How often do you use hypotheses on a daily basis? Chances are you might be formulating hypotheses more often than you realize. This is because, in very simple terms, hypotheses are guesses about expected results, outcomes, or relationships between aspects of the world you intend to understand. Guesses vary widely, going from hunches and intuitions to more educated and structured predictions. While there can be value in following one's intuition to develop a hypothesis, research in psychology has exposed the potential risk that derives from trusting one's instinct. Tversky and Kahneman, in particular, spent a lifetime working on mental shortcuts, or heuristics. Their influential research and subsequent studies on how individuals make sense of reality highlight how there might be several fallacies in the process of drawing conclusions from what we observe. There are several heuristics that we use on a daily basis. So for this reason, we're outlining a more structured process that should limit the chance of biased decision making through the systematic articulation of hypotheses. Hypotheses start as answers to a specific question. For instance: Would customers use my product or service? Your answer to this question is essentially a series of guesses. Several of these guesses often tend to remain implicit, but we have seen in module one that two, such as Business Model Canvas or balanced scorecards, can be used to gain clarity on key objectives and activities. They can also be leveraged to explicitly formulate our guesses or hypotheses. In the rest of this video, we're going to see, first, what are hypotheses, and second, how to develop them. First, to be more precise, we consider hypotheses as testable and falsifiable propositions made on the basis of evidence as a starting point for investigation, but without any assumption of their truth. To break this definition down and to understand how to practically develop hypotheses, we will use as an example Rent the Runway, a business founded in 2009 by Jennifer Hyman and Jennifer Fleiss. Rent the Runway offers the opportunity to rent designer clothing for a four to eight day period, for as low as 10% of the retail price through an online platform. The target are young women that look for high-end dresses for special occasions, such as weddings and proms. Rent the Runway implemented a systematic way to develop and quickly test their hypotheses when trying to address the following question: Is Rent the Runway a viable business idea? Would customers use this service? First, the founders developed testable and falsifiable propositions. Hypotheses should be articulated as propositions that can be tested and either proven wrong or right. And just to give you an intuition of how this works, you cannot test this proposition: Customers will be using my product or service, but you can test this one: Young women are willing to rent designer clothes. Second, hypotheses should be made on the basis of evidence. Evidence means what we know, and that forms the basis of our theory. In the case of innovative products or services, it might be difficult, as we face scenarios that are uncertain or unknown. However, as Ash Maurya - the founder of LEANSTACK - says: ‘You need to accept the fact that you will never have perfect information, and that you need to make this kind of predictions anyway.’ Rent the Runway started thinking about offering a designer clothes rental service, when they realized that several companies were renting tuxedos for formal occasions. That is when they realized that women might have been interested in a comparable service. The success of companies renting tuxedos provided sufficient evidence that Rent the Runway was worth hypothesizing about. Third, hypotheses are a starting point for investigation. The key purpose of hypotheses is to guide one's search for results. Why is this important? Well, without hypotheses we would engage in a blind search process where we're likely to just find what we're hoping to find. This is due to our confirmation bias, or the tendency to interpret results in a way that confirms our prior beliefs or assumptions, which leads to our fourth and last point: Without any assumption of their truth. Assuming that our guesses are true, it might lead to biased results. Rent the Runway, for instance, developed hypotheses and tested them rigorously. This often resulted in them having to revisit their original assumptions. For instance, their business idea relied on the assumption that women were able to correctly estimate their size without trying on the dress. And so that they would be able to order the correct size when renting a dress online. They conducted some tests and found that women were systematically choosing a smaller size than the one that would actually fit. Even though the original hypothesis was not confirmed, they learned that they had to address this problem in order for their rental service to work. Their solution was to include a size-up with every order. Having looked at what hypotheses are, how can we practically formulate them when it comes to innovation-related decisions? In the final video of this module, we will explore examples of what entrepreneurs and innovators in different sectors have done to turn their business idea into hypotheses, so you can learn from concrete cases. At a broader level, these indications will guide you in developing testable hypotheses. First, start with a clear question you're trying to address. Following Rent the Runway's example, that would be: Would young women be willing to rent designer clothes online at a fraction of the retail price? Second, understand what assumptions, guesses, or hypotheses have to be true for your idea to work. Start from the most basic fact and build on it, following a logical and consistent pattern. Rent the Runway started with: Young women are willing to rent designer dresses. Young women are willing to rent designer dresses they can't try on. Young women are willing to try designer dresses they can't see in person. This shows that the order of hypotheses is hierarchical and follows a precise order. The most basic fact that needs to be true is that women must be willing to rent designer dresses. The second one is that they must be willing to rent them without trying them on, and so on. If the first hypothesis is not confirmed by tests, the others aren't likely to be. A crucial point is that hypotheses need to focus on one aspect at a time. This is because it's extremely hard to test multiple aspects together. Third, make sure that your hypotheses are precise, consistent, and falsifiable. Hypotheses need to be tested through observations or experimentation. For this reason, they have to be sufficiently precise, so that conflicts of interest can be measured and propositions can be found true or not. In the next video, we will focus on hypothesis testing and on their context.
