistance brought in by the above biases.
Sunk Cost Fallacy
How many times have we seen people stay in a bad relationship when they know that its not going to work? How many times have we tried to leave a movie in between because it was senseless or boring? The irrational human tendency to factor in costs or investments in the form of money, effort, time or emotions, that cannot be recovered, to make future decisions is known as sunk cost fallacy. This bias comes into play when we must decide whether to continue investing more energy into an existing activity/project or abandon it to look for opportunities with higher value. Let’s look at some of the most popular examples from the real world. Concorde, the first supersonic airliner was built by a group of British and French companies funded by their respective governments. It cost £ 1.134 billion to develop the 42 year program. Despite knowing that a staggering amount of taxpayer’s money have been wasted in the program, both the Governments kept backing the project to save their prestige. That’s the reason sunk cost fallacy is also know as the “Concorde Effect”. In the world of smartphones no one can forget about Nokia’s deep investments in Symbian when iOS started ruling the market. Nokia was trying to compete with iOS but with a strategy that took it of business.
In product management this fallacy makes the product manager continue building and putting resources into something that would reap no benefits in the future. Larger the sunk cost, more is the inclination to continue investing on the product or project. Researchers sometimes describe this as “throwing good money after bad”. In a way it’s the fear of admitting to a mistake and the guilt of being wasteful that makes Product Managers fall prey to this bias. I had been part of a product team that had a great vision for a banking product and believed that the product could be delivered to the market in 7–8 months. It continued for years because a lot of money was invested, huge commitments were made and none of the product people wanted to let go their passion for their vision. We might think that Agile framework would help us avoid this bias and its true. However, in reality we often see that in medium and large organizations we continue having sprints for years, putting money into a losing cause while believing in the idea that” We are doing Agile” and “Everything is going as per plan”. Though we work in iterations, there is always the triple constraint of: Scope, Schedule and Cost that Product Managers are worried about. One of the root causes is obviously the wrong understanding of Agile but sunk cost bias does play a big role in many of the failure stories.
According to Daniel Kahneman and Amos Tversky “Loss Aversion” is the reason why people get lured into the sunk cost fallacy. Loss Aversion is the human preference to avoid losses compared to making gains. The pain of losing is almost 1.5x to 2.5x times stronger than the pleasure of gaining.
So product people tend to take more risks to avoid failure by continuing to invest in a product whose future ROI is low. They believe that if they keep the product alive and work hard to build better features then they can generate better value in the future.
Ways to combat cognitive bias
a. Healthy Conflicts — Opposing viewpoint should be welcomed in a team discussion and based on the merit of the hypothesis it should be tested or put aside. Someone playing devil’s advocate is only helping make the business case stronger.
b. 5 Why’s — Bias exists because we do not ask the right questions about our beliefs and hypothesis. Five Whys helps drill down any wrong assumptions that we might have taken
c. Conflicting data — Look at the metrics and data that prove your intent as wrong. This will help make better product decisions before building a product or after a launch. The KPIs set before building a feature is very important to avoid confirmation bias
d. Keep your ego in check because the team is working towards a common goal and not to prove our superiority over others
e. Use scientific methods such as AB testing and Usability testing to gather feedback and data to drive decisions instead of irrational emotions
f. Be brave to admit to your mistakes and that will help to avoid falling prey to Sun Cost fallacy bias
It was because of the sunk cost fallacy that I had to bear the pain of watching “Transformers: The Last Knight” till the end in the movie theater.
We will look into few more interesting cognitive biases in the next article.
References
· Kahneman, D., & Tversky, A. (1979). “Prospect theory: An analysis of decision under risk”. Econometrica, 47, 263–291
· Daniel Kahneman, Jack L. Knetsch, Richard H. Thaler (1991)Anomalies: The Endowment Effect, Loss Aversion, and Status Quo Bias The Journal of Economic Perspectives, 5(1), pp. 193–206
· https://hbr.org/2017/11/stop-doubling-down-on-your-failing-strategy
This post has been published on www.productschool.com communities.