What are the best ways to avoid confirmation bias in startup validation?
As an early-stage startup founder, you have a vision for your product and your market. But how do you know if your assumptions are valid and not just wishful thinking? Confirmation bias is the tendency to seek, interpret, and remember information that confirms what you already believe, while ignoring or discounting evidence that challenges or contradicts it. Confirmation bias can lead you to waste time, money, and resources on building something that nobody wants, needs, or pays for. In this article, you will learn some of the best ways to avoid confirmation bias in startup validation and experimentation.
One of the best ways to avoid confirmation bias is to test your riskiest assumptions first. These are the assumptions that, if proven wrong, would invalidate your whole idea or business model. For example, if you are building a subscription-based app, your riskiest assumption might be that people are willing to pay for it. Instead of spending months developing the app and launching it to the market, you should test this assumption as early as possible by creating a landing page, a mockup, or a prototype and asking potential customers to sign up or pre-order. This way, you can get real feedback and data on your value proposition and pricing before investing too much time and money.
-
Start your business by attempting to sell to people you don't know. Way too many founders begin by selling and serving their friends, colleagues, etc. You want to rapidly build muscle re: learning how to sell, hearing tough feedback, losing customers who don't have any other relationship with you, and ringing the cash register.
Another way to avoid confirmation bias is to seek disconfirming evidence. This means actively looking for information that challenges or contradicts your assumptions, rather than only focusing on the positive or supportive ones. For example, if you are conducting customer interviews, you should ask questions that explore the problems, needs, and alternatives of your target segment, rather than leading them to agree with your solution or praise your features. You should also seek feedback from different sources and perspectives, such as potential customers, competitors, experts, mentors, or peers, and compare and contrast their opinions and insights.
-
When interviewing people in your proposed target market, ask open-ended questions about the problem you're trying to solve. Most founders ask questions about their solution, and this is where confirmation bias creeps in. This is hard to do, especially for founders who are feeling this pain point themselves! Be curious. Ask why. Go down rabbit holes. Stay committed to solving the problem and not committed to your specific solution.
A third way to avoid confirmation bias is to use the scientific method. This means following a systematic and rigorous process of formulating hypotheses, designing experiments, collecting and analyzing data, and drawing conclusions based on evidence. The scientific method helps you to avoid confirmation bias by forcing you to be clear and explicit about your assumptions, to test them with measurable and objective criteria, and to update or revise them based on the results. For example, if you are running an online survey, you should define your target population, sample size, and margin of error, use unbiased and relevant questions, and use statistical tools and methods to interpret the data and infer the implications.
A fourth way to avoid confirmation bias is to embrace uncertainty and failure. This means accepting that you don't know everything, that you might be wrong, and that you can learn from your mistakes. Embracing uncertainty and failure helps you to avoid confirmation bias by encouraging you to be curious, open-minded, and humble, rather than defensive, stubborn, or arrogant. For example, if you are launching a minimum viable product (MVP), you should treat it as an experiment, not as a final product, and be ready to iterate, pivot, or kill it based on the feedback and data you collect.
-
Failure is a key part of the startup process. Launching an MVP isn't about getting it perfect on the first try; it's about real-world testing and learning. It's through this testing that we truly understand our customers and what our business needs to succeed. Failure isn't the end—it's the feedback we need to make our ideas stronger and our businesses more adaptable. Let's not fear failure; instead, let's use it as a tool to build something better, one iteration at a time.
A fifth way to avoid confirmation bias is to build a diverse and honest team. This means working with people who have different backgrounds, skills, experiences, and perspectives, and who can challenge and support each other. Building a diverse and honest team helps you to avoid confirmation bias by exposing you to different viewpoints, ideas, and opinions, and by providing you with constructive and candid feedback. For example, if you are brainstorming solutions, you should invite people from different departments, roles, or functions, and encourage them to share their thoughts, questions, and criticisms.
A sixth way to avoid confirmation bias is to use tools and frameworks. These are methods, models, or templates that help you to structure your thinking, organize your information, and guide your actions. Using tools and frameworks helps you to avoid confirmation bias by providing you with a common language, a clear process, and a consistent approach. For example, if you are validating your problem-solution fit, you can use tools and frameworks such as the lean canvas, the value proposition canvas, or the jobs-to-be-done theory.
-
An additional insight for avoiding confirmation bias in startup validation is to actively seek out diverse perspectives and feedback from individuals who represent different social and regional backgrounds. By intentionally involving a range of voices, you can gain a more comprehensive understanding of how your product or solution may be perceived and utilized in various contexts. This approach helps counteract the tendency to rely solely on feedback that aligns with pre-existing beliefs, leading to a more balanced and accurate assessment of your startup's potential impact and value across diverse communities and regions.
-
Calculate revenue figures, and metrics like churn, cost of acquisition, lifetime value, in a way that disregards any special cases, interested party transactions, and bespoke or custom work. Otherwise, you might just be confirming that you can book revenue by selling to your mom. Getting your first sales, and customer fulfillment and feedback, are very important wherever they come from. It proves you can make, sell, and successfully deliver a product… plus your mom may have some great feedback. So by all means, sell to mom. But don't convince yourself that what you did is replicable at scale. There's only one of her, and millions of arm's length customers who are not.
-
As an entrepreneur who has spent eight years on creating a new service I can attest to the importance of understanding the confirmation bias and diffusing it early on, it has to be your first job. Sometimes you may nail everything perfectly right, but sadly you are not the right agent for the market you are trying to liberate from the siege of the conventional and existing solutions and players in that market. For example, you are selling a telemedicine video chat service, you have interviewed 100 doctors, even a regional insurance company has agreed to work with you, because you are a superhero in convincing people, but since you do not have the right connections nationwide, your solution will not simply scale nationally.
-
Sometimes your "mom" represents your target market and has worked with hundreds of start-ups so don't discount her experience because she disagrees with formulaic (tired not true) venture dudes. When we built the first online platform for early childhood education, we faxed every school and school district in the country. Within a few years, we had millions of users for a product no one had heard of in schools with no connections or computers. Our goal was to help poor children learn to read. Not to get rich. In the end, we gave away hundreds of millions of free lessons to children in 200 countries. We would have made a bigger impact if more than 3% of self-replicating venture went to women-owned companies but heck, I will take the win.
Rate this article
More relevant reading
-
Startup DevelopmentYou’re in the startup industry. How can you prepare for the next wave of innovation?
-
Startup DevelopmentHow can you solve problems faster to stay ahead in Startup Development?
-
Start-up VenturesHow do you scale and pivot your startup based on your experiment insights and customer feedback?
-
Startup DevelopmentHere's how you can manage feedback from multiple sources in a startup.