Research shows “ghost articles” are seriously scary for ROI
Newsroom Transformation Initiative Newsletter Blog | 16 July 2024
How many of the articles you publish don’t get the intended results, whether that’s pageviews, subscription conversions, or some other metric? Do you track that regularly and evolve your content strategy based off of it?
As an industry, we’re usually good at analysing the top of the list — seeing what types of content performs well and celebrating our successes. But we’re not as good at digging into the bottom of the list and understanding what isn’t resonating with audiences and what we should stop doing.
DRIVE, a joint initiative of 30 regional publishers from Germany, Austria, and Switzerland, spent time in recent months studying what they dubbed “ghost articles” and how to write fewer of them. Read on for some interesting insights from that work.
And sign up now for my next Webinar, How The New York Times and Washington Post Are Approaching Elections Coverage, featuring Hannah Poferl, assistant managing editor, chief data officer and head of audience at The New York Times, and Philip Rucker, national editor at The Washington Post. They’ll talk about putting audiences at the centre of their coverage plans for the presidential elections. Takes place July 24; sign up here.
And I’d love to hear from you: amalie.nash@inma.org.
Amalie
P.S.: Have you downloaded my new report, Strategies for Continuously Transforming Your Newsroom? It focuses on how to position your newsroom for the future and how to instill the concept of transformation into the muscle memory of your organisation.
DRIVE publishers focus on eliminating “ghost articles”
Organised by German press agency dpa and consultancy firm Highberg, the DRIVE (Digital Revenue Initiative) project was launched in 2020 and now includes 30 publishers from Germany, Austria, and Switzerland. The cohort tackles initiatives, delves into best practices on shared problems, and learns from data and insights.
One such example is a recent focus on “ghost articles.”
In Germany, 80% of articles fell into the “ghost article” state at the beginning of the analysis, said Hendrik Langen, managing partner at Highberg.
What does that mean? In total, an article received less than five hours of media time — or reading time — from all visits to the article. If the average reading time is one minute, that would translate into 300 visitors reading the article for one minute.
That metric was chosen as the baseline, assuming the average article takes five hours to report and write, Langen said. So if an article gets less reading time than it takes to produce, it’s considered a “ghost article.”
“Our core metric shows 80% of the content is worthless in terms of engagement and media time,” Langen said. “If you put more time into writing an article than people are reading the article, then something is wrong.”
It could be the topic, he said. Or, it may be the distribution and the promotion of the article, said Ole Fehling, manager/data science at Highberg.
“This is also about improving the distribution and personalisation of content distribution,” Fehling said. “It’s also important to look at this from an ongoing basis. Audience interests shift, and this is a very usable metric.”
Two of the German publishers that analysed “ghost articles” moved from 70% in that category at the outset to 30% now, Fehling said.
“Reducing ‘ghost articles’ saves resources,” Langen said. “You have to do this to have free time to do other things and have a growth mindset. Really, the first key thing in the initiative is awareness.”
Part of looking at “ghost articles” is understanding user needs, Fehling said. For instance, the topic of living performs well for “divert me.”
“Helping to introduce user needs to newsrooms is not just one formula,” Langen said. “It’s a very individual mixture of user needs. Planning the right formula is important.”
So how do user needs fit in here?
DRIVE also hones in on user needs model
In addition to the ghost article analysis, DRIVE publishers worked with Dmitry Shishkin, CEO of Ringier Media and a former BBC journalist, on implementing the user needs framework.
The hypothesis: Relevance is a key success factor in digital journalism. Content that specifically addresses readers’ interests and expectations has been shown to increase their willingness to pay.
The problem: The majority of stories published do not meet this requirement, according to data on user behaviour on regional Web sites.
Among DRIVE publishers in the initial analysis:
90% of paid content articles did not generate a single digital subscription.
80% of published articles were read by too few readers or not read at all.
5% of published articles accounted for 50% of total reading time.
“User needs require various adaptations, like you need additional time to plan for articles,” Fehling said. “We developed a whole concept for newsroom change to support the work around user needs.”
The first content analysis, based on more than 5 million archived pieces of content, showed which categories of needs DRIVE publishers have addressed and which have achieved high engagement and conversion rates.
