The culture of experimentation: from digital products to the creation of new companies.

Do you know what companies like Amazon, Google, or Netflix have in common? They have all made experimentation their main muscle of innovation.

In an increasingly uncertain, changing, and competitive environment, the ability to adapt makes the difference. As neodarwinist theories suggested: it's not the biggest or the oldest that survive, but those who best adjust to change. In that context, experimenting is no longer an option, but a necessity. It has allowed tech giants to evolve quickly, minimize risks, and discover new opportunities ahead of others.

But what does it really mean to experiment? And how can this principle be applied, not only to improving products but also to creating new businesses from scratch?

1. From intuition to evidence.

For decades, many strategic decisions have been based on intuitions, market trends, or the vision of a leader. But in recent years, we have witnessed a radical transformation: the need to validate every step has emerged. From designing a feature in an app to launching a new business unit, experimentation has solidified as the smartest (and least expensive) way to reduce uncertainty.

The evolution of the digital world has been key. In digital products, the culture of A/B testing, continuous user testing, and the data-driven approach have generated learnings that today extend beyond software. The real challenge now is to transfer that logic to something much more ambitious: the creation of new companies.

2. Experimentation in Venture Building.

Venture Building is, by definition, a discipline based on constant validation. It consists of building companies from scratch, but not from assumptions, but from evidence (or at least, as close to it as possible). In this context, experimentation is not a one-time phase, but a continuous mindset that guides the entire creation process.

The venture building process is structured in different phases that may vary in nomenclature, but share a common goal: to test hypotheses and learn quickly. Broadly speaking, these phases usually include:

• Exploration and Discovery: Identify a real problem, understand the user, map the market. Here it is validated that the problem is relevant.

• Validation: We build prototypes and develop experiments that allow us to validate our hypotheses with real customers to reduce risk until we feel it is time to launch the business.

• Problem-Solution Fit: Define a viable solution and test it with users. Does it fit? Does it really solve the problem?

• Product-Market Fit: Analyze if there is enough demand, and if the business model can scale.

• Go-to-Market and Scaling: Once everything previously mentioned has been validated, the company is built and launched to the market with a clear and refined proposal.

In each of these phases, experiments are designed: interviews, landing pages, acquisition campaigns, low-medium fidelity prototypes, pricing tests, controlled pilots… and we call all these elements “artifacts.” The goal is not to build a perfect product from the start, but to gather enough evidence to make informed decisions and, in some way, reduce uncertainty as much as we can. And, of course (this is equally important), kill ideas in time if the data doesn't support them.

No one wants to invest vast amounts of money in something they still don't know will work. Typically, many projects fail and stagnate because they are not solving a real problem or do not have the necessary metrics. Many times, large sums of money are invested in solutions that end up in the drawer. Therefore, what we try to do is reduce risk by validating first.

With a market validation and user-oriented approach, our “loops” (that's what we call the different validation phases) are the best way to validate opportunities and hypotheses quickly and efficiently. For this, we build experiments, prototypes, and MVPs with the necessary features to test our hypotheses with early adopters and improve the solution until we find the “problem-solution fit,” to generate transactions in a scalable model.

3. What if we do it from a large corporation?

When this approach is applied in the context of a Corporate Venture Builder, the advantages multiply. Experimenting from (or alongside) a large company allows access to resources, channels, data, and knowledge that a traditional startup does not have. But above all, it allows concepts to be validated more quickly, more deeply, and more aligned with the real market context.

Among the most relevant benefits are:

• Direct access to real customers to test value propositions quickly.

• Existing distribution and sales channels, which can accelerate the Go-to-Market.

• Reputation and brand that generates trust in early stages.

• Market data and internal “know-how” that allow for a better definition of starting hypotheses.

• The not-so-famous “know-who”, which turns out to be fundamental in corporate environments.

• Initial investment capacity, reducing financial pressure in the early phases.

That said, for this formula to work, the corporation must adopt a mindset different from the usual: less control, more autonomy; less closed planning, more openness to change. Success in Corporate Venture Building does not come from predicting the future, but from building it through experiments.

56% of companies in Spain acknowledge that certain aspects of their organizational culture prevent them from being as innovative as they would like.

And this is despite nearly 15 years since Eric Ries' Lean Startup. For the uninitiated, Lean Startup is a system for transforming uncertainty into validated knowledge and building sustainable businesses through continuous experimentation. And no, it seems we are not learning.

4. The evolution of experimentation: from product to business.

In the last ten years, the culture of experimentation in products has gone from something marginal to becoming standard. Today, any digital team is expected to iterate quickly, listen to users, and constantly measure the impact of their decisions. That same approach must now be transferred to the creation of companies.

The most common mistake in corporate innovation is assuming that launching a company requires having a solid business plan, a solution ready to scale, and a complete team from day one. In reality, what is needed is a testing attitude, a modular structure, and a radical openness to change. Because launching a venture is an exercise in continuous learning. 

And because, in most cases, the value lies in what is discarded, not just in what is launched.

5. Validate to build with meaning.

Validating is not just testing if something “works.” Validating is constructing meaning. It is knowing why something has value, for whom it has value, and under what conditions. It is understanding which part of your hypothesis supports the weight of reality and which does not. And in this process, each experiment is a compass that points whether you are closer or farther from building something valuable.

In summary, experimenting is no longer optional. It is the only viable path to innovate intelligently. Whether you are optimizing a digital product or launching the next big company from within a corporation, the key is to validate, iterate, and learn. 

And you, are you building with certainties… or still with assumptions?