Martin Moore

This post was first published at Media Standards Trust on 23rd March 2011

Since churnalism.com launched on 23^rd February a lot of people have asked us if they could ‘search the other way’. In other words, could they see news articles that appeared to be based on press releases rather than vice versa.

We now have enough press releases in the system to make that worthwhile. So today we’re launching an ‘Explore’ page. Here you can see all the news articles that appear to based on press releases loaded into our system (over 5,000 so far). You can filter by press release source – such as the government’s news distribution service, Sainsbury’s or Eurekalert (science releases). And, you can filter by news outlet (UK nationals, BBC news online and Sky).

Go to churnalism.com ‘Explore’ page

This reveals things like the:

• 24 BBC News articles since November 2010 that appear to be based on BBC press releases
• 221 MailOnline articles since March 2009 that appear to be based on OnePoll surveys
• 84 Guardian articles since early 2010 appear to be based on Central Office of Information releases
• 32 Telegraph articles since 2007 appear to be based on press releases from Marks and Spencers
• 31 Financial Times articles since January 2010 appear to be based on CBI press releases

Our system tracks back most national news articles for about three years. Press releases are, necessarily, still more sporadic since they rely on people uploading and on a select handful of popular press release sources.

We will be adding more press release sources as fast as we can. In the meantime, if you’d like to help you can:

• Tell us where we can find good sources of press releases online
• Help us write scrapers (if you’re feeling really techie). We’ve put up a short ‘how to’ guide here. If you have any issues please do drop us an email (team AT journalisted DOT com)

Any pages you find when exploring churnalism.com you can share with a shortened url via twitter or Facebook or elsewhere.

This post was first published at Media Standards Trust on 10th March 2011

Surveys are a good way to generate ‘news’. News outlets like to be seen to reflect public attitudes and concerns, and surveys appear to do this. Because of their news value, surveys are frequently commissioned by commercial organisations to promote their products and services. There is nothing wrong with commissioning such a survey, except that these surveys are then often presented in mainstream media as ‘independent news’, which they are not.

OnePoll conducts many such commercial surveys. OnePoll is certainly not the only organisation to use surveys as a news peg, Travelodge do as well for example, but this post focuses on them because it helps to illustrate why churnalism.com is performing a useful function in helping to make the practice more transparent.

Since launching churnalism.com we’ve been in an on-off dialogue with South West News Group (SWNS), the owner of One Poll surveys. SWNS have strongly objected to the press copy based on their OnePoll surveys being highlighted on churnalism.com (see SWNS post here). To summarise their objections:

• The OnePoll surveys are not written up as ‘press releases’ but as ‘news copy’ by professionally trained (quote) ‘news agency journalists’
• As such they are ‘factually accurate, rigorously checked news copy which needs little or no subbing’
• This news copy is then ‘distributed direct to national news desks via the SWNS newswire’
• Because it is written as ‘page-ready news copy’ it is designed to be published almost without alteration in the mainstream media.

It should first be noted that, based on this post, SWNS may have misunderstood the nature and purpose of churnalism.com. It is a tool that allows the public to compare bodies of text with news. Though we encourage people to paste in press releases, they can – and do – paste in lots of other text, such as news agency copy, news articles, Wikipedia entries, even text from plays and books. The site then finds news articles that appear to be closely based on this pasted text. It is not churnalism.com’s job to police what people choose to compare.

But there are a number of reasons why, now I know more about how the SWNS system works, I think it is even more important that churnalism.com raises public awareness about the use of this type of survey ‘news’ by the UK press:

