It’s not just journalists who are overwhelmed with information. Audiences are too. People are consuming media all day long, in one form or another. So instead of merely contributing to information overload, news organizations need to help audiences navigate it. And that means focusing on quality over quantity.
Many publishers are actually growing audiences by producing less content. They include (to pick a few) Gannett, the Guardian, the Times of London, the French newspaper Le Monde, the Buffalo News, the San Diego Union-Tribune, and the Post and Courier of Charleston, S.C. These publishers significantly cut back on the number of stories they were doing — and then saw higher traffic, more engagement and ultimately more subscribers.
Gannett embarked on an initiative a couple years ago to cut out underperforming stories across all of its newsrooms, after research showed that the bottom half of Gannett content accounted for only about 6% or 7% of overall readership. It ended up cutting the number of stories its newsrooms were posting nearly in half; doing fewer stories but more high-quality, original reporting. Article views went up during that same time period.
“Publishers … discover that when they cut away the not valuable, nobody realizes that it is gone,” said media analyst Thomas Baekdal in an interview with Digiday. And not only do they not realize it’s gone, they apparently start engaging more with what’s left.
Identify stories that don’t serve any part of your ‘funnel’
But what is the “not valuable”? When Gannett cut nearly 50% of its output in 2018, it was guided by page views. The types of stories that hardly anyone was looking at went on the chopping block. But the company has since changed how it defines value: It’s not just about broad reach but also about which stories help gain and retain subscribers, Josh Awtry, vice president of content strategy at the USA TODAY Network, told API.
Several metrics can help you determine that. The key is making sure they represent each stage of the audience “funnel,” from casual user to paying subscriber or member. Stories that don’t serve any stage of your funnel can be cut — and few people will even realize that they’re gone.
So let’s break it down by stage so you know which metrics you should be looking at.
Metrics include: Page views.
Page views may just be one piece of the puzzle, but they’re still an important piece. “Every loyal subscriber started out as a new reader once. If we’re not thinking about reaching broad swaths of the community in search of a more diverse readership to bring on the subscriber journey, we’ll never be able to build up enough subscribers to support our newsrooms,” Awtry said.
The Milwaukee Journal Sentinel found that digital traffic for its community blotter was almost nonexistent. So it stopped doing those stories and focused on the crime stories that really mattered. Over the next month, they produced nearly 100 fewer stories — but total digital traffic went up by 19%. “Our reporters have more time to focus on the stories they really want to write and have increased the median page views by 62 percent,” editors wrote.
Metrics include: Time spent reading a story, scroll depth, reader comments and social shares; plus all of these metrics for audience segments like newsletter subscribers and frequent returners to your site.
How long readers are spending on a story is one great indicator of which content is deepening your relationships with readers. If a story has modest reach, but its readers are spending an average of three minutes on it, that’s a sign the story is resonating with certain people.
Metrics include: Page views of a story by readers who purchased a subscription shortly after, number of subscriptions purchased from the story page, number of paywall or meter “stops.”
Subscription conversion is a key metric for news organizations that are shifting to a reader revenue-supported business model versus an advertising-based model.
Tynin Fries, digital strategist at the Denver Post, says they “made a conscious effort to stop writing stories that don’t drive subscriptions.”
Although it hasn’t led to them producing less content, she said, it did lead to them shedding “a lot of low-level crime and sentencing stories. Now we try to focus on only active situations or fatal crimes [and] crashes. That gives our breaking news reporters more time to watch for trend stories that better fit what our audience wants.”
Metrics include: Any data that shows what content subscribers are engaging with; could be subscriber page views, time spent on a story, subscriber comments, etc.
A recent study found that 49% of digital subscribers to local news are “zombies” who visit the website less than once a month. Let that sink in for a moment. Most of your subscribers are barely interacting with your reporting — which is not good news for retention. Research also shows that regular reading habits is the topmost factor in whether subscribers end up sticking around.
So examining which stories your subscribers are engaging with is absolutely critical to understanding what’s worth doing and what’s not. Spend less time churning out stories and more time looking at what subscribers want, and finding more opportunities to get that in front of them.
