The economics of aggregation

By Felix Salmon
March 16, 2010
economics of aggregation post? Here's the question he asks:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

How geektastic is Mark Thoma’s economics of aggregation post? Here’s the question he asks:

If you run a website that depends upon advertising, what is the optimal number of aggregator sites (sites that run part of your original posts plus a link back to the original)? What is the optimal length of an excerpt?

And here’s his answer to the first part:

RsN[sC(CrN(rNA + rAN))N + sNA] = -RsNsAN

In English, he concludes that aggregators are good for original content providers if they provide a lot of clickthroughs, and that longer excerpts can also be good for original content providers “if increasing the excerpt length has little detrimental effect on the number of clickthroughs”.

This is all jolly good fun, if not particularly useful in the real world, but I fear that Mark, here, is missing something very important. The holy grail for content websites is loyal readers — people who come back to your site multiple times per week or even per day. Most visitors to a site, whether they come from Google or from social-media links or from aggregators, will simply read the article they came to read, and then go away. A small proportion of those visitors, however, will stick around for a while, look at the rest of the site, and then, possibly, decide to keep on coming back.

I’m not a great browser of websites, but I certainly do something similar: if I follow a link and find myself on an interesting new blog, I’ll subscribe to that blog, and then might well come across more great content there while looking at the stories in my feedreader. Alternatively, I won’t subscribe to a blog the first time I visit, but once I find myself visiting the same site three or four different times, I’ll get the picture and click on the RSS icon. (Of course, if the RSS feed turns out to be truncated, I’ll probably delete it immediately, but that’s another issue.)

So it seems to me that Mark’s model overemphasizes the importance of what you might call drive-by traffic, while underemphasizing the importance of building a loyal repeat audience. There are millions of potential loyal readers out there, and the most difficult thing to do online is to reach out and touch them, somehow, just so that you get onto their radar screen so there’s a chance they might become actual loyal readers. Aggregators are a wonderful way of doing that: they make it their job to find people posting original content, and to show it to lots of potential loyal readers. I think of them as free marketing.

People like Rupert Murdoch, and his former employee Heidi Moore, think that what the aggregators are doing is stealing content and making money off it. But I like it when people make money from reading my blog! As far as I’m concerned, that’s a good thing. And Murdoch should similarly be very happy that the dynamics of the internet mean that there’s a strong incentive for people to read his websites assiduously, and then link back to hundreds of his stories, driving traffic and helping to build that crucial base of loyal readers. Historically, Murdoch got his readers by spending hundreds of millions of dollars printing physical newspapers and distributing them all over town, this is a much easier and much cheaper way of getting his brands into the consciousness of consumers.

So I don’t worry when people read my stuff on an aggregator and don’t click through to my website — and I wouldn’t worry about that even if I was supported solely by advertising. (Which, incidentally, is actually a very uncommon blog business model: most bloggers, insofar as they monetize, will do so through job offers, book deals, or even blog acquisition deals, rather than ad revenues.) If you make it easy for people to find you and to read your stuff, they will eventually become hugely valuable to you. If you make it hard, they will be worth nothing to you. The choice is easy.

Update: Heidi responds, in the comments, saying that “honest aggregators” are those who provide “only a brief summary”:

It is true that I am a former employee of the Wall Street Journal, which is owned by Rupert Murdoch, but it’s misleading to say I was an employee of Rupert Murdoch, is it not? And I did, after all, decide to leave the Journal. I have nothing against Rupert Murdoch, as he was nothing but polite the only time I met him, but for the sake of accuracy that should be clarified.

Similarly, I’m sad to say my views here have been misquoted. Unlike Rupert Murdoch, I do not see Google as a problem for journalism – I see Google as a boon to driving traffic and opening journalism up to more Web-surfing readers.

I believe, Felix, you are referring to a Twitter discussion we had about plagiarism, which is quite a different topic. In that conversation I said that SOME blogs, in the name of aggregation, had resorted to cutting-and-pasting the content of news sites or other blogs, adding no new analysis or reporting of their own, and distributed the content with a new headline as if it were their own work.

I made the point that this practice, if it were copied in print, would be considered plagiarism – and that it should be defined as plagiarism on the Web because it steals traffic by taking the most important parts of a story and corraling elsewhere the traffic due to the author and original publication.

To be clear, this has nothing to do with linking, or analyzing an article written elsewhere, which is the currency of the Web and its honest marketplace of ideas. My point was purely about the practice of re-headlining the work of others and then copying that work on such a scale that it would divert readers from clicking on the original. This is, I believe, plagiarism that can often be hidden in the name of “aggregation.”

It means nothing for honest aggregators, however, who provide readers with only a brief summary and direct credit and link to the original.

More From Felix Salmon
Post Felix
The Piketty pessimist
The most expensive lottery ticket in the world
The problems of HFT, Joe Stiglitz edition
Private equity math, Nuveen edition
Five explanations for Greece’s bond yield
Comments
12 comments so far

Felix–Back in the early days of the economic downturn, I first began looking for blogs that seemed to have good insights, ideas, analysis, and commentary on what was going on. I found aggregators a great way to do that.

Over time, I would find myself clicking through to some of the same sites repeatedly. These I bookmarked–including your site here.

Just a long way of saying, if you don’t drive by a few times, you’re never going to stop.

Posted by Lilguy | Report as abusive

From the point of view of an advertiser I would say loyal readers are a pest. Let’s say I’m trying to sell some wares by advertising on your site. If your site receives 100 visits all by the same guy and he’s not interested, then that’s it. If it’s 100 different people at least there’s a chance one of them will be interested.

The solution to the loyal reader problem is to show them targeted ads. But that means violating their privacy, reading their cookies or whatever to see what they buy online. I suspect some sites do that because the ads are uncannily relevant to me. Once I was alerted this way that a website I had previously bought from was doing some sales. That was pretty useful.

