Assumptions

Every experiment starts with an assumption to validate. You have an assumption about the world around you that needs to hold true, if your project is going to be a success.

Examples:
Let’s say you have the idea of creating a webshop. Some important assumptions:

1. People want to buy what you are selling
2. You can reach these people through an online webshop
3. They are prepared to buy online
4. You can sell your products for a price that allows you to have a profit
5. …

For every startup or new product or service it is possible to come up with a number of assumptions that you will need to prove, in order to front-load the risk that is associated with actually building and launching.

In the case of the webshop, it pays off to test the listed assumptions before building any complex multi-language international webstores.

For example, this article from Business Insider writes ‘Jeff Bezos had a vision for the company’s explosive growth and e-commerce domination. He knew from the very beginning, he wanted Amazon to be “an everything store.”’ Nevertheless, Amazon started with books, because they had the (proven) assumption those would be easier to buy online than, say, clothing. Only later did Amazon grow into the online shopping walhalla we know today. They started small, first proving they could get customers, before working on better technology and infrastructure.

In fact, in the same Business Insider article, it says, ‘Amazon got started out of Bezos’ garage and the servers that the company used required so much power that Bezos and his wife couldn’t run a hair dryer or a vacuum in the house without blowing a fuse.’ How’s that for being scrappy!

As an entrepreneur or innovator, you’re responsible to tackle the biggest risks that can kill your startup early on. You start out with a vague idea with a lot of misconceptions and unproven assumptions, and you need to learn, you need to build a mental model that is an accurate reflection of reality.

That mental model will always be based on assumptions, there is no way around that. And that doesn’t have to be bad: some assumptions may be correct, while others may be wrong, but have little impact on the result.

Correct assumptions and the ones that have little impact are not the assumptions you should waste time on. You’re looking for are the assumptions that have a huge impact when they’re incorrect. Those are the ones you need to validate.

It’s your job to make sure that you find those dangerous assumptions and validate them, the riskiest one first.

Riskiest Assumption
So, how do you find the most risky assumption? Which one should you validate first?

First of all, understand that you are looking for the assumption that is the most risky at this moment. Later on, there may be other risks that are even riskier at that moment, but the one you’re looking for is the one that dictates a ‘go/no go’ at this very moment.

One way of looking at it is to think of risky assumptions as ‘gating’ each other. Some assumptions only become problematic if other assumptions are already proven. But if that earlier assumption is wrong, there won’t be a ‘later’.

Example:
In the Amazon example above, the quote about the garage using too much electrical power is a clear problem that needs to be solved. Blowing too many fuses at some point meant they would go out of business. But it could only become a big problem after it turned out they could find so many customers. The assumption ‘we can get customers to buy our books online’ was more fundamental than the assumption ‘we need a better server infrastructure to make sure customers can buy our books online’ or even ‘we can get thousands of customers to buy our books online’.

The search for the riskiest assumption is not different from first surfacing and then prioritizing your assumptions, and tackling the most fundamental one first.

So, if you would list all of the assumptions that are underpinning your point of view, and rank them from most important to least important, you’d ideally want to start with testing the most important one. In the Lean Startup, this is called ‘The Riskiest Assumption’. It’s called that, since, if this assumption turns out to be incorrect, you will need to rethink your entire startup.

Luckily, there is a rule of thumb to follow. Each stage in the innovation journey has a number of ‘default’ associated Riskiest Assumptions. Only when those are validated, you are ready to progress. They are:

Idea / problem-market fit stage:

• Customers understand the idea
• You are able to find customers that experience the problem and care about it (Problem Market Fit)

Problem-solution fit stage:

• You are able to find solution options that resonate with customers
• Customers really prefer your solution over alternatives (Problem Solution Fit)

Product-market fit stage:

• You can reach interested customers
• You can get customers to commit
• You can get customers to come back
• You can get customers to bring new customers
• You can get customers to pay
• Over 40% of customers do not want to stop using your product (Product Market Fit)

These ‘default’ riskiest assumptions are the ones you should at the very least consider when finding your current Riskiest Assumption. They can often be split up in even smaller assumptions, or be modified to fit with specifics of your project. But if you can’t validate these steps, you’re in trouble.