“The greatest enemy of knowledge is not ignorance; it is the illusion of knowledge” – Stephen Hawking
It is the quintessential Silicon Valley story. Elizabeth Holmes, a 19 year old Presidents Scholar at Stanford drops out of college to launch her startup. Holmes’ big idea was to perform multiple tests on a single drop of blood and deliver the results wirelessly to doctors. Her story was straight out of the Silicon Valley script – blood testing is a $75B market and as such was ripe for disruption: the technology was hailed as the iPod of healthcare and Holmes was touted as the next Steve Jobs. By the end of 2014 Theranos had received $700M in funding and had a valuation of $9B and Holmes with a net worth of $4.5B was the youngest self-made female billionaire in the world. The funding came from high profile investors: Rupert Murdoch, Carlos Slim, Betsy de Vos and the Walton and Cox families. The Theranos board consisted of such luminaries as Dick Kovacevich (Former CEO of Wells fargo), Larry Ellisson (Founder of Oracle), Jim Mattis (Secretary of Defense) and George Schultz (former Secretary of State).
However in 2015, in a sensational exposure by the Wall Street Journal, the test results were found to have been falsified. After a subsequent investigation by the Securities and Exchange Commission, Holmes was found guilty of an elaborate fraud and was charged with making misleading statements about the company’s technology, business and financial performance. Holmes and Theranos are currently on the verge of bankruptcy and the investors have lost almost all of their money.
The story illustrates the phenomenon of confirmation bias in the very highest echelons of business. Confirmation bias is the tendency to search for, interpret, favour and recall information that conforms to pre-existing beliefs. In the Theranos saga, the investors and board members were all unfortunate victims; they did not perform the required due diligence and consistently neglected taking a hard look at the metrics. The most apt example here is George Schultz, who despite been shown strong evidence of manipulated test results from his own nephew, Tyler Schultz – a senior employee at Theranos who was the actual whistleblower – doggedly refused to accept the evidence. Later, and only after many years, did Schultz admit that he fell in love with Holmes’ vision and determination and did not want to miss out on the big opportunity.
“Companies and the government ultimately respond to the collective consciousness of the people, and the best way to communicate with people is through the press,” [Tyler] Shultz said he learned. “Talking to the Wall Street Journal was way more effective than talking to [government regulators] or the company itself.” (Ellie Kincaid, 2017, October 03, Forbes)
Peter Wason, the English psychologist who coined the term ‘confirmation bias’, conducted a series of experiments in the 1960’s. Wason in his study successfully demonstrated that most people do not proceed optimally and aim to confirm pre-existing beliefs instead of attempting to falsify a hypothesis. Believe it or not, man has always been a biased animal. We are all prejudiced against change and nowhere is this more apparent than in organizations. Anyone who has been in a key decision-making meeting may have witnessed various manifestations of this scourge.
Consider a business that is launching a new product. The manager has an idea about the “next big thing” so he directs his team to conduct market research in order to explore feasibility. The team proceeds to conducts surveys and engages in competitive analyses but with the end result in mind, and confirmation bias has set in before the project even gets off the ground. First, the manager is using market research as a facade to confirm his own preconceived notions about a product idea. Next, the team launches into the product development process knowing exactly what their boss wants. As a result the research will very likely be biased in order to give the manager the answers he seeks. Whereas this is a hypothetical scenario, situations like this are all too common in many organizations.
Confirmation bias is dangerous for many reasons, but most notably because it leads to flawed decision-making in organizations. Here are a few techniques you can employ to help mitigate the problem:
Identify all pre-conceived notions that you are emotionally attached to. Ascertain when confirmation bias occurs and stop yourself mid-process.
Guard Against Overconfidence
The more certain you are about an opinion, the more likely it is that you will ignore dissenting evidence. Often, this is the best time to stop and question yourself.
Seek Out Contrary Evidence
Karl Popper’s theory of falsification states that a falsifiable hypothesis is one which can be tested, and conceivably be refuted. Actively search for evidence to disprove your theory rather than for confirmatory evidence.
Look for expert opinion, reference manuals, credible research or eyewitness accounts and consider the sample size of studies.
Be Open to Alterations
When you find evidence that contradicts your premise, modify incorrect aspects of your original theory instead of rejecting the evidence or abandoning the theory altogether.
Play Devil’s Advocate
A devil’s advocate takes up an opposing position for the sake of healthy debate. Appoint a few people from your team to play this role when crucial decisions need to be made.
Most of us see ourselves as intelligent and open minded. So, how can our beliefs persevere in the face of clear empirical evidence to the contrary? Even when facts are proven beyond doubt to be false, Wasson’s experiments found that many ostensibly sane people continue to find ways to alleviate the ensuing cognitive dissonance.
Confirmation bias is normal but extremely difficult to overcome; employing some of the strategies above will help overcome the issue and create a more rational and successful organization.
About the Author
General Manager and Vice President – Shared Services
With over 18 years of experience in the IT industry, Rasika is the Vice President, General Manager and Head of Shared Services at Sysco LABS in Sri Lanka. He played a key role in launching the Sysco LABS Sri Lanka operation and growing it to a high caliber engineering organization. He has played multiple roles in delivery, operations, HR, IT, finance, marketing, administration, and facilities management over the course of the last several years. Prior to Sysco LABS, he was Delivery Manager at Millennium IT where he led global delivery for some of the company’s largest accounts in North America, Europe, Asia, and Africa. He has also worked at Informatics where he developed systems for key local and international clients.
Rasika has a BSc in Computer Science from the Manchester Metropolitan University, an MBA from the University of Hull, and is a Project Management Professional (PMP).