The local-mobile paradox: Missed innovation opportunities at local newspapers

Literature Review and Conceptual Background

The idea of mobile news delivery has been around since the late 1990s, when some publishers experimented with pushing headlines to pagers (Westlund, 2013), but mobile news became a widespread reality with the 2007 launch of Apple’s iPhone. Later, industry leaders pondered how smartphones would change news production and provided regular primers for working journalists curious about the new technology (Society of Professional Journalists, 2018). Consensus grew that the future of journalism itself was mobile (Breiner, 2014), and some observers began warning that news outlets needed to harness mobile opportunities or risk missing another wave of innovation (Hindman, 2015).

As digital news audiences have migrated from desktop to mobile device, it has become clear that mobile news consumers are not homogeneous, with significant differences emerging between those who consume news via apps and those who use browsers (Nelson & Lei, 2018). Recent years have brought some digital success to larger publishers thank in part to robust mobile strategies, and a variety of news organizations are experimenting with mobile news delivery and storytelling (Howe, Bajak, Kraft, & Wihbey, 2017). Less is known, however, about the state of mobile-related innovation in rural and suburban newsrooms.

Innovation, or the lack of it, is a central concern of contemporary journalism. Why the news business did not innovate sufficiently to thrive in the early years of the Internet revolution—thus helping to precipitate a massive crisis across the industry—remains a key question. This perceived lack of innovation is at the core of the industry’s self-understanding and framing of future challenges (Huey, Nisenholtz, & Sagan, 2013). Scholarship on innovation in newsrooms has continued to examine a wide variety of dynamics, factors, approaches, and predictors with regard to both legacy organizations and startups, and external actors such as foundations have funded major efforts and employed various strategies to try to spur innovation (Lewis, 2011; Lowrey, 2011; Usher, 2017).

Empirical work by Lowrey (2011, p. 74) suggests that newsroom innovation is typically limited and constrained, but “where actual product innovation occurs, it is driven to a large degree by factors related to resources and markets, such as public ownership, organizational size, and audience feedback.” News outlets sometimes follow and mimic industry trends in terms of innovation, although these modes of trend-following and imitation sometimes result in questionable outcomes, diminishing the supply of distinctive, high-quality news (Boczkowski, 2010).

The adoption of new technologies by local news organizations might be understood more broadly through the five-stage diffusion of innovations theory: innovators, early adopters, early majority, late majority, and laggards (Rogers, 2003). Given their relative lack of resources and diminished access to innovation hubs in metropolitan areas, local, legacy news outlets are likely to be in the late majority or laggard categories. As Deuze (2004) has suggested, innovation in news work has a complex and unique set of dynamics at many different levels, with internal resistance to technological innovation issuing from deeply embedded values in journalism. Deuze’s idea of a “multimedia logic” helps furnish broader perspective that can take account of particular organizational, institutional, technological, and producer/user norms and perspectives in the news business. The push and pull of forces around change and adoption of new technologies in newsrooms is multifaceted and often transcends simple threshold models that characterize consumer product adoption. Explanation of the logic of innovation, or its absence, in newsrooms often requires ethnographic study of individual firms (Anderson, 2013; Usher, 2014).

Research Questions

Given the ubiquity of mobile technology, we might expect that, by now, local newspapers would provide reasonably high-quality mobile experiences for their readers and advertisers. Yet many local publishers also face serious financial challenges, a reality that may stymie technological innovation. The overall picture for local mobile news remains unclear, and this study’s goal is to provide both a granular picture and a more general exploration of the intersection between local news and mobile technology to evaluate non-metropolitan newspapers’ readiness for the current phase of digital disruption.

Method

Data collected and analyzed in this mixed-methods study have two facets: (a) an analysis of newspapers’ online platforms across the United States (the “website sample”) and (b) a survey with local news leaders of legacy outlets in New York state (the “survey sample.”). The two methods are complementary. One provides granular perspectives from leaders of local news organizations across an important state media ecosystem. The other features, national objective data about a sample of newspaper websites and their degree of mobile optimization.

