That’s the mixed and unsettling picture emerging from an analysis comparing the traffic at the 30 most active newspaper sites between October, 2006, and October, 2007. (Complete data are in the tables below.)
The good news is that the number of unique visitors at the 30 largest sites rose 21.2% over the 12 months to 97.5 million in October of this year, according to data provided by Nielsen Online, the independent ratings agency.
The bad news is that the time visitors spent on the sites fell by 5.1% to an average of 12 minutes and 20 seconds per month, or a measly 24.67 seconds per day. The year-to-year decline was broad-based, too, with visiting time falling at 19 newspapers and rising at only 11 of them.
The decline of visiting time was more significant than the 5.1% average suggests. Gains of 293.3% and 164.8% in Phoenix and San Diego helped to statistically offset the double-digit percentage declines suffered by a dozen papers. If you eliminate Phoenix and San Diego from the reckoning, the industry suffered an average 13.7% decline in the time spent on its sites. In the worst of the cases, visiting time at the Miami Herald fell to an average of 3 minutes and 6 seconds this year from 6 minutes and 48 seconds in 2006.
The Teflon effect at nearly two-thirds of the top newspaper sites is puzzling in light of the energy most publishers in the last year have put into building traffic with such features as 24-hour news, video, blogs, podcasts, slide shows, interactive commentary and user-generated, hyper-local content.
With newspapers offering more goodies on their web sites than ever before, they should be seeing more stickiness, not less. So, what’s wrong?
One explanation could be that the proliferating alternatives are fragmenting the audience faster than newspapers can coalesce them. For further discussion on this pooint, see Paul Farhi's must-read piece here in the American Jounralism Review.
But another possibility is that surfers aren’t as riveted by dreary Associated Press videos of yammering politicians as they are by You Tube's cutesy clips of dancing Gumbys. Or, they may not be particularly intrigued to learn about stray terriers in Virginia or whether a certain Chicago-area reporter will serve Brussels sprouts on Thanksgiving.
Declining stickiness is the last thing the newspaper industry needs, since it already severely trails slews of online competitors who are vying for the same audience and advertiser dollars.
As discussed previously here, the average 12-plus minutes that visitors sampled newspaper sites in October is overwhelmed by the 1 hour, 14 minutes and 40 seconds that the average visitor spends in a month at the sites operated by the 10 largest web companies.
The impressive gain in the number of unique site visitors over the last year may have been a mixed blessing.
On one hand, it gave newspapers the opportunity to capture the loyalty of thousands, if not millions, of fresh visitors. But the weak content that evidently shortened the average length of visits at nearly two-thirds of the sites also could have turned off, potentially permanently, a substantial portion of the notoriously fickle online crowd.
If newspapers intend to compete in the world of interactive media, they are going to have to do a ton of market research, product development and soul-searching to create sufficiently compelling sites to sustain the interest of the crowds they have been skillful, or lucky, enough to draw.
They have no time to lose. The broad decline in reader engagement in the last year makes clear that newspaper sites need more industrial-strength staying power than they pack today.
Excellent analysis, Alan. A couple of thoughts:
ReplyDeletea) Single visits at newspaper sites are skewed because so much traffic comes in from search and sites like Drudge and blogs--newspaper sites are sort of the ultimate referring destination. Thus, somebody visits just for a particular story, and then leaves, uninterested in anything else on the site.
b) Then again, newspaper sites generally haven't kept up with the state of the art in Web development. As I recently wrote in my blog (http://recoveringjournalist.typepad.com/recovering_journalist/2007/11/digging-an-even.html), they're pretty stodgy. They're barely using things like video, Ajax, social tools and other things that keep people around other sites. Their search engines generally are lousy. And their navigation tends to be extremely hierarchal--once you get down into the silo of a story, there's often little incentive to stay to read anything else. Horizontal navigation, which is a great source of stickiness, is often absent. Newspaper sites need to get sharper about using the newer Web tools and about deliberately creating reasons for visitors to stay on the site.
We don't need "a ton" of market research and soul searching. (Oh please God not more soul searching in the newspaper industry.)
ReplyDeleteHere's the deal: teenagers and college kids have managed to launch several hugely trafficked sites (digg, Slashdot, myspace, facebook, reddit) with massive stickiness and zero market research, soul searching or even original content.
How do you think they do that?
Well, for starters, they keep their sites simple, unlike the screen vomit that is the typical American newspaper site front page.
It helps that they don't have ten corporate digital VPs signing away the digital real estate in empty "partnership" deals.
They don't auction off story text to the highest bidder for spam links.
They don't pop up ad windows.
They don't demand to know your zip code and in what year you were born just to read the site.
And, hey, get this: their text has LINKS! Tons of links to relevant content inside and outside the site. Not made by a robot link generation system. Not tied to some random bizdev directory thing.
No, they have links made by actual humans who judge what sorts of links to what sorts of content might be logical and useful to a reader. This remains pathetically rare on newspaper sites, because it is so much cheaper just to dump the story contents into a CMS. Oh sure, Web editors say they do a quick scan for where they might add links, but all too often the end result tells a very different story.
For goodness sake, the last thing the industry needs right now is more process. More consultants and focus groups to tell us that what we need is dancing Gumbys and Britney Spears podcasts and dayparting and oh by the way ask your bloggers to use the word "meh" more, it's huge with the kids.
Alan, you knock dreary video of politicians, but sites like daily kos, talking points memo and instapundit have built large and, yes, sticky audiences based on this supposedly boring content.
They've done this, and done it on the cheap, because they care, because they have a human voice, because they have a human site -- not designed by a committee beholden to a marketing VP and 10 other people, bursting with awkward ads, inappropriate links and a million other reasons to click Back.
What is needed is a simpler, more honest exposition of the news.
It is that simple and that hard. Simpler, more honest. Simpler, more honest. Iterate your site over and over again until along those lines until it becomes the very essence of the topic it covers.
Personally, I skim the large papers for headlines but I never read the stories. You know what they are going to say before you read the article. Mainstream news is rarely new. Its recycled and unoriginal.
ReplyDeleteI've heard it before.