Skip to content


How aggregators help newspapers (now with real numbers!)

I couldn’t help but notice a tweet last night from Steve Yelvington: “Is it still legal to link to this asinine crap?”

Yes, it is asinine crap, and there’s no point dwelling on it. Instead, let’s look at are actual statistics from a real, live newspaper about how many of our visits come from aggregators. All stats (and screenshots) are taken from Google Analytics data for ColumbiaMissourian.com and VoxMagazine.com.

All Web stats are taken from the past six months — starting Jan. 1, 2009 and running through June 28, 2009.

First off, let’s establish how many visitors the Missourian has had over that time period. As Google says, “The number of visits your site receives is the most basic measure of how effectively you promote your site.” Since January 1, we’ve had 1,658,484 visits, or 9,265.27 visits/day.

By far our largest day in terms of visits was April 13, with 22,168 (1.34% of the total). April 12 was also high, but more on that later. Here’s the visual:

Visitors

The next step is to look at where our visitors are coming from. That’s easy as well — just go to “traffic sources.” Now this is interesting: only 24.26 percent of our visits are from “direct” sources (i.e., someone typing “ColumbiaMissourian.com” into a browser’s address bar). Of course, if you flip that around, it means that 75.74 percent of our visits are from either search or referring sites.

Here’s the visual:

Traffic sources

Not exactly something that we want to see die, eh? Maybe those aggregators aren’t so bad.

Now let’s look at the April 12-13 data. Wonder where all those visits came from? Looks like Fark.com, for whatever reason, linked to this AP story. Good thing our servers survived. Here’s the visual:

April 12-13

The moral of the story? Aggregators and referring sites aren’t bad. In fact, they probably drive most traffic at most sites. I’d love to get some data from some other newspapers to confirm or deny this …

Posted in Missourian, Newspapers, Web site.

Tagged with , , .


5 Responses

Stay in touch with the conversation, subscribe to the RSS feed for comments on this post.

  1. Josh Young (@jny2) says

    Just because aggregators deliver you X visits/day doesn’t mean that in a world without aggregators you wouldn’t receive X+N visits/day, where N is positive. The logic fail is ignoring relevant possible alternatives, and the practical point is that though aggregators may legitimately send news sites loads traffic, they may also siphon off a great deal as well.

    See http://en.wikipedia.org/wiki/Counterfactual_conditional.

    [Rob's note: edited to fix the Web link]

  2. Shafqat says

    Hey Rob – great post and well done backing up with data. You mentioned you’d love to see other data. I did a quick and dirty post with numbers from Hitwise UK (they kindly dug around a bit for me and produce the stats). On average, UK newspapers received 32% of their traffic from Google, with a further 5% from Yahoo and Facebook. It’s amazing that near numbers match up almost perfectly to your internal statistics. You can read the rest of my tongue in cheek post here: http://blog.newscred.com/?p=182

    Anyway, we need more of these kinds of posts. Once people like Ms. Schultz see these numbers repeated over and over again, she’ll realize how misguided her suggestions really are.

  3. The Truth says

    @Josh Young You’re absolutely right on this. It amazes me how many people miss the logical flaw in the aggregator argument. It is entirely possible that if those referring sites did not exist, that those users might simply go to your site instead. And simply going back in time to look at historical traffic before aggregators would not work as more users have moved online, consume more content, and benefit from better viewing experiences on news sites than they did in the past.

  4. Stephen Larson says

    Copyright law doesn’t need to change, newspaper need a way to enforce it with minimal expense.

    Your study doesn’t address what traffic would be at websites that actually generate real content if the aggregators were stopped with existing copyright law. With no aggregators, readers would have fewer websites to go to for news and likely increase traffic.

Continuing the Discussion

  1. Some responses to responses on aggregation – New media at old Mizzou linked to this post on June 30, 2009

    [...] I don’t have comment threading (lame, I know) I decided to respond to the comments about aggregation [...]