Our analysis of mobile newspaper online sites focused on a sample of one daily and one non-daily news organization in each of the 50 states. Although some local publishers offer smartphone apps, the mainstream expectation across the industry is for websites to be responsive—a term that indicates their ability to conform to mobile devices. There is an ongoing industry debate about native apps versus responsive sites, with publishers backing away from apps for a number of years and then, more recently, renewing their interest in that method of news delivery (Knight Foundation, 2016; Owen, 2017). Websites accessed through browsers allow for greater public discovery of stories through search engines and for circulation of content over social media. News audiences also behave differently across mobile platforms, making it useful to draw distinctions between those who consume news through apps and those who use browsers (Nelson & Lei, 2018). In one of the most detailed field studies, Nielsen and the Knight Foundation examined a large panel of U.S. smartphone users over a 2-year period and concluded that the audience for news apps is relatively small (Knight Foundation, 2016). The number of people accessing news sites through the mobile web greatly exceeds the audience using an app (often a 3:1 or 5:1 ratio), even for exemplar sites with high-end, custom apps. Because of this context, this study focuses only on browsers.

We included both daily and non-daily publications in our website sample because of the important role weekly and even monthly newspapers play in rural communities. We deliberately took a stratified sample in order to represent all 50 states. (We recognize that this may give outsized importance to smaller states with fewer news outlets.) We then selected these sites randomly within each state from the Library of Congress U.S. Newspaper Directory (n.d.), using the first newspaper with a website that had been updated within the preceding 30 days. Our goal was to look at rural and suburban publications, but we had to include several larger publications because of limited newspapers in small states such as Delaware and Rhode Island. Our sample reflects diversity in terms of geography, circulation, and ownership model. Digital publishing is ephemeral, especially at the local level, so this sample should be viewed as a snapshot of the local mobile ecosystem.

Our technical analysis had three components: a computer-generated metric (the Google PageSpeed Insights described below), human observation of site responsiveness, and ad placement (Google PageSpeed Insights, n.d.). Data related to these factors were collected during the week of 5 February 2018, and verified by both authors. Best practices in mobile news design and delivery are still evolving, but several norms have emerged in recent years. Leaders in mobile news delivery employ a number of shared features that we used as a baseline to compare with the local mobile sites. We selected three characteristics, common among these exemplar publications: (a) fully responsive design (a website that conformed automatically to the size of the mobile device), (b) ads interspersed at regular intervals throughout the mobile version of the site, and (c) mobile optimization as measured by Google PageSpeed Insights. To form our baseline, we looked at six exemplar publications with a variety of owners: The New York Times, The Washington Post, The Los Angeles Times, Vox, BuzzFeed News, and The Guardian’s U.S. site. All six featured fully responsive mobile sites. Five were designed with ad stacks interspersed throughout the news content. The sixth, BuzzFeed, had no traditional ads, but it did include links to sponsored content.

To measure mobile optimization, we chose Google’s Page Speed Insights mobile optimization score calculator, which evaluates “how well a page follows common performance best practices and computes a score from 0-100 that estimates its performance headroom.” In lay terms, the optimization score shows how much a particular site could be improved technically relative to its unique structure. This, we believe, is a more objective measure than page load speed, which is contingent on transient factors. As Google’s Page Speed Insights tool documentation notes the following:

Since the performance of a network connection varies considerably, the optimization addresses the network-independent aspects of page performance: the server configuration, the HTML structure of a page, and its use of external resources such as images, JavaScript, and CSS. Implementing the suggestions should improve the relative performance of the page.

The second data collection method was an anonymous online survey with newspaper publishers (owner-operators) across New York state, conducted with the help of the New York Press Association (NYPA), the nation’s oldest and largest state professional organization. NYPA agreed to facilitate the independent research project at no cost, and no compensation was provided to the authors. Although it would have been advantageous to have a larger, national sample, ours allows for more fine-grained comparison of outlets within a particular regional news ecosystem with common features. Such news ecosystems can vary widely, and may depend on the density and character of a population and resources in a given state, and the degree of resilience in the face of the industry’s decline (Matsa & Boyles, 2014). As an analytical study, comparisons within the sample have validity, even if the sample does not allow for general statements about local news outlets in all states. The sample therefore provides a snapshot, not a full picture.

