The first and biggest question you need to ask, before designing a product that can be used by everyone is 'whom are you building for?'

When building their AstroBot Amazon made a list of potential user groups: software developers, business analysts, data scientists, designers, researchers, marketers. The product team reached out to potential users in those roles and spoke with them about their pain points and needs. 

On the basis of those conversations, the company determined that software developers were most likely to use the product.

The best way to determine that their product is useful is to put it in front of the people most likely to use it. This can mean market testing, or it can mean asking people on your own team--ideally, those in roles it's designed for--to give it a go.

Treat your prototypes not as finished products but as conversation pieces. You're not only trying to find faults in your design, but also validate that the use cases that you identified are correct and identify if there is something for the future that you can put on the road map.