{"id":2451899,"date":"2026-06-09T20:05:45","date_gmt":"2026-06-09T20:05:45","guid":{"rendered":"https:\/\/celebrity.land\/en\/?p=2451899"},"modified":"2026-06-09T20:05:45","modified_gmt":"2026-06-09T20:05:45","slug":"what-kind-of-stories-are-best-at-turning-local-news-readers-into-subscribers-its-hard-news-not-the-soft-stuff","status":"publish","type":"post","link":"https:\/\/celebrity.land\/en\/what-kind-of-stories-are-best-at-turning-local-news-readers-into-subscribers-its-hard-news-not-the-soft-stuff\/","title":{"rendered":"What kind of stories are best at turning local news readers into subscribers? It\u2019s hard news, not the soft stuff"},"content":{"rendered":"

<\/p>\n

\n

Let\u2019s start with the good news. What types of news stories are most likely to make a reader subscribe on a local newspaper\u2019s website? Is it celebrity news, horoscopes, sports scores, the gardening column? Nope \u2014 it\u2019s hard news. Local government, public health, politics \u2014 the sort of stuff that makes for a healthy democracy. Those stories are much more likely to turn a reader into a subscriber than the softer stuff.<\/p>\n

The bad news? Even those hard news stories don\u2019t convert enough readers to sustain the cost of producing them.<\/p>\n

Those findings come out of one of the most remarkable bits of journalism research<\/a> I\u2019ve ever read \u2014\u00a0a granular analysis of a newspaper\u2019s web traffic at a scale we\u2019ve never seen before. We\u2019re talking more than 1.2 billion<\/em> user sessions, covering more than 600 million<\/em> individual article visits, all of them tied to unique user profiles, over a four-year period. Researchers were able to track each reader\u2019s path \u2014 how often they visited, what types of articles drew their attention, and what they did each time they were confronted with a paywall and a decision: offer up a credit card or go find something else to read online.<\/p>\n

\u201cI think, at least among people who study communication, the conventional wisdom is that most people are interested in entertainment and sports, only incidentally exposed to politics coverage at all \u2014 they don\u2019t really seek it out,\u201d said Gregory J. Martin<\/a> of Stanford University, the paper\u2019s lead author. \u201cIf they get it at all, it\u2019s by accident. That, I think, is kind of the conventional wisdom, both among scholars of journalism as well as among people who actually run newspapers.<\/p>\n

\u201cOur paper is making the point that that is basically true \u2014 if you look at visits. Those are the sort of articles that generate the most traffic. But willingness to pay in attention is really different than willingness to pay in dollars.\u201d<\/p>\n

The paper\u2019s title echoes a century\u2019s worth of publisher audience surveys \u2014 \u201cWhat do news readers want?<\/a>\u201d and it\u2019s by Martin, Shoshana Vasserman<\/a>, and Cameron Pfiffer<\/a>. (Vasserman\u2019s also at Stanford; Pfiffer now describes himself as a \u201crecovering financial economist<\/a>.\u201d)<\/p>\n

The researchers\u2019 data comes from a single newspaper, which they have anonymized here. It\u2019s described only as a \u201cmetropolitan daily newspaper headquartered in a large U.S. city,\u201d with the additional detail that it is \u201ccurrently owned by a private-equity-controlled holding company.\u201d So it\u2019s probably a reasonable guess that it\u2019s a paper owned by Alden Global Capital (MediaNews Group<\/a>, Tribune Publishing<\/a>) or Chatham Asset Management (McClatchy<\/a>). Digital subscriptions account for only about 40% of the paper\u2019s total subscribers, the remainder still in print \u2014 but of course print has done nothing but dwindle for many, many years.<\/p>\n

Online, the paper has your standard metered paywall, one whose boundaries have varied over time \u2014 five articles every 30 days, three articles every 60 days, and so on. Whenever a user hit those boundaries, a paywall would appear, offering a cheap intro rate to subscribe and keep reading. The data researchers had about these readers\u2019 behavior was extremely rich. (Creepily rich, for people with certain views about digital privacy \u2014 though of course it was all anonymized for research purposes.) How deep they read into each individual article; how many words (estimated) they had consumed in the previous six weeks; how many times they\u2019d bumped into a paywall and bounced right off.<\/p>\n

On the flip side, they had rich data on the articles themselves and who produced them. Stories were divided up via content analysis into eight distinct \u201cbeats\u201d: Sports, Entertainment, Local News, Health, Business, Local Events, Editorial, and Crime. They tracked whether stories mentioned at least one local place name. Staff-written articles were separated from wire stories. Pieces were also categorized based on whether they met eight \u201cCommunity Information Needs\u201d as defined through an FCC report (things like Emergencies and Public Safety, Environment and Planning, Economic Development, and Civic Life) as well as six others researchers defined (like Real Estate, Things to Do, and Opinion Columns). <\/p>\n

