>That's... not an easy thing to do.

Offering something easy to post that's looking good across devices?

Gimme a break, they have a single basic template, and minimal authoring tools. With these constraints, there are literally tens of thousands of Wordpress, Ghost and vanilla HTML templates by various developers/designers with 1/1000 the resources of Medium that look just as good across devices.

Well they did that by basically limiting what the user can do. The editing options were extremely limited. There was also no customisation until recently. They built a basic-function responsive blog platform. How is that hard?

It's very hard to tell people to use your software that has 1/100th of features of the incumbents but it's done better. Apple did it with iPhone for instance, but normally startups die trying.

That's because the iPhone didn't have 1/100 the features of competitors -- it had 80/100 of the features, but focused on the 80% that mattered (no "fm radio" or "mobile flash" for example).

Imho that was mostly achieved by marketing , not technical innovation (which they claim)

Also helps a lot with the amount of connections Ev Williams and co have. I would guess that having all the right connections was a huge part of Medium's success, especially really early on. With enough connections you can grow (almost) any half-decent service.

hard != innovative, was OP's point.

OTOH, solving well-established hard problems sometime requires more true innovation than developing new ideas. I prefer to do both.