This is a fairly basic principle in PM, but the two are often confused so I would like to discuss them in detail.

I’ll start with assumptions. Assumptions simplify complex problems. Making assumptions is a core role of a product manager. Product managers make assumptions about users (customers), human behavior, context, psychology, cognitive bias, etc. Product Managers usually create new features so making assumptions about the uncertain future is natural. However, these assumptions have to be scientific and methodical.

Identifying your assumptions is one of the first step towards creating something. And to do this, you must ask why to every single little thing you’re doing. Why would a user do use this? Why would a user pay for this? Why would a user come back to the website to use this again?

Assumptions cannot be proved or disproved – they’re not really data-driven. However, hypotheses can be backed up by data.

Here are examples of some common assumptions:
“In order for my idea to be successful, the following must be true…”
“My user has problems x,y,z”
“Productivity matters to my customers”
“There is no satisfactory competition to my product”

If a lot assumptions have to be true for your product to be successful, test the riskiest one first so that you can save on opportunity costs. (Risky as in if that assumption is not true, your entire product will fail without a doubt). Usually the riskiest assumption is the one that assumes users specifically have the problem you want to solve.

Hypothesis is something that translates your assumption that is testable for the purpose of a scientific experiment.

Example of a hypothesis:
We believe that adults aged between 20~30 will pay for our product because they need to find efficient means of transportation during rush hour.

A good format that I always use is:
We believe [subject] has a [problem] because [reason]. If we [action], this [metric] will improve.