• The OnePoll surveys that have been written up (and are listed on their site) are commissioned by commercial organisations in order to help promote a product or service. For example:
  • ‘People find true happiness by reaching the age of 50’ says a survey by Engage Mutual Assurance who have an ‘over 50s life cover’ insurance policy (20-1-11)
  • ‘The honeymoon period in marriage ends after just 14 months’ partly because partners put less care into personal hygiene, says a survey by Better Bathrooms (17-02-11)
  • ‘Eight out of ten singletons check out potential future partners online before they ever meet them’ says a survey by www.UKDating.com (5-01-11)
  • ‘Dedicated dog owners walk 23,739 miles during their pet’s lifetime’ according to a survey by esure pet insurance (11-02-11)
• A key reason commercial organisations commission OnePoll is because the surveys will be written up by journalists in the style of independent news
• They also know that, because the OnePoll write-ups are distributed via the SWNS ‘news wire’ mainstream news organisations are less likely to view them as a press releases, but rather as independent news (which they are not, at the very least they are ‘branded stories’ to use an SWNS phrase)
• Because mainstream news organisations do not view them as press releases, they are more likely to publish them virtually unchanged – as with Associated Press or Reuters news copy.

This is great for the commercial organisation which has commissioned the survey because:

• It has managed to promote its product or service without paying for advertising, but instead by generating news – according to Philips, ’72 Point [another part of the SWNS Group] is fast and efficient with a quick turn-around on ideas and news copy, making them our first choice when planning a survey campaign to generate national news coverage’
• Its product or service has featured in an ‘independent news’ article – therefore making the public more likely to read it and, the company hopes, believe it and take action as a result.

This is not so great from the perspective of the public which has been led to believe that that this is not a promotional article written to help sell a product or service but independent news. The public have been encouraged to think, for example:

• That 3 million women suffer from foot problems and, if they want to avoid such problems, then it might be a good idea to go and see comfort shoe specialist Hotter Shoes
• That one in three women would not dare to venture outside the house without make-up, and that make-up can significantly boost confidence, according to a survey from make-up seller Superdrug
• That women spend more on money on looks than they do on health, according to a survey by private healthcare provider Benenden.

Nor are OnePoll surveys occasional items of ‘news’ that only come out once a week or once a month. In the six months from September 2010 to February 2011 there were 176 (as recorded on the OnePoll press release archive). This is over 29 each month, or almost one-a-day. Many of the write-ups of these surveys are being republished almost unchanged in the UK press (as shown by churnalism.com).

Clearly the responsibility for making this more transparent to the public does not fall entirely on SWNS or similar companies that conduct and write up such surveys. SWNS has, at least, engaged in dialogue on this and sought to make the process by which OnePoll surveys make it into the press clearer.

Much of the onus is on the mainstream media to:

• Be as sceptical about OnePoll ‘news copy’ surveys and other surveys as they are about press releases from commercial organisations
• Do some of their own journalism on these surveys so, even if they do publish them, they at least add some context and scepticism to the claims
• Make it very clear to the public that these are ‘sponsored news’ stories by labelling them as such.

I hope the press will do this but I’m not holding my breath. Which is why churnalism.com needs to keep highlighting the use of OnePoll news copy and similar such copy from other organisations by the UK press.

This post was first published on the PBS MediaShift Idea Lab on 9th March 2011

Two weeks in, and the public response to Churnalism.com has been fantastic.

Since we launched the site on February 23, we have had 50,000 unique visitors, over 330,000 page impressions, and hundreds of press releases pasted in and saved. According to Google Analytics the site has been visited by people in 134 countries.

People have tracked down churnalism about eye-catching new products (such as “Baby Gaga,” ice cream made with breast milk), about new research findings from universities (for example, on the “protective properties of green tea”), about new police initiatives (e.g., the recruitment of teenagers by police to prevent cyber-bullying), about the “happiest time of the week” (7:26 pm on a Saturday, says a poll sponsored by a multivitamin company), and about the prose of Jane Austen (which might not be all hers after all, according to an Oxford study). People have pointed us to stores of press releases like www.eurekalert.org and www.alphagalileo.org so we can build up a bigger bank of comparisons. And there have been discussions about what might constitute “signals of churnalism.”

As importantly for us, the site has sparked lots of debate about churnalism. Here are some of the top questions that have come up:

Do the public care if journalists are churning out press releases?

Some felt the site’s exposure of churnalism would not much bother the public.

Mark Stringer of Pretty Green told PR week he was “not sure why anyone would want to go to the time and effort of producing a website to prove something that no one really cares about.”

