February 11th, 2006

Is the Long Tail a Lit Fuse?


The demise of Publishing 2.0 was predicted early on — I’m still chugging along, but Phil’s point here is spot on — starting to blog is easy but blogging successfully over the long term is really, really hard. Which makes me wonder about the future of consumer-created media, especially in light of a fascinating analysis by Matt Galloway, which combines David Sifry’s latest numbers on the growth of the blogosphere with Umbria Communication’s comprehensive report on the state of splogs. Here are the results of Matt’s analysis:

I created an approximation of Sifry’s trend and carried it forward (assuming a constant rate of growth of 12.75%) out to May, 2006. I then created an approximation of Umbria’s splog trend for the same time period, using a constant 31% increase in the percent splog (same as 48% increase in number of splogs.) Finally, I subtracted the number of Umbria splogs from the number of Sifry blogs to yield an approximate number of non-spam splogs.

Then I graphed it and got this…

Yulp, you’re reading it right. Assuming that both Sifry and Umbria are accurate and that these growth rates are remaining constant – the non-spam Blogosphere has peaked in size and is now contracting. Furthermore, by March, blog spam will represent half of all blogs.

Anybody want to take a stab at what the heck is going on here?

I’ll take a stab at what’s going on, using Matt’s analytic technique — consider Matt’s analysis alongside this Wired report on “Podfading”:

Although hard figures are elusive, host Rob Walch of the podcaster-interview program Podcast411 estimates at least a fifth of podcasters don’t make it to their 10th show; he expects the podcast graveyard to become even more crowded as podcasting becomes easier. Walch instituted a rule that he won’t interview a podcaster until the show has at least 10 episodes.

“Podcasting is one of those things that’s cheap and easy to begin to do but takes a tremendous amount of time to keep going with no payoff,” said freelance writer and blogger Brian Reid of Alexandria, Virginia, former host of the gender-issues program Sex Talk, who quit in August. “There was no money in it and it did nothing to push my career forward. I’ve got a lot of other things in my life, paying work being one and my family is another. It’s not like blogging, where you can do it for 15 minutes at a time and get away with it.”

Consumer-created media takes a lot of time and energy — unless we develop economic models to meaningfully compensate the long tail, the ego payoff for most people won’t be enough to justify the effort. The cost of entry to create content is low in terms of dollars, but the cost of sustainable content creation is very high in terms of time, which in this short life is our most valuable commodity.

Open markets have a way of purging the excess — the long tail will shrink as people go back to their lives, making high quality content less of a commodity. Perhaps the bubble in media will deflate itself and, as happened with the last bubble, the best brands will survive and ultimately ascend.

David Sifry just alerted me that his numbers have in fact been scrubbed of splogs and other forms of spam, so his graph below represents only “human-created blogs.”

Sifry Blog Count

Well, I’m getting that bubbly feeling again. The 27 million blog question is whether Sifry’s graph will turn out to look like Matt’s graph at some point down the road.

It would be interesting to see stats on blogs that post daily vs. blogs that post at least weekly vs. blogs that appear to be abandonned.

Comments (18 Responses so far)

  1. journal over a period of time, and the absence of any financial reward for most of the writers. How is all this relevant to Yo Chicago ‘s business model, which depends on enlisting large numbers of unpaid volunteers to write about their neighborhoods and on topics they know well? We

  2. Today’s bit of wisdom comes from Scott Karp at “Publishing 2.0″: Consumer-created media takes a lot of time and energy — unless we develop economic models to meaningfully compensate the long tail, the ego payoff for most people won’t be enough to justify

  3. [IMG] Scott Karp / Publishing 2.0:Is the Long Tail a Lit Fuse?

  4. Scott,

    I really enjoyed your comments. Lots of folks read my post, but you’re the first to take a stab at deciphering it. I think your point about the time and energy of CGM is a good one. I tend to think that, at some point, the number of blogs (and podcast for that matter) is going to plateau – but the level at which they plateau will be much lower than the number of blogs at the peak of the trend.

    There are lots of unknowns about Sifry and Umbria’s numbers, so it’s hard to say anything definitive about the bolgosphere based on my graph, but I tend to think that we’ve nearing the Peak of Inflated Expectation and are heading quickly for the Trough of Dissolutionment.

    Not that the Blogosphere is going aware, but that it’s maturing, and the average active bloggers will be a little more serious and commited than ever before. I think people measure the length of the Long Tail before they jump in. If the Tail is too long, there’s little chance of being heard – and that’s really the point with CGM. When the Blogosphere was 8M, it was worth a shot, but at 27M why bother?

    Being a blogger who feels that interacting with the community is its own reward, I don’t feel this was, but I think the average Joe might. Starting a blog becomes easier everyday, but getting it read get harder.

    I think we’re at some sort of inflection point, but I’m not sure what that inflection point is. If nothing else, it will be interesting to watch.

    Thanks for posting Scott. I really enjoyed it.


  5. Matt,

    Thank you for taking the time to pull out your calculator and drill down into the numbers — I find this kind of disciplined analysis all too rare.

