<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Kurt Gessler on Medium]]></title>
        <description><![CDATA[Stories by Kurt Gessler on Medium]]></description>
        <link>https://medium.com/@kurtgessler?source=rss-dd9f4cc96152------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*Rt0ymmitGeAVKu_2BTUN0A.jpeg</url>
            <title>Stories by Kurt Gessler on Medium</title>
            <link>https://medium.com/@kurtgessler?source=rss-dd9f4cc96152------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 18:46:44 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@kurtgessler/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Generative AI and newsrooms: Tips and best practices for how publishers can use this transformative…]]></title>
            <link>https://kurtgessler.medium.com/generative-ai-and-newsrooms-tips-and-best-practices-for-how-publishers-can-use-this-transformative-f86f4b9b228?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/f86f4b9b228</guid>
            <category><![CDATA[genai]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[generative-ai-tools]]></category>
            <category><![CDATA[generative-ai]]></category>
            <category><![CDATA[generative-ai-use-cases]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Tue, 22 Aug 2023 21:51:23 GMT</pubDate>
            <atom:updated>2023-08-22T21:53:18.135Z</atom:updated>
            <content:encoded><![CDATA[<h3>Generative AI and newsrooms: Tips and best practices for how publishers can use this transformative tech</h3><p><em>I originally published this article on </em><a href="https://blog.taboola.com/generative-ai-for-publishers/"><em>blogs.taboola.com</em></a><em>, which I’m a senior product manager working with editorial tools and generative AI.</em></p><p>Generative AI is proving to be a Rorschach test for publishers globally. Some editors stare at the inkblot and see an existential threat or a parasite that’s attached itself to their journalism. While others see incredible possibilities, with generative AI products used to automate onerous — even impossible — tasks or to create hyper-personalized content at scale.</p><p>There are certainly elements of truth in each perspective, but the reality of generative AI is neither subjective nor open to interpretation. ChatGPT, Dall-E, Bard and their kin aren’t another “<a href="https://reutersinstitute.politics.ox.ac.uk/our-research/time-step-away-bright-shiny-things-towards-sustainable-model-journalism-innovation-era">bright, shiny thing</a>” easily discounted and ignored. They’re extremely powerful tools in their infancy. Generative AI has the potential to be transformative to many industries, changing how we work in the classroom, the courtroom and, yes, even the newsroom. In a recently released <a href="https://www.technologyreview.com/2023/07/25/1076532/generative-ai-is-empowering-the-digital-workforce/">Adobe workforce survey</a>, 92% of respondents say AI is having a “positive impact on their work” and more than one-quarter (26%) call AI a “miracle.”</p><p>With the right prompts and guidelines, generative AI can excel at tasks like summarization, content optimization and transformation between content types. Because generative AI understands language syntax so well, the results are surprisingly good at turning a longform news article into a podcast script or a Twitter thread. And the promise of conversational search and prompt-based data analysis is exciting. In fact, when we at Taboola mapped about four dozen newsroom tasks around discovery, creation, optimization and distribution, we found generative AI had the potential to assist with more than 90% of them.</p><p>But it’s not engineered for every task. As is well documented, generative AI can be unreliable with facts and citations, subject to “<a href="https://www.nytimes.com/2023/05/01/business/ai-chatbots-hallucination.html">hallucinations</a>“. This is one of the primary reasons publishers have such different postures toward GenAI — and why it’s critical to have a methodical approach to adoption.</p><h4>Establishing Rules of Engagement</h4><p>First and foremost, it’s important you begin by articulating your company’s initial position on the use of GenAI text, images, audio and video. You have a social media handbook detailing what you allow and don’t. Generative AI requires the same care. What will you use generative AI for? Where will it never be used? Be specific, for both internal and external audiences. For example, maybe you will allow generative AI images — but only on tech or medical articles to illustrate complex and theoretical concepts. And remember, this won’t be carved in stone. The technology is quickly evolving, so be prepared to reevaluate your position and guidelines periodically.</p><p>Insider, for example, published a <a href="https://www.businessinsider.com/how-insider-newsroom-will-use-ai-2023-4">letter from the editor</a> detailing its position on AI, taking an approach leaning into experimentation and innovation. “I’ve spent many hours working with ChatGPT, and I can already tell having access to it is going to make me a better global editor-in-chief for Insider,” Nicholas Carlson <a href="https://gizmodo.com/insider-chatgpt-journalism-ai-chatbot-1850332799">said in a memo</a>. “My takeaway after a fair amount of experimentation with ChatGPT is that generative AI can make all of you better editors, reporters, and producers, too.”