An overview of the key findings:
Mismatch between supply and demand: Regional publishers largely offer traditional news on their sites. However, the data shows readers don’t demand this level of coverage.
Engagement drivers: Emotional and inspiring stories generate the highest user engagement, but these are rarely found on regional sites.
Conversion stars: Visitors to regional sites are more likely than average to read inspiring and opinionated articles before subscribing to a digital subscription.
“We see that more and more publishers understand the need to focus on audience and softer topics,” Fehling said. “We have publishers who are showing how a digital newsroom should work.”
Not surprisingly, “inspire me” stories over-perform in terms of engaged time and conversions, Fehling said. Interestingly, an analysis of crime coverage showed it performed well in the “divert me” category — but perhaps that’s not surprising when you consider the interest in true-crime podcasts among the audiences of these publishers, Fehling said.
Also, the DRIVE initiative was able to analyse user needs by date and time. What they found, for instance, was that readers on Mondays are looking for “divert me” articles.
“If you search on Friday afternoon, you may be looking for ‘help me’ articles on what to do this weekend,” Fehling said. “We find people are looking for more lengthy articles and ‘give me perspective’ at specific times.”
The results:
News publishers have gained an average of 75% more subscribers with needs-based stories than with traditional articles.
Reader engagement has more than doubled, increasing by 125%.
And recommendations for implementing the model:
Acceptance in the newsroom is higher when the newsroom itself decides on adaptations.
The introduction of user needs models has shown that intensive support of editorial teams by a training team has a maximum learning effect.
Each story needs to be assigned a needs category in the content management system.
User needs models are particularly useful for building up ongoing coverage of a topic. Instead of writing one comprehensive article covering all facets, it is a good idea to produce and publish several articles that correspond to the respective user needs categories.
Broadcasting and production schedules based on user needs are an important part of the editorial change process.
What have you seen with user needs? E-mail me: amalie.nash@inma.org.
A quick note on headlines: Simplicity is better
Readers prefer simpler headlines. But journalists like complex headlines.
Science Advances published an analysis of nearly 20,000 headlines in The Washington Post, the popularity of which was determined by the click-through rate, and discovered simpler headlines got higher click rates.
And simple headlines are not necessarily shorter, the research finds. While using common words, an informal style and better readability were associated with higher click rates, character count was not.
Interestingly, in follow-up surveys, researchers discovered that while most people like simpler headlines, there was a “notable departure” with journalists — who themselves preferred more complex headlines.
That, according to the article, is a main takeaway for journalists: Be aware that your experience and your audiences’ experience when interpreting headlines may be leagues apart — and lean into simplicity.
“Thousands of field experiments across traditional (i.e., The Washington Post) and non-traditional news sites (i.e., Upworthy) showed news readers are more likely to click on and engage with simple headlines than complex ones,” the authors write.
Thoughts on headlines? E-mail me: amalie.nash@inma.org.
Mark your calendars
Upcoming INMA events that shouldn’t be missed:
July 24: My next Newsroom Transformation Initiative Webinar on “How The New York Times and Washington Post Are Approaching Elections Coverage” centers on how major publishers in the United States are putting audiences at the centre of their coverage plans for the presidential elections. Moderated by me, it features guest speakers Hannah Poferl, assistant managing editor, chief data officer, and head of audience at The New York Times; and Philip Rucker, national editor for The Washington Post. Register here.
July 31: “A Solution for the Sticky Subject of Commenting” will focus on making comment moderation easier and less time-consuming. This INMA Product & Tech Initiative Webinar is presented by Laura Badura, product manager at Spiegel. Register here.
September 23-27: INMA’s upcoming in-person event, Media Innovation Week, takes place in Helsinki, Finland. I’ll be there! And we’ll talk about future-proofing sustainable news brands in the sustainability capital of Europe. Learn more and register.
About this newsletter
Today’s newsletter is written by Amalie Nash, based in Denver, Colorado, United States, and lead for the INMA Newsroom Transformation Initiative. Amalie will share research, case studies, and thought leadership on the topic of bringing newsrooms into the business of news.
This newsletter is a public face of the Newsroom Transformation Initiative by INMA, outlined here. E-mail Amalie at amalie.nash@inma.org or connect with her on INMA’s Slack channel with thoughts, suggestions, and questions.