    I like your terms “Peak of Inflated Expectation” and “Trough of Dissolutionment” (or perhaps it’s the Trough of Disillusionment). I see potentially two tiers of bloggers — the handful who find a way to do it professionally (i.e. make a living at it) and then the second who find a way to rationalize the time either as a hobby or as an adjunct to their day job.

    And yes, it will be very interesting to watch how it all plays out.

  6. If you want to see the future of ‘blogs’, look at my space. Lots of people there ‘blog’ but they’re not creating any value, they’re mostly passing notes in class in electronic form. A funny link here, a joke there, a “I don’t know what to eat for breakfast here”.

    However, that’s not what matters. Blogs, since the very beginning, have been mostly amateur crap. However, even 1/10th of one percent of a million will still yield valuable content that can both provide some level of economic satisfaction for the author (either a higher profile for his day job, or actual renumeration). Even if the numbers have plateaued in some sense, there are still new people reaching the age of 20 or so every day, and there are new countries getting online faster.

    Perhaps the numbers do show that we’re approaching a peak in the ‘everyone has a blog’ age, but that still doesn’t mean that the personal publishing revolution is dead, because even with a 50% drop in new blogs, we’re still looking at millions of words of good writing being produced, cheaply and with low distro costs.

  7. Ted, I don’t think the publishing revolution is dead — quite the contrary. I think the shrinking of the “amateur crap” end of the tail (I might have chosen kinder words) will only increase the value of the “millions of words of good writing being produced.”

    And I don’t think My Space has anything to do with blogs — at least not blogs that see themselves as publishers. My Space is about socializing online.

  8. [...] Rather than take recent stats on the seemingly relentless growth of blogs at face value, Matt Galloway and then Scott Karp got out their slide rules. [...]

  9. Great post, but allow me to clarify on the Technorati numbers. The numbers I presented are NET numbers, in other words, we’ve already subtracted out all of the spam and spam blogs that we see. So, if I included all of the spam blogs that we’ve tracked and killed, the numbers would be much much higher. The State of the Blogosphere numbers are all about human-created blogs, not splogs or spings.

  10. “It would be interesting to see stats on blogs that post daily vs. blogs that post at least weekly vs. blogs that appear to be abandonned.”

    Exactly. If there’s no culling of the numbers, if every demo from the past five years is included as A BLOGGG!!!, then the curve is highly misleading. It might just as well be measuring the hype and marketing, and hence feeding on itself.

    Not the mention the confusion between people who are doing online personal diary (lots), and those who are trying for a substantial audience (much fewer).

  11. [...] Rather than take recent stats on the seemingly relentless growth of blogs at face value, Matt Galloway and then Scott Karp got out their slide rules. [...]

  12. Great exercise in reality-testing the press release. I had vaguely similar thoughts, but I slept through a lot of econ lectures. What I do know is that splogs are a noticeably annoying percentage of Technorati search results, which I find frustrating as a cranky user. I’ve been thinking I’d test some keywords on various such blogging utilities and see which is the most splog-free.

    If Sifry can detect and discount splogs, then surely he can exclude them from search results?

  13. Participatory Media and the Incentives Problem…

    Why do people blog? Questions about incentives are sure to arise soon after people begin talking about participatory media (or “user-generated content” as the business people call it). Yahoo! Research Berkeley has a whole team, led by Came…

  14. [...] justify the effort. Originally posted by rybesh from del.icio.us/rybesh, remediated by yatta on Feb 11, 2006 at 09:53 PM Comments Excerpt: Post acomment [...]

  15. [...] In blog entry, they show this graph which according to them indicates that number of blogs on the internet has started to decrease. tecnorati recently reported that there are 27 million blogs, and that the number of blogs is growing. however, what these guys seem to be saying is that more then half of new blogs created are spam, and that when you actually compare the growth of spam compared to growh in blogs, they actually expect decrease in the number of blogs. they are talking here about blogs that are active after three months. [...]

  16. [...] Zwei Blogger, Matt Galloway und Scott Karp, haben sich die Technorati-Zahlen mal genauer angeschaut und mit einer anderen Studie zum Wachstum von Spam-Blogs (splogs) gegengerechnet. [...]

  17. [...] Scott Karp’s recent entry “Is the Long Tail a Lit Fuse?” also raised the question for me of just who bloggers are. Karp writes, and correctly so, that while starting a blog is easy and cheap, keeping it going is not. This I’ve learned in my attempt to keep this thing going. He suggests that unless we find economic models to compensate blogger’s time, the number of non-spam blogs will shrink. While I completely agree with him as far as this applies to blogs and bloggers like him, who put care and effort into what he publishes, I think, like The Wall Street Journal and The Washington Post (though to a lesser extent) Scott is ignoring the large section of bloggers who make up the meat of the blogging network. These people may occasionally post something with careful thought behind it, meant to have wide effect, but generally they post about their lunch last Tuesday, or that sweet party they went to last night. Perhaps instead of seeing a widespread decline in blogging, we will see a strengthening of these more relaxed blogs. [...]

  18. [...] When you don’t have a steady stream of returning traffic it means that the only way you get hits is with appealing content. And as Scott Karp pointed out this last week, that’s actually a really, really difficult thing to maintain. [...]

Add Your Comment


Receive new posts by email