After the Arizona Republic made the highly unwelcome discovery that 42% of its digital subscribers weren’t even visiting its website once a month, and that those zombies made up 50% of subscription stops each month, it set out to examine which stories were resonating with subscribers (particularly the zombies). “We found that the most successful zombie-killer stories in the newsroom were those that had strong news elements combined with a bit of a hook — a unique spin on a viral story, a strong human element, or a format that grabbed readers quickly,” wrote John Adams and Alia Beard Rau. “And, most importantly, one not written on print deadline.”
Spend less time churning out stories and more time looking at what subscribers want, and finding more opportunities to get that in front of them.
Again, if you start noticing story types that don’t serve any part of the funnel — they’re not getting page views, readers who do land on the pages aren’t staying long, subscribers are ignoring them — it’s safe to say you can cut them. Consider them the low-hanging fruit in this exercise.
The Buffalo News uses API’s Metrics for News analytics tool to look at a customized blend of metrics called an Engagement Score — including page views, time on page and social media engagement — to see which story types readers care about the most.
Looking at this blend of metrics has allowed the Buffalo News to cut back significantly on daily stories that don’t get high engagement, such as quick updates from local government meetings, and instead channel reporters’ time into more enterprise reporting.
Even topics that weren’t fetching audiences when written about as daily stories — like the school board, police and narcotics in Niagara County — were more appealing to readers when approached from an enterprise angle.
“Cutting down on dailies is one part of the strategy, but probably a more important part is knowing when to invest more time and effort into a story so it’s not just a daily story with limited impact,” said enterprise editor Patrick Lakamp.
“The most important thing editors can do is say yes to the enterprise stories and make sure reporters have the time to do them … but also say no or discourage the daily-grind stories,” he added.
Doing that can also boost reporters’ morale.
Speaking about Gannett’s exercise in cutting content across its newsrooms, Awtry said, “Newsrooms received the training and focus well — it’s not often you get to tell journalists that they’re working too hard and that they should take a breath, get off of the hamster wheel, and think about how to use their talents to do more dot-connecting, enterprise work. It’s what got most of us into the business, after all.”
Find patterns in the data
Looking at more data won’t help if you can’t easily find patterns and themes in that data — especially when you’re using it to decide which stories to kill.
“Steering by ‘this story good, that story not as good’ only gets you so far,” Awtry said. “We also want to unearth the taxonomical link between the good stories and the stories that didn’t connect with readers.”
Your best bet for easily unearthing insights from your data is having consistent, thorough story tagging and categorization practices within your CMS.
The Times of London, for example, started tagging stories with 16 different pieces of metadata, including tone, headline type, article format and geography. It then plotted those tags against metrics like page views; time on page; and comments, saves or shares. It looks at these metrics for two audience segments: registered readers and subscribers.
The tagging system gave them valuable insights into what’s working for readers — and what’s not. They started publishing 15% fewer stories on the online Home News section and actually saw readers’ time in that section increase. And users of the Times mobile app started spending an average of 28 minutes daily on the Home News section, an increase of 25% year over year.
But what if your CMS isn’t currently built for tagging stories, and you don’t have the expertise to change that?
You can still do short, relatively quick experiments to identify which stories to kill. Take one reporter, beat or topic at a time. The Buffalo News, for example, examined its food coverage to see which story types were and were not resonating with readers.
Features editor Geoff Nason exported the analytics data for all food articles published over a certain time period to an Excel sheet and assigned “categories” to each story to help surface trends, things like story type (feature, guide, review and restaurant opening/closing articles, for example) and author. He then created pivot tables within the Excel sheet to make it easy to compare how well certain stories attracted readers.
Nason found that restaurant guides performed well, but a short feature called “Hot Dish,” which spotlighted one dish at a time from a local restaurant, was underperforming. At first the columnist tried a different tack, reviewing Buffalo-specific dishes at restaurants around town and experimenting with headlines, but when this approach also failed to engage readers, he was able to abandon the column with confidence.
Next (and last), we’ll look at what to do (if anything) with the time you’ve saved by getting rid of low-value work.