Thursday, August 16, 2007

Flat-lining online

Contrary to the chipper spin from publishers that online readership is growing robustly, the traffic at most newspaper websites has been flat – or worse – for the last year, according to a new report from Harvard University.

In a study of 160 newspapers, the Joan Shorenstein Center on the Press, Politics and Public Policy found that the “record” readership gains heralded by the industry were produced almost entirely by the New York Times, Washington Post and USA Today, where traffic on average rose “more than 10%” in the last year.

Although the Newspaper Association of America reported as recently as last month that a “record” 59 million unique individuals visited newspaper web sites in the second quarter of this year, the Harvard researchers reported that almost all the growth was generated by the high-volume sites operated by the name-brand national papers.

“The overall traffic level…hides important differences within the newspaper sector,” says the study. While the leading newspapers have been gaining traffic, “the websites of most other newspapers — whether in large, medium-sized, or small cities —have lost audience and their sites on average have substantially fewer visitors now than a year ago.”

Traffic fell in most mid-size markets in the period from April, 2006, to April, 2007, according to the study. “The sites of small-city dailies also are not growing,” said Thomas E. Patterson, the author of the study. “Although two of the nine [small-city] sites we sampled had a traffic increase of 20% or more from the previous April, five sites suffered a decline, including one that lost 20% of its web audience.”

UPDATE (8/16/07): The study provoked a sharp resposne from Howard Owens, director of digital media at Gatehouse Media, who called the report "bunk" in the Comments section below. Howard says traffic has improved at newspaper web sites since April, the month the study concluded. Statistics below for the 25 largest newspaper sites show that 14 gained and 11 lost traffic between April and July, yielding an over-all average increase in unique visitors of 6.3%. The gnarly challenge of accurately measuring site data is discussed here by Mark Potts.

ADD UPDATE (8/17/07): The NAA's "own data indicates that web traffic to newspapers has been more or less flat for two years," says Harvard researcher Thomas Patterson in his full Comments below. "With more than a thousand newspaper web sites, it is possible to cite other newspapers that are gaining traffic. But we were not interested in selecting data to fit a thesis.... I wish it were different.... Some in the business can put a spin on the trends, but, if others believe them, they will end up delaying the adjustments necessary if local papers are to thrive in the Internet age."

The performance of newspaper sites in the Harvard report paled in comparison with those of the broadcasters tracked in the yearlong study, to wit:

:: Unique web visitors gained an average of 40% at local TV websites and rose 35% for the national broadcast and cable networks. CNN bucked the trend, falling by 3.2 million visitors in the 12-month period.

:: Traffic at local commercial radio stations grew 23% in large markets and 14% in small ones. By contrast, unique visitors tumbled 20% at National Public Radio’s site.

In fairness, newspaper websites absolutely have more readers than broadcasters, because they have been publishing online much longer – and more effectively – than most broadcasters. Because papers have larger bases of site visitors, they have to add far more new readers than lower-traffic broadcast sites to achieve meaningful percentage gains in readership.

With that said, however, faltering traffic is still faltering traffic.

If it is not turned around, then newspapers can’t hope to maintain the double-digit increases in online advertising sales that have provided the sliver of silver lining in their otherwise gloomy earnings reports for the last five quarters.

As discussed here, newspapers and other mainstream media for the most part have tin ears when it comes to developing content that appeals to online consumers. They need fix that. Stat.

5 Comments:

Anonymous Anonymous said...

Thanks for that. somehow I totaly managed to miss it.

11:39 AM  
Anonymous Anonymous said...

I have direct access to audience metrics for several small newspaper sites, and I talk enough with other site managers and small- and mid-sized to know that the Harvard report is bunk.

Traffic is up.

It flat-lined for about a year, but a few months ago, started climbing again.

BTW: I'm not endorsing the NAA reports. I have my own questions about those. I'm just saying whomever was involved in the Harvard story was smoking beans.

3:25 PM  
Anonymous Anonymous said...

The Newspaper Association of America's own data indicates that web traffic to newspapers has been more or less flat for two years. They can't all be gaining audience if the industry as a whole has a flat traffic pattern.

And who are the big winners? Our data, and that of every other data service of which I am aware, indicate that a few sites, such as nytimes.com, are getting the lion's share.

Sure, with more than a thousand newspaper web sites, it is possible to cite other newspapers that are gaining traffic. But we were not interested in selecting data to fit a thesis. We selected the newspapers first, and then investigated their traffic.

I wish it were different. Local media are a bedrook of community democracy. Newspapers across the country are hurting--losing circulation and laying off journalists. Some in the business can put a spin on the trends, but, if others believe them, they will end up delaying the adjustments necessary if local papers are to thrive in the Internet age.

In any event, time will tell who's analysis is correct. And that was the point of the research piece, as a careful reader would have discerned. We carefully described our methodology and encouraged others to follow up with research of their own, now and in the future. The news business is changing, and we need to know in what ways it's changing, and what might be done in response.

10:17 AM  
Anonymous Anonymous said...

I'm puzzled by the figures in the table. What exactly is the unit being reported?

I see the title of the table is "Unique Visitors" and the data columns are for specific months, but is that TOTAL unique visitors over each ~30 day period or is it a DAILY AVERAGE of unique visitors?

Inquiring minds want to know....

5:20 AM  
Blogger Newsosaur said...

The chart above measures the total number of unique visitors to a site in each month.

If you visit a site once during month, you count as one unique visitor. If you visit the site every day in a month, you still count as only one unique visitor.

9:14 AM  

Post a Comment

<< Home