</p><p>The Guardian <a href="https://www.theguardian.com/help/insideguardian/2023/jun/16/the-guardians-approach-to-generative-ai#:~:text=When%20we%20use%20genAI%2C%20we,of%20time%2Dconsuming%20business%20processes.">published a similar memo</a>, but took a more cautious stance. “When we use genAI, we will focus on situations where it can improve the quality of our work, for example by helping journalists interrogate large data sets, assisting colleagues through corrections or suggestions, creating ideas for marketing campaigns, or reducing the bureaucracy of time-consuming business processes.” The New York Times was <a href="https://wan-ifra.org/2023/06/joseph-kahn-on-his-first-12-months-as-new-york-times-executive-editor/">even more conservative</a>, with Executive Editor Joe Kahn saying at the recent World News Media Congress in Taipei that the paper is “eager experimenters but extremely cautious in what we would be willing to present as a finished product to our readers.” In all cases, both employees and readers know what to expect.</p><p>One of the most common provisions for newsroom use of generative AI is external transparency. If an article or video is created or augmented by AI, it’s labeled as such. The Associated Press for years has relied on data-to-text AI to create short templated articles based on structured data like <a href="https://apnews.com/article/automated-insights-earnings-molson-coors-brewing-co-colorado-business-81435cf95a027b38771af5059da2caf8">earnings reports</a>. In all cases, they contain a tagline detailing its provenance and original source: “This story was generated by Automated Insights (http://automatedinsights.com/ap) using data from Zacks Investment Research. Access a Zacks stock report on ATRC at <a href="https://www.zacks.com/ap/ATRC">https://www.zacks.com/ap/ATRC</a>.” U.S. broadcast network Tegna, in an email newsletter, includes a disclaimer on generative AI assistance: “The subject of this email and all text below were written by GPT.” Messages like these can help set reader expectations and maintain trust, while attempting to provide a layer of insulation against errors. A few publishers like the <a href="https://www.washingtonpost.com/world/2023/08/04/novorossiysk-port-russia-feodosia-crimea/">Washington Post</a> have taken AI transparency so far that they are labeling non-algorithmically recommended areas as “hand curated” or “picked by editors.”</p><p>The News/Media Alliance <a href="https://www.newsmediaalliance.org/ai-principles/">guidance for AI</a> also calls for full disclosure: “Generative outputs should include clear and prominent attributions in a way that identifies to users the original sources of the output and encourages users to easily and directly navigate to those products, as well as to let them know when content is generated by GAI.”</p><p>Another common guardrail publishers are attaching to the nascent technology is that human review is generally required for any output seen by the public. In a business based on trust and accuracy, tools and workflows should include the opportunity for a validation step to prevent the dissemination of misinformation. Over time, as the technology improves, more end-to-end automation will absolutely be possible. But today, full automation would require some tolerance for risk.</p><h4>Setting a Development Roadmap</h4><p>Once you’ve established your framework for engagement with generative AI, the next step is to experiment with a purpose. Pick a project or two aimed directly at unlocking value or solving a problem in the newsroom. Make sure it has a reasonable development timeline — no XXL T-shirt-sized projects for your first endeavor. You want to get your hands dirty here and understand the boundaries.</p><p>Next, set very clear and measurable goals. If you are trying to optimize headlines, what’s the definition of success? Will you track adoption and usage metrics, like an increase in the number of tests? Or maybe a performance metric like CTR. And where does the product live? If it’s yet another browser tab your audience team has to have open, it might face an adoption barrier. And training a newsroom in a new product carries a cost as well beyond development hours or story points. Also make sure whatever product team or innovation squad developing generative AI tools isn’t isolated. Bottom-up suggestions on workflow efficiencies come with the benefit that the people doing the work are stakeholders in that new tool’s success.</p><p>One early adopter to AI initiatives is Forbes. The publisher’s content management system, Bertie, can <a href="https://digiday.com/media/forbes-built-a-robot-to-pre-write-articles-for-its-contributors/">recommend article topics to contributors based on their earlier offerings</a>. A politics writer would get suggestions on politics while a tech writer might be spoon-fed ideas from Silicon Valley. It also can suggest headlines and images to reduce production time.