3 ms·
Clearly, a subscription of a few dollars a month is MUCH more valuable that aggregated article hits to the same total value: a subscriber is someone who will re
by cornholio 9y ago
Clearly, a subscription of a few dollars a month is MUCH more valuable that aggregated article hits to the same total value: a subscriber is someone who will return because he chose you from the alternatives, he has a sunk cost in the subscription and will keep consuming, building familiarity and brand loyalty, develop favorite columns and writers, bookmarks and homepages etc. All these will make him return in the future and extend the subscription as long as you don't screw up in a major way.
A pay-per-article viewer is a mercenary that will come back only for a viral or exclusive content. These could be quite profitable when they happen, but are highly unpredictable thus have lower present value.
Maybe the correct approach to this market is via the lower value, ad supported content, where publishers would gladly accept a few cents instead of displaying the ads. Ad blockers seem uniquely positioned to bootstrap this business, the publishers are deeply aware of the ad blockers to the point they modify their sites to detect them - an API to accept the micropayment instead of nagging the user to turn off the blocker would be golden.
Once the system is in place and develops large revenue streams, the owners of premium content should grudgingly fold and accept pay-per-article.
- kenhwang 9y agoFunny you mention adblocker. You'd think it's the biggest concern for publishers, but they're not scared of what they don't know. News publishers don't exactly have nimble cutting edge engineering departments dedicated to min/maxing advertising revenue. Adblockers also block many general purpose analytics scripts that would inform the publisher that the adblocker is a threat. Very few publishers are willing to implement a hard paywall that could only be passed via confirmed payment. Especially when it relies on some 3rd party integration working. When there's a chance that that micropayment API doesn't work, no one thinks it's the payers fault, the blame always lies with the publisher in the user's eye, even when it's almost always the payer's fault. And there's nothing worse than pissing off a paying user. And hard paywalls require engineering investment, which newspapers tend to not have. It's practically effortless pushing static content up to a CDN with some ad serving scripts. It's a lot of work accepting micropayments on every request.