Who’s on top? News Web sites

Average number of pages viewed per visitor per month
(Total page views divided by total unique vistors)

38 – Atlanta Journal-Constitution
35 – NYTimes.com
28 – The Houston Chronicle
19 – washingtonpost.com
19 – USATODAY.com
18 – Star Tribune
18 – Boston.com
18 – Philly.com
16 – SFGate.com
15 – Chicago Tribune

Average number of seconds spent per visitor per month
(min:sec per visitor per day)

2229 (1:20/day) NYTimes.com
1889 (1:07/day) Atlanta Journal-Constitution
1412 (0:50/day) Star Tribune
1328 (0:47/day) USATODAY.com
1256 (0:45/day) Boston.com
1244 (0:44/day) The Houston Chronicle
1228 (0:44/day) washingtonpost.com
1052 (0:38/day) Philly.com
950 (0:34/day) Wall Street Journal Online
896 (0:32/day) SFGate.com

These lists do not show the way Nielsen represents the data. But they should make us think. Do we care how many people pop in per month or per day? Or do we care how long they stay, and what they stay for?

Original data from February 2007 reported in E&P and attributed to Nielsen//NetRatings.

Download my Excel spreadsheet (31 KB). Yeah, I took an IRE workshop once.

Technorati tags: | | |

4 Comments on “Who’s on top? News Web sites

  1. I saw those numbers attached to a ‘top 30’ news web sites list elsewhere and was also struck by the bare minimum of time people were spending on many sites.

    Not sure it matters what number we show advertisers – they seem attached to PV/CPM at the moment. But, as content creators and information/page designers – we should be seriously concerned if our readers are only spending 30 seconds a day with us.

    Even as we build PVs and uniques – having an average time spent that low seems like a dangerous leading indicator of failure.

  2. Given the way the inflatability of page view numbers, I’m not sure the first metric has too much value these days.

    I still like the time-spent-on-site measurement as a way to check how sticky news pages are, overall.

    It’s telling to see the NYT high not just in page views, but in time spent. Site design is a factor, but there’s also the all-important fact that NYT reporters write stories I want to read all the way through to the kicker quote.

  3. I agree, Ryan, the quality of the NYT writing hits home again and again for me. But I realized that I probably open fewer than 35 pages at nytimes.com in an average month … then again, I don’t live in New York.

    I also agree that page-view inflation makes this number close to meaningless. But that’s even more reason for concern: It means the real number of pages viewed is much lower than these.

  4. I’m looking forward to the day when the page view/CPM model is ditched (I’m going to be waiting a long, long time …)

    The time-spent vs. page view argument is running headlong into Web 2.0 apps (disclosure: I recently wrote about this topic on Poynter’s E-Media Tidbits), yet the latest apps don’t work with the current revenue model because they don’t generate copious page views.

    Perhaps we’re heading toward a broadcast variation where time-spent becomes an important metric. I’m certainly not an advertising genius, but it seems to me that the longer someone stays on a page the better the chance they’ll see an ad or brand.

    Somewhat related question: Anyone know which traffic services/software packages the major news sites use?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.