After reading this article, I have many questions and doubts. First, how do you know medium’s user base cares about the direct messaging and inbox feature? Have you done any research to support that? For example, I do not wish Medium to become another social network personally and if I want to network with someone, I probably will choose LinkedIn or FB or other social media. I come to Medium to consume content and if I want any engagement with the author, I would probably respond to their articles as I’m doing now. So I don’t see how “obviously” Medium users would embrace this feature.

Secondly, what’s the rationale behind your design decisions for all wireframes or high-fi prototypes? How does that impact user behaviors?

Thirdly, I disagree where you titled: “Gather feedback and iterate: In a real scenario, this phase would happen after the feature is developed and made live to the general public.” User research should be done before you start the redesign and testings can always be done before a feature is built. No companies are going to develop a feature without testing it either internally or with a small group of users before it goes public, simply because the cost/risk will be too big if you don’t validate it enough.

I appreciate the thought on improving the platform/service but I also think you need to incorporate some validation there to make your design justified, even it’s just a design exercise.