Because a lot of news sites know you're not loading them intentionally in your browser, but landing on them because of shares or search. Meaning you may only se

Because a lot of news sites know you're not loading them intentionally in your browser, but landing on them because of shares or search. Meaning you may only see their pages 10x a month or similar even without a paywall. The hope with the paywall is if you keep seeing it, you'll go ahead and pay to unlock it, and the amount you're paying offsets the transaction fees. 1000 users each paying 20 cents for ten articles (so assuming -double- the rate you propose, and charging only at the end of the monthly billing cycle) is still a loss of likely around 1/4th the revenue even with a cut rate payment provider (due to the large fixed cost of 2-4%). So might make $50. It's -negative- if using something pleasant to integrate with (because they will charge for that convenience with a larger fixed cost, like Stripe, who will charge $.30 + small percentage per charge), unless you work with them to design a new pricing model for you. Converting 50 of those thousand (so just 5%) of those people to paid users at $5 each is going to be north of $200 you keep, even if you use Stripe and don't negotiate anything. And for those users that -will- load up a news site intentionally, the economics are even starker; a subscription means you'll likely go to their site, and stay there. Pay-as-you-go pricing incentivizes you to not visit the site, and instead find other, cheaper news sources. AWS pricing works because of the scale of the resources people tend to use, and that for businesses, using those resources = additional revenue. It is spending money to make money.