New York state has about 55 dailies and about 650 weekly newspapers; roughly two-thirds of the weekly newspapers belong to NYPA, which has 420 members. Half of those members have print circulation figures below 5,000, according to NYPA’s own figures. On August 1, 2017, NYPA distributed an online Qualtrics survey devised independently by the authors to roughly half of the members (230 news leaders); a follow-up reminder was sent a week later, and the survey was completed on August 21. Seventy-seven publishers completed the survey (33% response rate), providing an overall sample representing about 11% of all newspapers in the state. This response rate compares reasonably favorably with rates found across other types of online surveys (Cook, Heath, & Thompson, 2000; Kaplowitz, Hadlock, & Levine, 2004). In the sample were 8 dailies and 69 weeklies/other non-dailies such as monthlies. Thus, the sample distribution (10% dailies) roughly conforms to the distribution of dailies and weeklies/non-dailies overall in the state (8% dailies).

Results and Discussion

To answer RQ1, RQ2, and RQ3, the analysis for all 100 websites was recorded for each metric. We compared the means of these numbers against those of a six leading news sites known for digital innovation. For RQ4, RQ5, and RQ6, a linear regression was conducted on the relevant variables.

All local websites sampled performed well below the exemplar sites on Google PageSpeed Insights. The mean Google optimization score of the sample was 73.61 (SD = 12.98). The exemplar sites that provided a baseline of comparison had an average mobile optimization score of 90.8. The bulk of the local sites were, however, responsive. Of 100 local newspaper mobile sites, 80 were fully responsive and 8 exhibited some responsive components but also featured page elements that did not conform to a mobile screen. The remaining 12 were not responsive. Ad placement, meanwhile, was variable: 49 sites featured advertising positions interspersed throughout news content; 29 clustered advertisements at the top or bottom; and 14 sites had no advertisements when displayed on a mobile device. The remaining eight sites were among those that are not responsive and displayed desktop advertisement stacks on mobile.

The survey collected specific data that could provide insight into the nature of mobile news at the local level and the strategy of local news leaders, who were asked to respond to three questions. These three questions asked respondents to evaluate their own mobile outlets and efforts on a five-point scale (“a great deal,” “a lot,” “a moderate amount,” “a little,” or “none at all.”) The questions were the following: (a) “How much of your news outlet’s digital readership would you estimate comes from mobile devices?”; (b) “How much of your online advertising revenue comes through mobile devices?”; (c) “How much attention has your outlet devoted to making changes to better serve your mobile readership?” The survey’s initial screening questions also collected data about the outlet’s size; a five-point scale was utilized to indicate print circulation and annual unique online visitors to the news website.

Responses to the mobile-related questions and outlet size/online visitors questions were then each recoded 1 to 5, and surveys with any omitted responses (or those indicating “no website”) in the relevant questions were removed. A linear regression was performed among variables with the strongest theoretical justification and relevance to the research questions. To evaluate RQ4, the amount of attention publishers devoted to mobile product was used as the independent variable and the amount of advertising revenue through mobile was the response variable; for RQ5, the newspaper’s size as measured by print circulation was the independent variable and the amount of attention the outlet devoted to improving its mobile product was the response variable; for RQ6, the size of a publisher’s online audience was the independent variable and the amount of mobile traffic to the news outlet was the dependent variable.

The distribution for print circulation among relevant respondents (n = 75) was reasonably balanced between small and large local news outlets: 1 to 5,000 (31%); 5,000 to 10,000 (13%); 10,000 to 50,000 (36%); 50,000 to 100,000 (16%); and more than 100,000 (4%). Likewise, annual unique visitors of each news outlet saw a reasonably wide spread distribution: 10,000 or fewer (16%); 10,001 to 100,000 (20%); 100,001 to 500,000 (26%); 500,001 to 1 million (11%); more than 1 million (17%); and “no website” was another option (10%).