Each story was tied to the reporter(s) who produced it, tracking their relative frequency of publication. And stories were flagged as being \u201cinvestigative\u201d or not using a creative measure<\/a> that looked at how much an individual story influenced future coverage of the same subject. (I think \u201cimportant\u201d might be a better term for what they\u2019re measuring than \u201cinvestigative,\u201d but that\u2019s a quibble.)<\/p>\n

They also divided all of the site\u2019s non-subscribers, based on their behavior, into three different \u201cbins,\u201d ranging from casual, one-off readers to those eager enough to bump into paywalls regularly. (\u201cBin 3 users are more than 100 times as likely to subscribe as those in bin 1, conditional on encountering a paywall.\u201d)<\/p>\n

Basically, they had near god-like visibility into the content this newspaper produced, all the ways readers consumed it, and the intersections in between. Let\u2019s go through some of the most interesting findings.<\/p>\n

First of all, this paper loved<\/em> to cover sports. When articles are broken down by the \u201cinformation needs\u201d they meet, Sports is far and away No. 1 in both staff-written and non-staff content. The only other \u201cinformation need\u201d near it among staff articles is \u201cEmergencies and Public Safety\u201d \u2014 which overwhelmingly means crime stories.<\/p>\n

<\/p>\n

But what happens when you look at how those information needs aligned with the two output metrics the authors are measuring \u2014 how many visits they generate and how many subscriptions they generate? The somewhat confusing chart below is actually two charts \u2014 non-staff articles on the left and staff articles on the right. Each point on the chart represents how much value those articles offered in terms of visits (x-axis) and subscriptions (y-axis) compared to the site\u2019s average. <\/p>\n

\"\"<\/p>\n

In the bottom left, you can see that non-staff articles are all below average in both visits and subscriptions \u2014 with the single exception of columns, which are a big winner in visits but still a loser for subscriptions. (Think advice columns or syndicated opinion columnists.)<\/p>\n

Meanwhile, among staff-written articles, the \u201chard news\u201d article types \u2014\u00a0marked in red \u2014 fared better in both visits and subscriptions than the \u201csoft news\u201d types marked in blue. <\/p>\n

(This is as good a place as any to note that huge outlier in the upper right \u2014 health stories. This analysis covers January 2020 to December 2023 \u2014 which means it includes an enormous number of Covid stories, which drove enormous<\/em> reader interest, including a boomlet in subscriptions. So the fact that health stories look wildly<\/em> more successful than anything else the newspaper produces is in large part an artifact of the pandemic. Martin told me that, if you only look at the later years of the study period, health stories still performed well \u2014 just not as absurdly<\/em> better than every other type of story. Still, if you wanted to get someone to convert someone from casual reader to subscriber, there has basically never been a tool as effective as putting a Covid article behind the paywall circa 2020.)<\/p>\n

Here\u2019s how each beat contributed on visits and subscriptions within each of the three user bins they\u2019ve defined. (Bin 1 is casual readers who will basically never subscribe. Bins 2 and 3 are each increasingly more frequent and dedicated readers.)<\/p>\n

\"\"<\/p>\n

\n

Unsurprisingly given the subscription rates, Bin 1 subscription utilities are uniformly much lower than the other two reader types. For the higher-propensity bins, however, hard news beats like Business, Health and Local News\u2026generally outperform the soft news beats like Entertainment and Sports.<\/span> Almost all in-house beats outperform wire-sourced articles on both dimensions for Bins 2 and 3. For Bin 1, wire-sourced articles are at the bottom in traffic generation but average in subscription utility.<\/p>\n<\/blockquote>\n

\u201cEven for people who, most of the time in their past history, read sports and weather articles and things like that, their potential to subscribe was still higher when they encountered a paywall on a story about politics, or about public health, or about one of our other hard news topics,\u201d Martin told me. \u201cSo I don\u2019t think it\u2019s just that it\u2019s a different person who is on the margin of subscribing versus visiting\u2026People are able to recognize what\u2019s valuable, and that\u2019s different from what they\u2019re willing to click on to read.\u201d<\/p>\n

Martin et al. then engage in a bit of fantasy-sports-for-newsrooms. If you wanted to optimize your newsroom for web traffic or for digital subscriptions, how would you allocate your resources? Which beats would you devote more reporters to, and which ones would you cover less?<\/p>\n