Others thought the opposite was true.

“If you tell someone who is a punter rather than a journo that it’s pretty standard practice to ctrl+C and ctrl+V huge chunks of a press release into a story,” Steven Baxter wrote in his New Statesman blog, you’ll get a revealing reaction. “I call it the ‘Really?’ face. People look at you as if to say ‘Really? Is that what you do?’”

Our own experience to date appears to support Baxter’s view rather than Stringer’s.

Does the re-use of wire copy count as ‘churnalism’?

There has been a fascinating discussion about the re-use of wire copy, especially when it is re-used almost verbatim, often with a byline from the news outlet added.

People have pointed out that news outlets subscribe to wire services to broaden their access to news, so why shouldn’t they publish it?

Others have countered that using wire copy is not the problem, but passing it off as your own is.

“If you have to churn,”Minority Thought blogged, “at least be honest about it.”

How can news organizations make their use of press releases more transparent?

On Memeburn, Tom Foremski wrote about a suggestion he made a few years back to color-code text that came from a press release. For example, distinguishing text “copied from a release or outside source (red)” from original text in black — and potentially other colors to represent separate conflicts of interest. Others suggested just noting or linking to the release.

Professor George Brock, head of journalism at City University London, worried that rather than push journalists towards footnoting sources, Churnalism.com might discourage them.

Will Churnalism.com help reduce the production line approach to press releases?

A prominent communications professional, Mark Borkowski, welcomed the site, hoping it might help kill off the mass production of poor press releases.

So many are now produced, Borkowski wrote, that “the level of noise makes it hard for the true craft of the publicist to flourish.”

Is all churnalism bad?

Alan Twigg of Seventy Seven PR told PR Week that “this site is making it sound like [public relations officers] getting coverage is a doddle and that PROs are taking over the media. If only it was that easy.” Sounding a similar note, Stuart Skinner of PHA Media took to PR’s defense on the same website, saying that “news is not a product of collusion between shady PROs and lazy journalists.”

It is worth noting that the site does not say churnalism is easy, nor indeed that the reproduction of parts of press releases is necessarily unsavory.

“Of course not all churnalism is bad,” the site’s FAQ section says. “Some press releases are clearly in the public interest (medical breakthroughs, government announcements, school closures and so on). But even in these cases, it is better that people should know what press release the article is based on than for the source of the article to remain hidden.”

Richard Sambrook also made an important point in his blog, that “there is of course Good PR and Bad PR just as there is Good Journalism and Bad Journalism.”

Does Churnalism.com illustrate the self-correcting power of the web?

In the Guardian’s online comment section Jimmy Wales, co-founder of Wikipedia, suggested that plagipedia and Churnalism.com “show us that the Internet is perfectly capable of correcting its own follies.”

What’s an equivalent word for “churnalism” in Spanish?

Great question. 1001Medios began a Twitter-hunt for a word in Spanish that captured the idea of “churnalism.” Sadly, my Spanish is not good enough to work out if they’ve found one yet.

Building Buzz Without Legacy Media

The tremendous public response and debate almost certainly would not have happened without social media, blogs, and Chris Atkins. Chris’ news stunts — particularly about the chastity garter, the penazzle and Larry (or Jo) the cat — captured public attention at the same time as making a serious point about how churn makes it into the mainstream media. (You can see Chris’ film describing the stunts on the Guardian website, and his blog about it here.)

They also helped kick-start discussion about churnalism on social media, notably Twitter and Facebook. Thousands of people have tweeted about the “churnalism” problem, about Churnalism.com as a way to address the problem, about evidence of churn they have found, and yes, about Larry the Cat and the penazzle. It has been humbling and somewhat overwhelming to observe the level of public response and engagement.

Indeed, without social media and blogs there is every chance the site might have gone virtually unnoticed. The Guardian, which published the original “reveal” article about the news stunts, is still the only UK national newspaper site to have mentioned Churnalism.com.

