Blendle [1] has had success with such an approach in the Netherlands where they have brought on board quality publications, letting users pay per article. They also let users ask for refunds if an article is not what it seemed, which punishes clickbait articles.
Although, to be honest, I'm not sure how to fix that problem. Trying to create a system that provides news people need instead of news people want is a fundamentally hard problem when those two things don't overlap.
Full disclosure, I work at PressReader. But it surprises me sometimes when we're not mentioned in these conversations. We do huge business internationally, have millions of active users, have been profitable for years, and are growing at an absolutely insane rate. It's a win-win business model because readers get content (often for free since it's sponsored by a brand they're a customer of), publishers make money (we pay royalties when their content is read), and brands have the opportunity to offer something tangible and personalized to their customers.
Better yet, visit the PressReader HotSpot Map and you'll see all the places you can get full access to PressReader for free. You just have to access it while you're connected to their WiFi: http://www.pressreader.com/hotspot/map
I guess Blendle is less like Netflix than PressReader, and more like iTunes in that you generally buy individual articles for a small price (often < 1 euro).
First, pay by the article means I can easily go through the annual amount I'm willing to spend on "news" in just a few active months. Then I have nothing left for the rest of the year. I don't want to go through the mental effort based on each article and how it _might_ compare to others in the future in order to decide what to read, regardless of having to make that decision before or after I read the article. In other words, it may be a fine article on in it's own right and fully worth 25 cents. But not compared to all the articles I might read in a year. For example, if I have $150 per year for "subscriptions" that means I get 600 (25 cent) articles per year or 20 per month. Is this article worth it?
Second, as mentioned in numerous blogs, paying by the article creates a race to push more articles with just the minimum quality to get the pay click.
Generally I've been happy with the articles I paid for based on its popularity and what publication it was from. But if I recall correctly I asked for my money back once or twice.
The reason I like Blendle is precisely that I've found few magazines or newspapers that are worth their subscription money. It's cheaper for me to pay for that occasional article that is written by a friend or recommended to me.
True micropayments would be more like paying a nickel to read an article. All we can say is that this has been talked about for at least 15 years and it hasn't happened. [1] Some sort of subscription that pooled a bunch of sources might work. The ones I've seen around magazines aren't nearly comprehensive enough though.
[1] http://www.openp2p.com/lpt/a//p2p/2000/12/19/micropayments.h...
Hopefully it would be more granular than just at the website level, perhaps being able to drill down to a specific journalist / category / article / whatever.
Maybe that feedback loop would shape things in a way that I would be more willing to subscribe to a entire publication vs cherry picking individual authors or content.
As an example, I really like vice.com's reporting and perspectives on Syria, ISIS, etc. However, I'm not a fan of their reporting on other topics, like gun violence, environmental issues, etc.
I would be much more excited about micropayments where I pay only for what I care about. In the same way that I like "on demand movies" over say, a subscription to HBO.