The three mobile-focused questions also saw a range of responses (see Figure 1). In terms of how much an outlet’s digital readership comes from mobile devices, publishers reported wide-ranging outcomes across a five-point scale: a great deal (21%), a lot (24%), a moderate amount (23%), a little (17%), and none at all (15%). Relatively few publishers reported seeing significant advertising revenue coming from mobile devices: a great deal (3%), a lot (5%), a moderate amount (19%), a little (24%), and none at all (49%). Finally, publishers on the whole reported that they had not devoted much attention to serving mobile readerships. Distribution of responses was the following: a great deal (8%), a lot (8%), a moderate amount (28%), a little (40%), and none at all (17%).

OPEN IN VIEWER

Figure 1

Local Newspaper Owners in New York on Questions of Mobile Readership, Advertising and Strategy

Our website sample results suggest that, although mobile innovation may be present at some local news outlets, many have a long way to go. The majority of the sites we examined are responsive, but most are not taking full advantage of mobile design in terms of advertisement placement. Although 80% of sites were fully responsive, only about half had altered their advertising display to reflect mobile best practices. Local newspapers also had a substantially lower optimization score in comparison with exemplar sites (17.2-point difference on a scale of 100). This points to a repetition of the “shovelware” phenomenon common during the Web’s early days when newsrooms shoveled print content online with little regard for the demands of digital platforms. Local newspapers are also lagging behind industry leaders in terms of mobile site optimization. This is true even though reducing technical “headspace” in websites can be accomplished, for the most part, without purchasing high performance servers or caching services. This implies a lack of technical literacy among local news publishers that could be contributing to stalled mobile innovation.

Our survey results suggest real benefits for publishers who embrace mobile delivery and that even small publishers may be positioned to leverage those gains. The model indicated that, at a significant level, directing more attention and resources toward mobile optimization was associated with greater advertising revenue through mobile (β = .327, SE = 0.119, p < .01). It also showed that being a larger newspaper with greater print circulation was not necessarily associated with greater attention to mobile innovation (β = .0867, SE = 0.114, ns) and that newspaper size was not associated with attention to mobile.

The survey also showed that the size of a publisher’s online audience was associated at a highly significant level with the overall percentage of traffic from mobile devices (β = .544, SE = 0.096, p < .001). Evidence supporting the last point was especially robust and hints at the difficulties for local publishers in terms of growing audiences and revenue. It is somewhat striking, for instance, that more websites are not employing best practices for mobile ad display. Although digital advertising is not a major revenue source, demonstrating mobile proficiency might help publishers build longer term, more lucrative relationships with local businesses struggling to figure out their own mobile strategies. Offering a sub-par mobile experience may also discourage casual readers, virtually assuring diminished traffic and further cutting into monetization opportunities, whether through increased ad rates, sponsored content opportunities or more traffic into a potential subscriber funnel. This is clearly a missed opportunity; indeed, an American Press Institute (2018) survey that shows 60% of people who had recently subscribed to a newspaper were motivated by a desire for local news and 31% wanted to support local journalism. Local publishers would be wise to woo—and financially benefit from—that audience by devoting at least some institutional attention to creating an appealing mobile site.

The survey sample results also help us understand how local news owners see this domain. Focus on mobile audiences is associated at a significant level with more revenue from mobile. Of course, this may include both the creation of native apps as well as optimization for the mobile web. The results do not suggest that mobile innovation is necessarily associated with firm size, as judged by print circulation, a finding that fits with the anecdotal picture drawn from the website sample and technical analysis. Outlets of all sizes can seize opportunities in the mobile space.

