When designing MVPs you must optimize for learning, instead of purely delivery. With this in mind, the word failure has another meaning.

An experiment will result into something. When thinking about a hypothesis or a feature, you might be thinking about it in a positive way. It is fine to do so. But it is very important to organize your product architecture to handle localised “failure” as a possible outcome. Failure is a learning opportunity. You need it, but you must be well prepared to handle it.

Organize your product architecture to handle localised “failure” as a possible outcome. Failure is a learning opportunity. You need it, but you must be well prepared to handle it. Localised failure is a learning opportunity. #MVP . Tweet This.

Below are a few photos of an oil tanker and its internal architecture.

petroleiro tanker2 oiltanker layouts
Even though they are very big, they are designed with many smaller oil tanks compartments. In case a compartment has a problem, the boat does not sink. This design prevents an oil tank problem (such as leakage) to capsize the boat.

This is how your MVP (and features) should be architected. You must be ready for new features experimentation, without capsizing your product.

The top most important consideration is your boat (the overall product). It must always remain stable as the new features (compartments) are being worked on. These are the new experiments. They are important for validating new learning. But they cannot put your product at risk.

Innovative organisations handle it nicely. They run Lean Inceptions to align about the MVP. They are growing their products similarly to the large oil tankers. They handle new experiments on top of already proved successes, without risk of capsizing the boat.

lean-enterprise-oil-tanker

What about your organisation? Is it prepared to work this way? We’re not all in the same boat, but I hope yours is well prepared to handle the upcoming learnings.