Assuming that overall headcount remained constant, the researchers say that reducing coverage of crime would improve both<\/em> visits and subscriptions. Increasing<\/em> coverage of health would do the same \u2014 though note the caveat above about the uniqueness of Covid. For other beats, though, chasing visits and chasing subs point in opposite directions. Add more entertainment reporters? You\u2019ll increase visits but reduce subscriptions. Add more local news reporters? You\u2019ll decrease visits but increase subscriptions.<\/p>\n

\"\"<\/p>\n

\"\"<\/a><\/div>\n

All of that sounds like good news for those of us who would like local newspapers to protect its most civically useful beats \u2014 the \u201ciron core\u201d of journalism<\/a> \u2014\u00a0whenever there\u2019s another round of cuts to be had. If your newsroom still lives and dies by Chartbeat \u2014 if pageviews are all that matters to management \u2014 it\u2019s missing out on some critical intel. The stories that get visits might be the ones you should be doing fewer<\/em> of if your goal is chasing subscriptions. Smarter newsrooms have known this, at least intellectually, for a while, of course. But here\u2019s hard data proving it.<\/p>\n

But what about that bad news? Because Martin et al. have all this data tying reporters to stories to visits to subscriptions, they also have a go at testing whether hiring an additional journalist might even pay for itself. If more local news means more digital subscriptions, could we be at a point where a reporter\u2019s salary might be covered by the extra subscriptions that her work generated? If that were true, it\u2019d be an excellent<\/em> case for further investment in newsroom capacity.<\/p>\n

Unfortunately\u2026it\u2019s not. Even in the most optimistic scenarios, the authors find, one reporter\u2019s digital subscriptions don\u2019t come close to paying one reporter\u2019s salary.<\/p>\n

Here\u2019s a chart showing the relative share of a marginal reporter\u2019s salary covered by marginal digital sub revenue. (Note that the researchers don\u2019t have access to this newspaper\u2019s reporters\u2019 actual salaries; they\u2019re using market averages.) Adding a local news reporter will generate digital subscriptions all right \u2014 but only enough to cover something like 1\/4 of their salary. Even during peak Covid, a health reporter\u2019s digital subs would only cover around 60% or so of their salary.<\/p>\n

\"\"<\/p>\n

To be fair, Martin notes that this methodology only accounts for the digital subscription revenue that an individual reporter might generate. Newspapers make money in other ways \u2014 from print (somehow!) and from online ads (theoretically!). But neither of those is going in the right direction, and the connection between an individual reporter\u2019s work and revenue is much more abstract. \u201cIn a world where newspapers were exclusively online, for the staff, the digital subscriptions alone wouldn\u2019t have covered the the cost, at least during this period,\u201d Martin told me.<\/p>\n

So that\u2019s the paper\u2019s central conundrum. If a newsroom wants to optimize for digital subscriptions \u2014 which for more than a decade has been the closest approximation of a sustainable business model for high-quality local news \u2014\u00a0it should lean into hard news. But no matter how hard it leans, the underlying numbers remain dangerously unstable. <\/p>\n<\/div>\n

\u2018 The preceding article may include information circulated by third parties \u2019 <\/em><\/p>\n

\u2018 Some details of this article were extracted from the following source www.niemanlab.org \u2019 <\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

Let\u2019s start with the good news. What types of news stories are most likely to make a reader subscribe on a local newspaper\u2019s website? Is it celebrity news, horoscopes, sports scores, the gardening column? Nope \u2014 it\u2019s hard news. Local government, public health, politics \u2014 the sort of stuff that makes for a healthy democracy. […]<\/p>\n","protected":false},"author":2,"featured_media":2451900,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_jetpack_memberships_contains_paid_content":false,"jnews-multi-image_gallery":[],"jnews_single_post":[],"jnews_primary_category":[],"jnews_social_meta":[],"footnotes":""},"categories":[25177],"tags":[],"class_list":["post-2451899","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-celebrities"],"jetpack_featured_media_url":"https:\/\/celebrity.land\/en\/wp-content\/uploads\/2026\/06\/What-kind-of-stories-are-best-at-turning-local-news.png","jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/posts\/2451899","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/comments?post=2451899"}],"version-history":[{"count":1,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/posts\/2451899\/revisions"}],"predecessor-version":[{"id":2451901,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/posts\/2451899\/revisions\/2451901"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/media\/2451900"}],"wp:attachment":[{"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/media?parent=2451899"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/categories?post=2451899"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/celebrity.land\/en\/wp-json\/wp\/v2\/tags?post=2451899"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}