All of that said, we must consider the subtle but important implications of another survey finding, namely that higher website traffic is associated at a significant level with a greater percentage of overall online traffic coming from mobile. This reveals a difficult dilemma for publishers, what we call the “paradox of local-mobile news.” News sites need to generate sufficient web traffic to modify user behavior and promote mobile access habits. Yet without an optimized mobile experience, those audiences may never materialize. The lack of mobile innovation constrains the potential news audience, leading to a vicious cycle where consumers may turn toward other optimized media apps and platforms such as Facebook.

Given the grave concerns about the decline of local news in the country (Coll, 2016) and the parallel rise of mobile devices, this paradox may explain a crucial aspect of our disruptive media moment. Coupled with the other evidence of constrained innovation provided in this study, it may inform the conversation about the future of local news in the United States. The lack of fast-loading, engaging, and relevant local content may have fueled opportunities for the increasingly dominant Internet platform companies to fill the vacuum and substantially undercut the ability of small publishers to support their newsroom (Benton, 2015). Industry, reports in 2018 that both Facebook and Google are now interested in promoting the circulation of more local news through their giant platforms speaks to the depth of local publishers’ problems but also the opportunities for new approaches (Kulwin, 2018).

The patterns observed in our data can, at least in part, be understood through the diffusion of innovations theory, with local news outlets often falling into the “late majority” or “laggard” segments on the adoption curve among news media outlets (Rogers, 2003). Deuze’s idea of a complex “multimedia logic” that is in some ways idiosyncratic and highly particular to individual firms, may help explain the variation in innovation, insofar as some small publishers have excellent mobile responsiveness and optimization while similar others do not. In any case, optimization of websites and responsiveness to mobile devices are technical innovations that were quickly adopted by exemplar news media outlets and are only now present across the sites of many legacy local publishers. Our findings suggest that the local-mobile landscape reflects many dynamics that were at work when news organizations failed to adapt to the web in the early 2000s. In short, history is repeating itself. Adoption of new publishing technology is uneven, and vital community news organizations are missing out on opportunities to connect with their audiences and generate revenue.

These findings could be symptoms of the stagnant newsroom cultures previously documented. The parallels are especially striking when we compare our results with Lowrey’s (2011) survey of local newspaper websites. He found “low product innovation” that indicated “news managers are rearranging deck chairs: restructuring newsrooms, pursuing relationships with the business side, and advocating online monitoring of audiences but doing relatively little to transform the product.” This also echoes the “big disconnect” the Local Media Association (2018) described in a report about the state of digital strategy.

Although the article’s findings are strongly suggestive, the research design had limitations. For example, sample sizes for both the website analysis and survey are limited. A national survey may provide more robustness, although as mentioned, the wide variation among state ecosystems would need to be taken into account in order to make generalizations about local news media overall. The survey instrument used for this study generated self-reported data, plus answer choices such as “great deal,” “a lot,” “moderate amount,” and “a little” probably do not have high reliability and internal validity; research that obtains and uses the actual online analytics of newspapers would be a stronger basis for conclusions. Furthermore, publishers could be reaching mobile audiences with messenger apps or other techniques that were not apparent in our sample. Further research is necessary to better understand how or whether local newspapers are using apps and push notifications to reach mobile audiences. Also worth further inquiry is the mobile readiness of non-newspaper journalism entities operating in rural and suburban communities. This might help researchers better understand the role print production routines play in preventing mobile innovation.

It should be noted that through this analysis we also found some smaller, independent newspapers with responsive sites with well-displayed local advertisements and seemingly ample original news content, anecdotal examples that bear out the survey finding that mobile innovation is not limited to big media companies. Stronger mobile strategies may also help publishers connect with younger audiences, an issue many local newsroom leaders worried about in a 2017 Tow Center survey (Ali & Radcliffe, 2017). Young adults consume much of their news through smartphones (Weiss, 2013), and it would be wise for local newspapers to find ways to serve the next generation of audience members.

Local newspapers face difficult challenges right now, and this study points to the need for a deeper understanding of the barriers between a typical local newspaper and meaningful technological innovation. There is, in short, an urgent need for more research in this area, which might provide the basis for greater collaboration between industry leaders and academic institutions, and ultimately facilitate a more sustainable future for local news.