</p><p>Lastly, you aren’t in this alone. There’s a lot of good work globally being done to help education and support publishers large and small. The Center for Cooperative Media published an excellent <a href="https://medium.com/centerforcooperativemedia/new-chatgpt-handbook-helps-publishers-use-ai-for-local-news-7214582e454">guide on prompt usage and use cases for generative AI</a> specifically aimed at publishers. The London School of Economics and Political Science’s <a href="https://www.lse.ac.uk/media-and-communications/polis/JournalismAI">JournalismAI initiative</a> offers everything from classes to case studies.</p><h4>What can generative AI do for me?</h4><p>When we at Taboola approached this question, we started by roughly mapping the end-to-end production workflow in a newsroom. While all newsrooms have different roles, generally the process is the same — you get an idea for something you want to cover, you gather info, you create it, publish, and then you work hard to get an audience to that piece of journalism. Following this progression, we identified many tasks and areas where AI shows promise.</p><h4>1. Discovery tasks</h4><p>How can generative AI help a reporter find what article to write today? There are actually many ways, surprisingly.</p><p>Gen AI can help with data analysis, source identification, analyzing trending topics and brainstorming/suggesting ideas, like with Forbes. I also can imagine a system where it could even help route news tips/messages. But these tools have obvious limits. Often the most powerful articles, the ones with impact, require human intelligence and a journalist’s intuition. Generative AI won’t directly help with source development, building that one-on-one rapport with individuals, even though it might help you find them.</p><h4>2. Creation tasks</h4><p>Here’s where generative AI and conditional AI have great potential but also very clear limits. AI can help with creating articles from structured data, transcribing notes, researching topics, editing to style, packaging content, transforming content into alternative story forms and smartly sharing content across networks. With oversight and fact checking, I’ve also seen generative AI help with research and create compelling data visualizations. But generative AI can’t take photos or video from a protest scene or conduct interviews with demonstrators. Again, there’s no replacing boots on the ground.</p><h4>3. Optimization/delivery tasks</h4><p>We found this is perhaps the richest area for workflow efficiency. Generative AI can help with translating content, SEO/SEM, optimizing headlines, social shares, browser and native app alerts, email alerts and newsletters, content curation and real time data analysis.</p><p>If you take a solution-based approach, you found many areas where generative AI can help a newsroom, suggesting related links or headline options, hopefully freeing up reporters to do impactful work and not commoditized tasks.</p><h4>What happens next?</h4><p>Really, even if your newsroom does nothing with generative AI, the world is changing around you.</p><p>Google’s <a href="https://blog.google/products/search/generative-ai-search/">Search Generative Experience</a> (SGE) is very much a work in progress, but <a href="https://searchengineland.com/google-testing-links-directly-in-search-generative-experience-snapshot-answers-430075">even with recent changes highlighting more sources</a>, it will likely prove to reduce referral traffic. How impactful is that? <strong>In a new </strong>Taboola survey of 3,700 sites live from 2019, Google in the first half of 2023 accounted for 34% of all referral sessions. For many individual publishers, Google can account for half or more of all referrals. Any search erosion means a significant audience hit.</p><p>Specifically, we believe utility searches — ones related to getting an answer to a specific question — likely will drop first and fastest. Questions like “when is the World Cup?”, “what’s my horoscope?” or “Sydney travel destinations” will be answered without a click. Searches related to brands, authors and breaking news events should be more durable. E-commerce and affiliate links tied to search performance may erode, as questions like “what’s the best blender?” could be aggregated and shown without needing to visit a site. However, Google Bard currently excludes medical, legal and financial advice, so those verticals may represent a longer-term opportunity. You also probably won’t be able offset search loss by using generative AI to create content at scale. By definition, Google says <a href="https://developers.google.com/search/docs/essentials/spam-policies#automatically-generated-content">AI-created content is spam</a> and would result in demotion or removal.</p><p>In addition to search challenges, subscription-focused publishers also are likely to see headwinds. First, search traffic reliably converts, far ahead of social, for example. So any decrease in search will have impact. Second, SGE/GenAI interfaces isolate information from its source, reducing brand trust and affinity. The world may be less aware where that information came from and, long term, but less aware of how valuable you are as a publisher.</p><h4>What does Taboola recommend?</h4><p>First, experiment with using generative AI, whether you develop it yourself or use any of the thousands of tools available. This is transformational technology that isn’t going away. Use it to make the jobs of journalists easier, freeing them up from mundane tasks. Also, have some fun with it. See what it can do.</p><p>Second, some search traffic loss is inevitable, probably not in 2023 but soon. Prepare yourself by performing a risk analysis of what would happen if SGE-vulnerable search went away. Because it might.</p><p>Lastly, become a destination site. Be a source of credibility in a world of chaotic content. Publishers that are able to maintain and grow direct relationships with readers and viewers will see greater success. Use and build tools that zero the distance between you and your audience. And that’s advice that’s as good today as it will be a decade from now.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f86f4b9b228" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[What happened when we disabled Google AMP at Tribune Publishing?]]></title>