Major news outlets that were fooled by Chris’ PR stunts have yet to acknowledge their mistakes — much less the website the hoaxes were intended to publicize. The BBC’s Radio 5 Live is — as far as we know — yet to tell its listeners that the “Jo the Cat” story, which they discussed at length on their lunchtime program, was a fabrication. The Daily Mail does not appear to have informed its readers that Margaret Sutcliffe is not pursuing her custody claim about the Prime Minister’s cat.

Contrast this with BBC Norfolk which immediately put its hands up and then used the hoax as a good way to start a discussion about churnalism.

Industry and International Attention

The public relations industry in the U.K. has been more direct in its response than the mainstream press. “PR Industry hits out at churnalism site” said an article on PRWeek.co.uk.

Various figures from the industry voiced their concern about the impact the site might have on the reputation of PR. Though in a measured and sensible leader, the editor Danny Rogers suggested churnalism was a genuine threat to both journalism and PR: “If organizations are churning out rubbish, and so-called journalists are mere accomplices in this process, we will all be taking part in a depressing downward spiral.”

One of the really encouraging things about the response to the site in its first two weeks has been the international reaction. In addition to many kind words of encouragement, we have had expressions of interest from people to extend the site to the U.S., Germany, Finland, Spain, and Australia. We’ve spoken to NPR radio in New York, to CBC radio in Canada, BBC Radio Norfolk, BBC Wales and to community radio in Essex. We’ve been contacted by news organizations in Germany, Belgium, Australia, the U.S. and Russia.

What’s Next for Churnalism.com?

Some of this interest is not in the site itself but in the technology that underlies it. The methodology we developed can be applied to many other uses beyond churnalism. It could be used, for example, to trace changes in the progress of legislation. It could be used to measure the re-use of Wikipedia. It could be applied to plagiarism in other parts of the web.

We’re still pedaling furiously to respond to many of the questions people have raised and issues identified. We are, for example, about to introduce a page that allows people to explore the use of press releases by news outlet or sector (i.e. government, science). We are now highlighting, on the home page, what comparisons people are sharing (since people seem to prefer to share than to rate). We are adding a report button so people can tell us when something definitely is not churn.

Finally, we will start to link the site more directly with the other Media Standards Trust transparency projects — notably journalisted.com and hNews. This should help us to create a whole toolbox of transparency and accountability mechanisms for online news and create an ecology that will foster and advantage original journalism.

This article was first published on www.CJR.org (Columbia Journalism Review) on March 3rd 2011.

The Media Standards Trust (U.K.) has just launched a website—churnalism.com—that lets people compare press releases with published news articles in order to help identify ‘churnalism’. Martin Moore, director of the MST, explains why they built it and how it works.

“Facebook ‘friends’ cause stress” (BBC, 2-16-11)

“New rations for Afghanistan troops” (Daily Express, 2-14-11)

“Men think putting out the bin is romantic says new survey” (Daily Mirror, 2-11-11)

“Home cooks help save traditional breakfast marmalade” (Daily Telegraph, 1-21-11)

“Golden age of happiness: Turning 50 is key to ‘content and comfortable’ life” (Daily Mail, 1-19-11)

“Immigrants’ family appeals costing taxpayers £1million a week” (Daily Telegraph, 1-2-10).

All these articles have something in common. They are all ‘churnalism’, the word made popular by Nick Davies in his seminal book Flat Earth News. A piece of ‘churnalism’ is a news article that is published as journalism, but is essentially a press release without much added.

Churnalism has been around a long while. Back in the 1920s Edward Bernays was writing about “The conscious and intelligent manipulation of the organized habits and opinions of the masses” as an “important element in democratic society.” In the 1950s Vance Packard warned us about “the large scale efforts being made, often with impressive success, to channel our unthinking habits, our purchasing decisions and our thought processes,” typically “beneath our level of awareness.”

But its power and extent have grown. In the U.S. and U.K. there are now more PR people than journalists. The PR industries in these two countries are numbers one and two in the world in terms of size. In the U.K., PR accounts for over £6.5 billion in revenues. PR is, in the words of Trevor Morris and Simon Goldsworthy, “faster growing, better paid and better resourced” than journalism. “Like it or loathe it, PR has become a key ingredient in many of our lives.”