Posted by EmilianoZ | Report as abusive

There are several unproven statements in this article that render it pointless, the most egregious is the idea that the “Holy Grail” is repeat customers.

While it is good of people come back MoM growth is usually more highly valued than have a small number of repeat visitors, so holy grail is at least hyperbole if not just wrong.

The other dubious point is about personalization, what most publishers have known for a long time is that if a person is reading the article then regardless of who they are they might be interesting in what you are selling.

The AI based algorithms are great VC fodder and have made some people rich but in structural breaks like this one lots of people make money on ideas that are “new” and unproven, most go on to lose investors money as it turns out that knowing the color of the readers hair is not that useful when selling most things.

Posted by jstaf | Report as abusive

It is true that I am a former employee of the Wall Street Journal, which is owned by Rupert Murdoch, but it’s misleading to say I was an employee of Rupert Murdoch, is it not? And I did, after all, decide to leave the Journal. I have nothing against Rupert Murdoch, as he was nothing but polite the only time I met him, but for the sake of accuracy that should be clarified.

Similarly, I’m sad to say my views here have been misquoted. Unlike Rupert Murdoch, I do not see Google as a problem for journalism – I see Google as a boon to driving traffic and opening journalism up to more Web-surfing readers.

I believe, Felix, you are referring to a Twitter discussion we had about plagiarism, which is quite a different topic. In that conversation I said that SOME blogs, in the name of aggregation, had resorted to cutting-and-pasting the content of news sites or other blogs, adding no new analysis or reporting of their own, and distributed the content with a new headline as if it were their own work.

I made the point that this practice, if it were copied in print, would be considered plagiarism – and that it should be defined as plagiarism on the Web because it steals traffic by taking the most important parts of a story and corraling elsewhere the traffic due to the author and original publication.

To be clear, this has nothing to do with linking, or analyzing an article written elsewhere, which is the currency of the Web and its honest marketplace of ideas. My point was purely about the practice of re-headlining the work of others and then copying that work on such a scale that it would divert readers from clicking on the original. This is, I believe, plagiarism that can often be hidden in the name of “aggregation.”

It means nothing for honest aggregators, however, who provide readers with only a brief summary and direct credit and link to the original.

I just thought it was important to clarify. Carry on.

Posted by HeidiNMoore | Report as abusive

This is perhaps the single greatest piece of financial journalism ever written.

Posted by Mindbomber | Report as abusive

I like Digg, especially.

Posted by IhateStoryburn | Report as abusive

Interesting thoughts here. You’re right in that aggregators are great tools for new traffic & clickthrough, but they are even more valuable if they bring readers that eventually transition to repeat visitors.

I think the main thing that content generators hate is syndicators and/or plagiarizers as Heidi had pointed out. If a syndicated article is on another site, readers have little reason to click to go to the original since they just read the article in its entirety. Aggregators that provide a ‘tease’ of a paragraph or two and then allow the reader to clickthrough to the original are often the best source of traffic for content generators.

Lastly, I felt the need to challenge some of your latter comments. You say that you wouldn’t worry about people not clicking through to your original website if you were on a solely advertising based model. With all due respect, I think you’re in a bit of a different situation than most bloggers, no? I’m making assumptions here so forgive me if they are wrong, but I would assume that you’re working on a salary. While your pay may be somewhat based on traffic incentives, you work for a major media company that already has an established traffic base.

You differ from the typical blogger by way of platform & compensation/business model. Most bloggers I would argue are highly dependent on traffic and absolutely need that clickthrough from aggregators. If they don’t get any, they don’t make money (through advertising or upselling to premium content). This is coming from a blogger that relies primarily on advertising for revenue. I would argue the complete opposite of your point in that I absolutely would need all traffic I can get because a) I’m not on a mainstream platform and b) I’m not on a salary. Any money made on the site is a direct result of advertising (which is reliant on traffic).

I think it’s a situational thing. This would assume that I am blogging full-time though, which I am not. So, that throws a wrench into the equation as many authors in the financial blogosphere have dayjobs as well. So, maybe revenue from their blog isn’t the priority to begin with and they aren’t as concerned with maximizing clickthrough.

Look forward to hearing more of your thoughts on these types of topics Felix.

Jay
@marketfolly

Posted by marketfolly | Report as abusive

When the aggregation gets rough, bloggers get existential.

Posted by HBC | Report as abusive

Jay, I wasn’t talking about myself, of course I’m not supported 100% by advertising. And even you note that you’re not really blogging for the ad dollars. The point that Mark Thoma and I were making is that it’s worth questioning the assumptions you make when people like you and Heidi simply assert, without any empirical evidence, that aggregators with only a “tease” paragraph are better sources of clickthroughs and long-term traffic than aggregators which quote at length. You might be right, but then again, we don’t really know that for sure, and it’s possible that the opposite is true.

Felix, do aggregators really need “empirical evidence” that stealing gobs of other people’s hard work is a bad strategy? It’s not clear why you’re thinking in terms of the economics of the aggregator and ignoring the economics of the originator.

Posted by HeidiNMoore | Report as abusive

Felix: “economics of aggregation” – shouldn’t that read ‘logarithmic economies of scale” ?

Please next time explain the equation, you can’t have a “+” on the left and a “-’ on the right, you will have a Algebraic Fatwa hanging over your head in no time. Or maybe it’s a half full glass…

And quite frankly, I am not interested in your and Heidi’s politics.

Posted by Ghandiolfini | Report as abusive

“This is perhaps the single greatest piece of financial journalism ever written”

I keep coming back to this just to crack myself up.

Posted by Uncle_Billy | Report as abusive
Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/