Tumblr will kill RSS

posted 13 March 2013

Google's shuttering of Reader is a failure to find a business model for RSS.

Back in 2001 or so, when RSS started getting really popular, my first question was: why would any ad-supported site do this? It's obviously really convenient for me, as a reader, to get the full content of your site without having to visit it, but that destroys page views and thus ad revenue for the publisher. Even if the RSS feed consists only of headlines, that eliminates one, perhaps several visits I might have made to the home page each day to check for new headlines manually. Great for me, terrible for the publisher.

The monetization idea for RSS was there in the name: syndication. In old media, syndication was a content-sharing agreement settled upon for a fixed time and a nontrivial fee, whereby a single author could get into dozens or hundreds of separate publications that, crucially, didn't compete with each other -- if your article appeared in both the Des Moines Register and the Miami Herald, neither publication was bothered about that, because their overlap in readers was negligible.

The problem for RSS is that the Internet doesn't work that way. There are no non-overlapping markets: if your RSS feed allows another website to display your content, they are going to steal your page views directly. That might be okay if you made more money that way, but RSS doesn't charge a fee! It's just giving away your content for no apparent reason.*

That problem might go away if you could find some way to monetize RSS directly. Enter Feedburner! It was supposed to provide you with readership stats for your feeds (to compensate for the apparent decline in readership of your site when people switch to RSS) and, eventually, provide you with income via ads in your feeds.

Google acquired Feedburner for something like $100 million in 2007. It then launched AdSense for Feeds, which was supposed to be the way to monetize the feeds. But it never worked. Either advertisers didn't buy the ads, or readers didn't click the ads, but last September Google shut down AdSense for Feeds.

If you can't monetize the feeds themselves, the only other thing you can try, if you're Google and you've paid all this money for FeedBurner, is to try monetizing the RSS reader application itself. With the shutdown of Google Reader today, that experiment is now over.

So now what?

Is RSS dead, and if so, what will replace it? In the short term, some service will spring up to replace Reader (and, when it is eventually killed off, Feedburner itself). In the longer term I think we need to look harder at the business model of RSS itself. The web grows fastest in ways that are mutually beneficial for everyone, and RSS's benefits seem one-sided.

For users, there is undoubtedly value in the time saved not having to go to the front page of every site every day to check for interesting articles. To some extent, that purpose is being served by ad-hoc social distillation of news via social media, especially Twitter and Facebook. There are also news aggregation sites like Techmeme (and its political cousin Memeorandum), as well as news/social hybrid sites like Hacker News and Reddit.

For publishers, particularly lower-volume publishers, there is value in having readers be able to "subscribe" to your site, i.e. getting notifications every time you publish something. However, notification is really all you want -- to drive visitors to your site. Giving away content for free is never going to be attractive to publishers who have to pay people to write it.

The clearest model I see for the future of content syndication is Tumblr. Big publishers have been hopping on for a while now: the New Yorker, the Economist, the New York Times (half-heartedly), the Guardian, etc., and they've been enthusiastically aided by Tumblr itself in doing so. The model is simple: readers subscribe to your content by following your Tumblr, which posts a curated subset of your content, edited for the quick-glance format of the Dashboard, in the hopes that readers will click through.

This seems to me to work better for everyone. Tumblr's built-in reblogging is the ultimate in social amplification mechanisms, ensuring that a publisher's best posts will be seen by far more people than actually subscribe to their feed. Going viral so easily is the carrot that brings publishers to the platform.

For users, instead of an inbox-like interface with thousands of "unread" posts** of equal importance, they get a continuous feed of new content on their dashboard, and reblogging ensures that particularly interesting content will be repeated multiple times as different friends reblog it, so you're less likely to miss it. This is a more natural, social, and less frustrating mechanism for surfacing the best content.

Coming up next

Of course, Tumblr isn't perfect -- its audience skews young, it's perceived as unserious, and it prefers quick visual hits to long-form writing. There are already dozens of potential competitors to become the source for socially-filtered content. But I believe Tumblr, or something very like it, will be the eventual winner in this space.


* There may be non-free ways to make syndication work. Felix Salmon is a fan of paid syndication of web content, though the drawbacks he lists in that post are enough to make continue to doubt it as a viable model or a desirable practice.

** The feature that always turned me off of RSS readers. You can't give me an "unread" count for the whole web and not expect that to drive me nuts.