There are now vast quantities of PR material produced every day, a good chunk of which makes it into the independent media. Research by Cardiff University, which Nick Davies used to inform his book, found that 54 percent of news articles in the U.K. press can be at least partially sourced to PR. And this is only the national press. Many local papers rely even more heavily on press releases.

This is a problem because, as Davies writes, “this material, whether or not it is truthful, is designed specifically to promote or suppress stories in order to serve the interests of political, commercial and other groups.” This promotion or suppression is more effective for public relations if it is disguised from public view. Nor do news outlets want to advertise their reliance on PR. Therefore the connection between the two normally remains hidden.

Promotion, advertising and distortion disguised as news

In many cases churnalism is about promotion of a service or product. The hotel chain Travelodge, for example, has an interest in promoting sleep. Sleep is what people generally do at Travelodge. The more Travelodge can get people to associate the chain with sleep, the more likely it is they will think of it when choosing a place to stay when they are travelling. To help people make this association, Travelodge often sends out press releases on the subject of sleep.

Here is a selection of Travelodge press releases: “Coldplay has the snooze factor – as it claims No.1 position in the UK Kip Charts” (10-22-10); “Over a third of British adults still sleep with a teddy bear” (8-16-10), “UK drained by 29 billion sleep debt” (5-27-10); “23 million Britons give big ‘Hola!’ to British siesta!” (6-6-07).

And here are some of the articles in the press that bear a distinct resemblance to these press releases: “Coldplay sends Britains to sleep” (Sunday Telegraph, 10-25-10); “Third of adults ‘still take teddy to bed’” (Daily Telegraph, 8-16-10); “Britons have lost almost an hour’s sleep a night during the recession, claims study” (Daily Telegraph, 5-27-10); “Wake up call on sleeping” (Daily Mirror, 6-3-10); “Britain Says ‘Si’ To Siesta Time At Work” (Daily Express, 6-7-07). Though these are essentially another form of promotion for Travelodge, like paid commercials or advertisements, the newspapers publish the press releases as news stories. The stories are not untrue, nor do they necessarily do people any harm. But they are manufactured, and are not what most of us would think of as “news.”

Yet the releases make headlines, and not just in the national press. In July 2010, The Scotsman and The Independent reported that “July is the grumpiest month,” attributing it to a Travelodge survey. The same story was published in the Hartlepool Mail, the Yorkshire Post, the Dundalk Democrat, the Kilkenny People, the Tipperary Star, and other local papers. Only a few years ago, on June 2006, Travelodge had sent out a press release reporting that June was the grumpiest month.

Travelodge is certainly not the only company to try to use the media to promote itself. All the major (and many minor) retailers use press releases and PR to promote their products and services. This is simply the flotsam and jetsam of the daily information cycle. But although those in professional communications and the press may be aware of this unspoken flow, it remains concealed from the wider public.

In some cases, lazy journalism is to blame. The press releases provide text, quotes, images, and sometimes even video footage. The story quite literally writes itself. Many news outlets ignore such puff, but many do not. As the Cardiff figures show, many get into the press without any sign that the journalist has even picked up the phone.

Serious churn

Not all churnalism comes from commercial sources. Much of it has political sources: public authorities trying to spin bad news, medical firms trying to obscure poor results, and political lobbying groups. For instance, Migration Watch UK lobbies against immigration to the U.K. At the beginning of January it sent out a press release reporting a “Massive Increase in Family Visitors Appeals” by migrants. The release stressed how expensive these appeals were to U.K. tax payers and how the problem urgently needed to be addressed. Articles based on the press release were published in three national newspapers. In his article for the Daily Express the following day, Macer Hall used 52 percent of the release (based on our analysis). Richard Edwards in the Telegraph used only slightly less, and James Slack in the Daily Mail used over a third.