            <link>https://kurtgessler.medium.com/what-happened-when-we-disabled-google-amp-at-tribune-publishing-2fa65a2f2a40?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/2fa65a2f2a40</guid>
            <category><![CDATA[web-design]]></category>
            <category><![CDATA[web-development]]></category>
            <category><![CDATA[search-engine-optimizati]]></category>
            <category><![CDATA[google]]></category>
            <category><![CDATA[seo]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Fri, 26 Aug 2022 06:03:57 GMT</pubDate>
            <atom:updated>2022-08-26T15:31:21.562Z</atom:updated>
            <content:encoded><![CDATA[<h3>What happened when we disabled Google AMP at Tribune Publishing? Shockingly little. So you should try it, too</h3><p>Ahead of its June 2021 algorithm update, Google made a significant announcement as part of its shift to emphasize overall page experience in determining its search ranking — it was lifting its thumb off the scales in favor of Accelerated Mobile Pages (AMP).</p><p>AMP is a Google-created open source HTML framework featuring lean pages optimized to be more performant on mobile devices. While encouraging good behavior, it also limited design and homogenized the content-level page experience. First appearing in 2016, AMP pages were quickly rolled out by most U.S. news publishers, incentivized by not just snappy load times, but by Google locking non-AMP pages out of key mobile search positions, like the “Top Stories” carousel.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/524/1*Pbhv1NYFyE_IgHZmN9EVNQ.png" /></figure><p>Even Twitter in 2017 linked to AMP pages from iOS and Android apps. AMP, like Facebook Instant Articles, was everywhere, becoming an offer many publishers couldn’t refuse.</p><p>But in late May 2021, amid continued complaints of loss of control over user experience and subscription integration nightmares, <a href="https://developers.google.com/search/blog/2020/05/evaluating-page-experience">Google said they were ending that AMP mandate</a> in its June 2021 Core Algorithm Update:</p><blockquote>“As part of this update, we’ll also incorporate the page experience metrics into our ranking criteria for the Top Stories feature in Search on mobile, and remove the AMP requirement from Top Stories eligibility.”</blockquote><p>Instead, Google would rely on its now ubiquitous <a href="https://developers.google.com/search/docs/advanced/experience/page-experience">Core Web Vitals metrics</a> and related signals. Five months later, <a href="https://www.theverge.com/2021/11/19/22791002/twitter-amp-ios-android">Twitter phased out support of AMP</a>. Around this time, we at Tribune Publishing made the decision to phase out AMP as we rolled out a redesign of our content-level pages in spring of 2022.</p><p>As search was by far our largest referral source to all our sites, including the Chicago Tribune, Baltimore Sun and New York Daily News — traffic that was a reliable source of subscriptions — this move was not without risk. Plus, in coordinating this with a redesign and codebase change — a situation which often has a detrimental impact on search—we could take a bad search situation and make it far, far worse. Also in the background, exacerbating our fears, was a significant erosion in search traffic off of record COVID highs. The needle was already pointed down.</p><p>But if it were a success, we’d not only fully control the on-page experience, but we wouldn’t be saddled with the tech debt of maintaining and updating AMP, an extremely attractive proposition, as <a href="https://searchengineland.com/amp-wont-be-required-for-googles-top-stories-section-335276">Barry Schwartz noted in Search Engine Land</a>. And most importantly, we know our mobile web pages are better than AMP at converting visitors into subscribers.</p><h4>THE CUTOVERS</h4><p>The first site to make the leap was the Baltimore Sun. We turned off AMP and cut over pages to the redesign by sections over four weeks starting in early March. Prior to the cutover, about 68% of daily mobile search users coming to the Sun were hitting AMP — a fairly large percentage.</p><p>For our data, we’ll use Google Analytics referral source and page type reports, filtered by device as needed. To clarify a question raised by Damon Kiesow shortly after publication, each line is inclusive. Red is total Google search users, of which mobile search users (yellow) is a subset, of which AMP users (blue) are a subset of that.