A Scientific Approach to Startups

Why is it that the vast majority of tech startups result in failure? According to some sources, up to 90% of startups will ultimately fail. But is there a way to increase your odds of success? Could there be a way to more confidently move forward with a new startup and increase your odds of being one of the success stories? 

In the competitive world of tech startups, many entrepreneurs jump into building their ideas without sufficient research. This often leads to difficulty getting to what we refer to as “Product-Market Fit”. Never reaching PMF is the primary reason early-stage startups fail. 

Over the years, my team and I have developed a very structured process for evaluating new startup ideas that significantly increases the odds of quickly getting to PMF with the ideas we choose to back. Our systematic, data-driven methodology helps us know that the companies we decide to invest in have a higher probability of success. 

Our approach has a lot in common with the scientific method and uses many of the same processes to test and validate new product concepts. In this blog post, we'll delve into the major steps of our scientific approach to launching new tech startups.

Step 1: Focus on Problems, Not Solutions

The first step is to identify a problem that needs solving. We try not to start with a specific solution in mind, but rather with a problem we think needs to be solved. Entrepreneurs should concentrate on first understanding the target customers' pain points and verifying that the issue at hand is genuine. This will require market research to confirm the problem's existence and ensure the market isn't already saturated with adequate solutions.

Step 2: Develop a Hypothesis

Once the problem has been identified and validated, it's time to develop a product hypothesis - a proposed solution that will appeal to the target customers. This hypothesis should be a unique, innovative way to address the problem. This is an opportunity to dream up how to solve the problem in a way that customers will love and that is going to be difficult for potential competitors to replicate. 

[For more on what makes for a good business idea, check out this blog post]

Step 3: Run Experiments to Test Your Hypothesis

With a product hypothesis in place, it's time to test it with real customers. The goal is to actively try to disprove your hypothesis through unbiased customer discovery. Begin with customer interviews to gather feedback on your proposed solution. At this stage, you haven’t “built” anything. You’re just asking for feedback on the idea itself. If the response is positive, you can then progress to wireframes or prototypes (for software) or proofs of concept (for hardware), enabling customers to provide even more direct feedback. 

Be open to criticism and avoid confirmation bias to ensure your idea is truly viable. This can be really difficult for first-time entrepreneurs to do. It is tempting to want to look for feedback that validates your initial ideas. But you should actively try to find reasons why your idea is not worth pursuing. The easiest way to increase your odds of success is to avoid burning too much time on the bad ideas in the first place.

Step 4: Replicable Results

After gathering feedback, continue to refine and retest your hypothesis until multiple customers confirm they would buy your product (if built). We typically recommend setting a goal for pre-sales using a high-fidelity prototype tool, such as Figma, to gauge customer interest. If you can't generate a significant waiting list based on a basic prototype demo, it might be a sign that your product will be difficult to sell.

Step 5: Confidently Move Forward with Development

Only after completing the previous steps successfully do we recommend moving forward with launching a new company to build and deliver the proposed new product. By rigorously testing a hypothesis and gathering pre-sales, we can confidently move forward with development, knowing that there's a real demand for the proposed solution. And at each step of the process, we are actively looking for reasons not to move forward. 

This scientific approach to launching new tech startups significantly increases the likelihood of success by ensuring we’re solving real problems in unique and defensible ways as validated by actual customers. By following a process such as this, you can dramatically increase the odds that your new startup will be one of the select few that survives and thrives.