Of course if they made it based on popularity / page views then it will just encourage click-bait articles even more so it's a somewhat damned-if-you-do scenario.
I've spent some time thinking about this (but eventually chose to pursue a project where the biggest challenge was technical rather than the business model), and there's another consideration you don't mention:
The total value that a publisher gets from a platform like Medium does not grow proportionally with the publisher's size, only a fraction of the value does. Specifically, the technical work that Medium does to scale the website, and the exposure that popular publishers get from being featured on the Medium homepage, Editor's picks, etc. However, the more popular a publisher gets, the more the balance tips in favor of leaving Medium for something with lower marginal cost to the publisher, like self-hosting. (These also tend to offer the publisher significantly more control.) As a result, to keep popular publishers on the platform, Medium has to lower their margins for them, whether by special deals or otherwise. This is the downside of Medium's platform having only a modest network effect: only modest network lock-in. (The upside was a minimal chicken-and-egg problem.)
----
Also, re click-bait:
I think it's worth drawing attention to the fact that readers are discouraged from supporting publishers with one-time payments (it's presumably still technically possible by signing-up for and then immediately canceling a monthly membership). If most of a publisher's revenue is from recurring payments due to the loyalty of readers, rather than one-time payments like ad impressions or micropayments or a Reddit gold-equivalent, that helps mitigate the incentives favoring cheap click-bait.
Comments
I've spent some time thinking about this (but eventually chose to pursue a project where the biggest challenge was technical rather than the business model), and there's another consideration you don't mention:
The total value that a publisher gets from a platform like Medium does not grow proportionally with the publisher's size, only a fraction of the value does. Specifically, the technical work that Medium does to scale the website, and the exposure that popular publishers get from being featured on the Medium homepage, Editor's picks, etc. However, the more popular a publisher gets, the more the balance tips in favor of leaving Medium for something with lower marginal cost to the publisher, like self-hosting. (These also tend to offer the publisher significantly more control.) As a result, to keep popular publishers on the platform, Medium has to lower their margins for them, whether by special deals or otherwise. This is the downside of Medium's platform having only a modest network effect: only modest network lock-in. (The upside was a minimal chicken-and-egg problem.)
----
Also, re click-bait:
I think it's worth drawing attention to the fact that readers are discouraged from supporting publishers with one-time payments (it's presumably still technically possible by signing-up for and then immediately canceling a monthly membership). If most of a publisher's revenue is from recurring payments due to the loyalty of readers, rather than one-time payments like ad impressions or micropayments or a Reddit gold-equivalent, that helps mitigate the incentives favoring cheap click-bait.