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*osy9mM03zdK8XMeuSCngFQ.png" /></figure><p>As the chart shows, despite some general overall attrition in search, users hitting AMP gracefully transitioned to mobile web search pages. Overall, from Jan. 1 to July 31, median daily search users fell only 12.41% post AMP, with most of those losses attributed to yearly highs in January.</p><p>We disabled AMP and cut over to the redesign on the South Florida Sun Sentinel April 11. Unlike Baltimore, turning off AMP here and for future sites was a one-day event. South Florida was less reliant on AMP, with only 39.8% of its mobile search users hitting the framework.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QSXa2xmp0POU2tySDgicOA.png" /></figure><p>Much like Baltimore, when AMP stopped, mobile web search picked up the slack. Median daily mobile search users fell only 16.23% post AMP, although we saw high levels of volatility throughout.</p><p>AMP was disabled on the Chicago Tribune days later on April 15. The Trib had the AMP lowest usage, with only 37.5% of its mobile search users being served AMP pages. But it also had the highest levels of overall search referrals.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2oVa0i_hbf7F_KNjHNqKhQ.png" /></figure><p>Chicago also proved extremely resilient, seeing an 18.15% decline in median daily mobile search users post AMP, mostly due to the general search decline.</p><p>Lastly, one week later on April 22, we disabled AMP on the New York Daily News. This was the riskiest proposition in that 73.4% of mobile search users were being served an AMP page. A greater percentage of the Daily News’ overall search traffic also is mobile vs. desktop.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YmNY99ZyItcEKQMpJkkFaw.png" /></figure><p>Looking at the trend, you can see how aligned mobile search users and AMP users were. Despite walking lockstep, the Daily News transitioned off of AMP well, declining 26.94% in median daily mobile search users. And like Baltimore and Chicago, most of this change is centered at the beginning of the year. You’d see an almost horizontal trend line from March to July.</p><h4>THE RESULTS</h4><p>Given all the factors, it’s impossible to make a truly clean search impact analysis. In addition to retiring AMP, we had a redesign and a codebase change. We’re at the tail end of a long organic search decline from the pandemic. How much of a factor was <a href="https://www.pewresearch.org/fact-tank/2020/02/26/almost-seven-in-ten-americans-have-news-fatigue-more-among-republicans/">news fatigue</a> on news-centric sites? And while mostly stable, there were fewer articles being published each month in 2022 post AMP on some of our sites—a factor which could meaningfully impact overall search referrals.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pe-_Fvl0H6SNx5WgEbgxpA.png" /></figure><p>Despite all these variances, all of them working <em>against</em> search success, Tribune Publishing generally saw very manageable declines in mobile search referrals post AMP. The search drop generally fell within the accepted near-term range of a standard redesign, and on the macro level, year-long drops mostly aligned to broader search trends.</p><p>Given the higher page RPMs and subscriber conversion rates of a non-AMP page, pulling the plug on AMP looks like an easy win for both programmatic and consumer revenue. And most importantly, we regain full control of the user experience. And that’s perhaps the biggest upside.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2fa65a2f2a40" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Facebook’s Breaking News Indicator valuable despite rumors of hidden risks, post cannibalization]]></title>
            <link>https://kurtgessler.medium.com/facebooks-breaking-news-indicator-valuable-despite-rumors-of-hidden-risks-post-cannibalization-8559cf5514c3?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/8559cf5514c3</guid>
            <category><![CDATA[metrics]]></category>
            <category><![CDATA[facebook]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[metrics-and-analytics]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Wed, 30 Jun 2021 18:10:46 GMT</pubDate>
            <atom:updated>2021-07-28T00:52:31.089Z</atom:updated>
            <content:encoded><![CDATA[<h3>Facebook’s Breaking News Indicator valuable to publishers despite rumors of hidden risks, post cannibalization</h3><p>Three years into Facebook’s attempt to take a bite from Twitter’s breaking news empire, there exists very little good data on the social media’s giant’s most visible effort — the Breaking News Indicator.</p><p>Facebook’s Breaking News Indicator is a bright red label that certain U.S. and global publishers can apply to a News Feed post, highlighting a general sense of urgency around its topic. The Breaking News Indicator was <a href="https://marketingland.com/facebook-opens-breaking-news-label-50-publishers-north-america-latin-america-europe-australia-235475">tested in late 2017</a> and more <a href="https://techcrunch.com/2018/03/05/facebook-breaking-news/">broadly expanded in March 2018</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9FgYiQcr0ZDMwPOFIGexIw.jpeg" /></figure><p>The rules of engagement around the Breaking News Indicator are that most publishers can use one per day plus a slush fund of five extra per month in cases where unexpected critical news needs to be conveyed. The label <a href="https://www.facebook.com/business/help/221476155366376">can be set</a> depending on the expected duration of the news, displaying for 30 minutes to six hours and can be manually extended.