Curious to see what was required to feed churnalism into the press, we helped set up an experiment. Chris Atkins, a film director with a track record of successful hoaxes (having directed Starsuckers and Taking Liberties), offered to explore how easy it was to get fake ‘news’ stories published. We agreed that the stories had to be entirely invented, harmless, and relatively easy to disprove. If any were successful, we would quickly make clear that the stories were made up.

It did not take Chris long before he had some success. He invented a product, the “chastity garter,” to be worn by women while their partners were away. Should the woman’s pulse rise above 120 BPM, and the moisture on her skin pass a particular level, the press release read, a text message would automatically be sent to her partner.

It would not take Woodward and Bernstein to see the flaws in this story. The press release even came with a wonderfully amateur photo of a garter with a cheap digital watch attached. Yet it proved irresistible to some in the press. The Daily Mail website (with readership of over 40 million unique users a month) published the story on its front page. For a while it was the most read story on the site. The Daily Star published a third of a page in the print paper and online. The story then went global, being published in the Times of India, CNET news, Express.de (Germany), Mako.il (Israel), Florida Today, and the Chicago Tribune, among others.

Of course, not all churnalism is bad. There are plenty of press releases that are in the public interest. It would be odd if news outlets did not publish news about medical breakthroughs, about major government announcements, about exciting new consumer products.

Moreover, huge numbers of press releases never make it into print. We never saw a news story about “The Valentines Day cucumber” shaped like a heart, for example (announced by Sainsbury’s in a press release in February).

But many do make it into print, and very few news outlets make the connection between the press release and the news article explicit to their readers. In the past, this lack of transparency was partly excusable given space constraints and given that newspapers never aspired to academic standards of sourcing. But now, given that many press releases are published online and are so easy to link to, any news outlet that wants to could easily link to a press release from the article.

By being more transparent about the sources of news, readers would be better able to judge where it comes from, whether it has an agenda, and whether it is just puff.

Frustrated by the current lack of transparency in the press, we (the Media Standards Trust) built churnalism.com. It is an independent, non-commercial site that lets people paste in press releases and compare them with all the articles published in the national press, the BBC, and Sky News online. It has been funded out of the grants we receive from charitable foundations, in order to raise awareness about churnalism—including, in the U.K., the Esmée Fairbairn Foundation, the Gatsby Foundation and the Joseph Rowntree Charitable Trust. In the US, the Trust has also previously been supported by the MacArthur Foundation and the Knight Foundation (we were a News Challenge award winner in 2008).

Paste in a press release, hit ‘compare’, and the site will compare the release with over three million articles from the national press, BBC, and Sky. It will then tell you what percentage of the press release has been cut and pasted and used in which news articles.

How churnalism.com works

From a technical perspective, figuring out whether something is churnalism is not entirely straightforward. We tried a bunch of different methodologies before using the one we ended up with. At first we looked for distinctive words that were in both the press release and news articles (which can be very helpful in identifying pieces about a similar subject). But we found this was highly erratic in pinpointing churn.

Eventually we created our own methodology, based on compression, re-indexing, and matching. Essentially, the site compresses all articles published on national newspaper websites, on BBC News, and Sky News online, into a series of numbers based on fifteen character strings (using a “hash function”) and then stores them in a fast access database.

When someone pastes in some text and clicks “compare,” the churn engine compresses the text entered and then searches for similar compressions (or “common hashes”). If the engine finds any articles where the similarity is greater than 20 percent, then it suggests the article may be churn. Churnalism.com is powered off the back of the database of over three million compressed articles in journalisted.com.

Churnalism.com is not going to “solve” the churnalism problem, nor is it supposed to. News outlets will continue to copy and paste press releases. Indeed listening to people in public relations, press releases are already “old school.” Much better to feed PR in via the news agencies than go straight to the news desks, and easier to strike long term commercial relationships with news outlets that enable constant cross promotion.

But churnalism.com will provide people with a tool to help distinguish between journalism and churnalism. And maybe it will make journalists think twice before putting their byline at the top of the next press release, and link to it instead. Who knows? It may even encourage more original journalism—which would be a very good thing.