Experimentation As A Business Competency

It's not a nice-to-have anymore.

Nearly every entrepreneur agrees that success is not a linear path. Rather it is a series of ups and downs, twists and turns which eventually may lead to a business model that works. In this post, we provide a basic system to manage this turbulence and give your business a better opportunity to build something customer care about. 

Finding a business model that works is perhaps the most important thing we can do as entrepreneurs. It is also probably the most difficult. Syncing our product design and development with the needs of a customer base is no easy feat. It becomes even more challenging when we are not actively seeking information that tells us if we are on the right track to finding a match. 

Many entrepreneurs fail big because they fail to fail small. They spend resources developing a solution without validating it solves a customer problem that is big enough to be worth pursuing. 

You’re unlikely to go “all-in” at a poker table on the first hand and you shouldn’t with your business. Instead, we search for product-market fit by making a bunch of small bets that help us get enough evidence to justify increasing the size of the investment. To reduce risk we significantly increase the rate at which we try things. 

We can systematically reduce the risk of our business or product idea by quickly collecting evidence that indicates whether it will solve the problem for customers. So instead of waiting for a product to be fully baked before we launch, we create a rhythm of launches that allows us to test the product idea(s) early and often to determine whether we should proceed:


Not This:

The only way we can do this in a rigorous manner is by having a system that emphasizes launching Minimal Viable Products (MVPs). MVPs allow us to be more systematic about the evidence we collect from customers and more adaptive to their needs. Contrary to popular belief, an MVP is not necessarily your first product. Rather it is the vehicle through which you launch your experiment to obtain your desired learnings. 

A basic application of this system in your business contains three parts:

  1. Evaluate and Prioritize Risk

  2. Design Experiment 

  3. Kill, Pivot, or Persevere Decision

Evaluate and Prioritize Risk

Understanding the most uncertain aspects of your business is essential to managing your business’ risk. A Business Model Canvas is a great tool to document all aspects of your business. You can then evaluate where the risk exists. 

A simple way to think of your business risk is to consider the assumptions you have with the business. An assumption is a “We believe…” statement that describes an element of the business model. For example, Quibi may have an assumption such as “We believe customers want to watch entertainment series in sub-10 minute clips while commuting.” 

Generally, there are three areas where our business assumptions exist:

  • Desirability: Customers want what we are building

  • Feasibility: We can build or deliver what customers want

  • Viability: We support a business delivering what customers want

All risk is not created equal, nor are assumptions. When identifying what to test first, we need to prioritize the assumptions that are most important to our business success and the most uncertain. An assumption map helps us to identify our Leap of Faith Assumption (the assumption that is mostly unknown and most important to our business’s success). The LOFA is what should kick off our experimentation.

Assumption Map

Design Experiment

Once we have our LOFA identified, we can begin to design an experiment. Experiment design needs to be rigorous for it to be useful, but it doesn’t need to be expensive, difficult, complex or time-consuming. The simple experiment you launch tomorrow is better than the complex one launched next month. The key is to be disciplined in how we are designing the experiment so that we can have confidence that we are collecting the right data.

A proper experiment has these elements:


A “We believe…” statement about an aspect of the business model. 


A falsifiable “If… then…” statement related to an assumption. 

MVP/Experiment Archetype:

The vehicle by which we will test your hypothesis.


The data that indicates whether your hypothesis was proven or disproven.


The insight that drives your next action.

Kill, Pivot, or Persevere Decision

As we launch experiments/MVPs it’s critical that we make a regular pivot, persevere or kill decisions based on the learnings and data. These decisions drive the direction of future experimentation and pave the path for product-market fit.