</p><p>Most importantly, according to Facebook, the label by design has no impact on News Feed prioritization. Any engagement benefits are from the visual cue — not a thumb on the scale of Facebook’s algorithm. According to Facebook’s <a href="https://www.facebook.com/journalismproject/facebook-breaking-news-label">six-week alpha test with the Washington Post</a>, posts with the Breaking News Indicator saw a 4% lift in click-through rate, 7% lift in likes, 4% lift in comments and an 11% lift in shares. Later <a href="https://www.facebook.com/journalismproject/facebook-breaking-news-label-expands">studies by Facebook</a> suggested “breaking news posts about politics, crime, disaster, and business perform best.”</p><p>However, even <a href="https://www.facebook.com/journalismproject/facebook-breaking-news-label-expands">Facebook qualifies those results</a> and potential upside: “We cannot guarantee that use of the Breaking News Indicator will result in a distribution boost.”</p><h4>THE CASE AGAINST</h4><p>One of the few <a href="https://www.echobox.com/resources/white-papers/should-publishers-use-facebooks-breaking-news-label/">studies</a> on Facebook’s Breaking News Indicator was conducted by Echobox in 2019, and it’s often cited as a reason against full adoption. The chief takeaway is that while the label does appear to be effective in boosting engagement around a post with the Breaking News Indicator, the study suggests that the engagement boost on that individual post didn’t translate to any overall page-level performance increases.</p><p>Reporting on the study for<a href="https://www.socialmediatoday.com/news/should-publishers-use-facebooks-breaking-news-label/551966/"> Social Media Today</a>, Sebastian Huempfer wrote: “The root cause is most likely that shares for which the Breaking News Label was not used performed relatively worse than they could have — it seems likely that this is at least partly a symptom of an underlying cannibalization effect.”</p><p>“In summary, while the Breaking News label evidently generates increased viewership for breaking stories, the findings show that it also decreases traffic to other posts, and therefore does not raise overall traffic from social media.”</p><p>I think there are a few problems with that cannibalization hypothesis if we assume Facebook is telling the truth that the tag has no impact on the News Feed algorithm. And Facebook assures me we should believe them.</p><p>First, due to the frequency limitations, any benefit is going to be small. If your page shares 40 posts a day and two have a Breaking News label getting the aforementioned modest engagement boost, that might not move the needle that week, let alone that month. It may be like a muon — it’s there, but that particle is small and difficult to detect.</p><p>Second, without true multivariate testing available to us — serving simultaneously the post with the label and the post without the label — it’s a degree of guesswork. What if you’re putting the label on posts that would have overperformed anyways? Lastly, we know the content of the article, its timeliness and the quality of the post matters most. A boring post to a procedural story is unlikely to overperform on engagement KPIs just because it has a red label.</p><p>So given all that, how can we assess the value and risks, if any, of the Breaking News Indicator?</p><h4>KEY QUESTIONS</h4><p>The per-share engagement question seems fairly straightforward to hone in on. The questions seem to be:</p><ul><li>Does a post with a Breaking News label get more interactions than ones without a label?</li><li>Does a post with a Breaking News label reach more people and get more clicks than ones without a label?</li></ul><p>In a subscription-focused business model, referral traffic is critical so I wanted to expand the analysis beyond just comments, likes and shares and tackle reach and link clicks.</p><p>The cannibalization question is more interesting and, honestly, harder to pin down. I’m going to approach it by trying to answer these questions:</p><ul><li>Did we see gains in interactions on the days when we used 1 or more Breaking News labels, or were the gains in those posts offset by losses in others?</li><li>Does the use of the Breaking News label on a post negatively affect posts immediately before or after that share?</li></ul><p>For the first two questions, since the default Breaking News indicator guideline is daily, it seems fair to incorporate that into our analysis. If the Breaking News label adds value, we would expect to have more total daily interactions vs. a day when we don’t use it — even if only by a little. And if we use the label more than once in a day, the likelihood of exceeding the average/median should further increase.</p><p>For the last question, I want to limit the scope to posts immediately before and after a Breaking News-labeled post, because it should not affect a post 18 hours later. If it’s going to have an impact, we probably should see it most acutely in that immediate time vicinity, stealing traffic from the post right before or right after the Breaking News post was shared.</p><p>For our dataset, Tribune Publishing has four markets with the Breaking News Indicator enabled — the Chicago Tribune, New York Daily News, South Florida Sun Sentinel and the Hartford Courant. And we’ll look at every post in 2020 to get a sample size of around 50,000 total posts, hopefully minimizing the effect of outlier social data. It’s also a good range of page sizes, from 143,000 likes to 3.2 million. The only wrinkle is that in 2020, Facebook loosened restrictions during COVID-19 outbreak, allowing twice the posts to be labeled Breaking News each month during the worst of the pandemic, I’m told. However, this could prove beneficial in the end by increasing our sample size a bit.</p><h4>TRIBUNE PUBLISHING ENGAGEMENT DATA</h4><p>To get the data to address the questions, CrowdTangle and Facebook Insights should be plenty. CrowdTangle has the useful function of logging all posts that had a Breaking News label, even after expiration. An important note, however: Posts with a Breaking News label from one account shared by another account can track falsely as having used the Breaking News Indicator by the account that reshared the post, so you have to be careful.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/664/1*HMuhaFno-fcFa5HYxuRWeg.png" /></figure><p>The usage data for 2020 was pretty easy to pull together. And we avoided a situation I was dreading. All of the 1,150 Tribune Publishing posts with a Breaking News Indicator in 2020 were on link posts, not video or photo posts. As engagement can vary by post type, we can limit our analysis to just link posts with the Breaking News Indicator and all link posts, excising photo and video posts.</p><p>Looking at the data, the Daily News (389) and the Sun Sentinel (419) more aggressively used the label, around one per day, but the Sun Sentinel did it in about half the link posts. The Tribune (212) used one every other day and the Courant (130) one every three days. The team at CrowdTangle confirmed my data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fiN3UhtLyFg2OO_H-cXECA.png" /></figure><p>As a percentage of total posts, the shares with the Breaking News label represented between 1.66% and 3.22% of total link posts in 2020 — so a fairly small percentage of total volume, even in the more aggressive cases.</p><p>Looking at a year’s worth of link posts from CrowdTangle, we can get average and median post data. Daily News has about 1,000 interactions per post, Tribune 500 and Sun Sentinel and Courant around 100.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IzSZj_WDGzugvy6DM5PyUA.png" /></figure><p>Looking at just posts with the Breaking News label we see much higher average and median interactions, from 3x to 20x.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WnyR5MENI2n8ts8vDvR7gg.png" /></figure><p>The pages with the most fans still have the most interactions, but it’s interesting to see the highest gains at the Courant, which has the fewest page likes, fewest link posts and fewest Breaking News posts.</p><p>Percentage wise, the difference between the Breaking News post and all posts is pretty gaudy — but again, there’s many factors, including obviously the quality of the story being shared and the quality of the post itself.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fY-D_GWHkeUp-iieH2iVQA.png" /></figure><p>But while not a cause, the above data align with Facebook’s data that show increased engagement with Breaking News-labeled posts. The same largely holds true with reach and link clicks, as shown by two per-post datasets I grabbed from Facebook Insights.</p><p>I laid Chicago Tribune’s data out by month to isolate any seasonal or news-based fluctuations. We saw average per-post link clicks range from 1,200 to 2,000 and median from about 400 to nearly 1,000. Average monthly per-post reach was in the mid-40,000s for much of the year with a monthly median around 30,000.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WlvuvGmScBDyA9D2E8DZUQ.png" /></figure><p>But looking at just the posts labeled Breaking News, we saw average per-post reach was higher in 11 of the 12 months and monthly median per-post reach was higher all year. The same pattern held with average and median monthly per-post link clicks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DQN0PYOLiEbFpkNAcJ9PWA.png" /></figure><p>Again, the results fall into line with Facebook’s data that suggest an increase in link clicks as well. In our case, we saw a healthy margin of 100%, 200% and even more than 400% in two of our months.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*XuZc4kEsqQtMcKcdTKR6ng.png" /></figure><p>So, whatever the cause, we’re clearly seeing per-post engagement track above the median and above average on posts with a Breaking News Indicator on a fairly decent dataset.</p><p>But that wasn’t really the core question, although it was valuable to validate. The central premise to probe is the thought that while the Breaking News label may help an individual post, it doesn’t help overall because it may hurt other posts. And that’s a tougher concept to approach.</p><h4>MARKET(ING) CANNIBALIZATION</h4><p>The most straightforward way to approach the first of our cannibalization questions is to look at the sum of daily interactions. Our markets have an extremely steady cadence when it comes to posting on Facebook.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Cble5jBTucKBOFguyOzsHw.png" /></figure><p>Given the steadiness of the post cadence, roughly 1 an hour, every day, all year for the Tribune, we should be able to address our first question — did we see gains in interactions on the days when we used 1 or more Breaking News labels, or were the gains given by Breaking News posts offset by losses in others? In other words, did those Breaking News posts cannibalize engagement to the extent that we saw no meaningful difference between a day where we didn’t apply a Breaking News Indicator to a post and a day where we used one or more Breaking News labels.</p><p>To set up the daily data in Excel, I downloaded a CrowdTangle interaction report by day and added a merged column for the number of Breaking News Indicators on that day, with 0 representing that none were used. Then I created another new column adding text if the label was used: =IF(G289&gt;0, “BRK TAG USED”,”NO BRK TAG”). Once I calculated the average and median interactions per day, I added two more columns checking if that interaction total was either below or above those daily totals. No day fell exactly on the average or adjusted median.</p><p>=IF(M289&gt;$M$369, “ABOVE AVG”,”BELOW AVG”)</p><p>=IF(M289&gt;$M$370, “ABOVE MED”,”BELOW MED”)</p><p>Then I tallied the number of days where the tag was used and if it was above or below the average or median. Here’s an example of the counters for averages.</p><p>=COUNTIFS(N2:N367,”ABOVE AVG”,P2:P367,”BRK TAG USED”)</p><p>=COUNTIFS(N2:N367,”BELOW AVG”,P2:P367,”BRK TAG USED”)</p><p>In 2020, the Tribune saw an average of 12,487.93 interactions per day and a median of 11,320 interactions per day. The Tribune applied the Breaking News label to at least one post on 170 different days in 2020. For 93 of those 170 days, or 54.7%, the Tribune had above average interactions levels. For 109 of the 170 days, 64.1%, interactions exceeded the daily median. So in both cases, average or median, the Tribune was more likely to exceed the typical number of daily interactions when the Breaking News Indicator was used on that day.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xDOau190jCmobZxt9TvmqQ.png" /></figure><p>The same was true for the Sun Sentinel. In 2020, the Sun Sentinel saw an average of 4,405.12 daily interactions and a daily median of 3,797. The Sun Sentinel used the Breaking News label on at least one post on 182 different days in 2020. For 115 of those 182 days, or 63.1%, the Sun Sentinel had above average interactions levels. For 126 of the 182 days, 69.3%, interactions were higher than the total daily median.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CtVLq9QOvUpKJY9EqqAB1w.png" /></figure><p>In both cases, we do see correlation between higher daily engagement and the use of the Breaking News Indicator. While that certainly isn’t causation, if there were a hidden cannibalization risk, we would expect to see those usage days to track more toward the average or median — not roughly two-thirds of the days when it was used to perform above normal engagement levels.</p><p>The second way to explore the cannibalization theory was attempting to quantify if the use of the Breaking News label on a post negatively affects posts immediately before or after that post. Basically, does that label cannibalize engagement from other posts, in essence, using up all the oxygen in the room?</p><p>I picked the posts immediately before and after because, in theory, those should be more susceptible to influence. Maybe it cuts short a post’s reach/exposure in favor of a new post with the Breaking News Indicator, or maybe it doesn’t surface as energetically a brand new post following a post with a Breaking News label. Again, not being able to imagine a reason a post published at 5 a.m. should have any bearing on a post at 8 p.m., I started with those immediately adjacent.</p><p>To test this theory, I used the same year’s worth of data from CrowdTangle and added a column marking a post as “pre brk,” “brk”, “post brk” or “pre brk post brk” and screened the rows. And the initial results were anecdotally intriguing. There did appear to be an interaction spike around Breaking News labels.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/846/1*U0YekjD5hK1BfLp6QZyTAg.gif" /></figure><p>Then I used an Excel sumproduct function to count each category.</p><p>=SUMPRODUCT(( — EXACT(“brk”,N2:N8708)))</p><p>=SUMPRODUCT(( — EXACT(“pre brk”,N2:N8708)))</p><p>=SUMPRODUCT(( — EXACT(“post brk”,N2:N8708)))</p><p>=SUMPRODUCT(( — EXACT(“pre brk post brk”,N2:N8708)))</p><p>The data show that we had those 212 Breaking News-labeled posts. However, due to timing irregularities, there weren’t 212 before and after. Sometimes we saw sequential breaking posts, or a post that both followed a Breaking News post and preceded another. In the end, we had 196 posts that preceded a Breaking News post, 196 that followed one and 3 that both preceded and followed.</p><p>Then I tallied the interactions and created averages for all four categories, looking for the labels I added in Column N and summing the total interactions in Column O.</p><p>=SUMIF($N$2:$N$8708, “pre brk”, $O$2:$O$8708)</p><p>=SUMIF($N$2:$N$8708, “pre brk”, $O$2:$O$8708)</p><p>=SUMIF($N$2:$N$8708, “brk”, $O$2:$O$8708)</p><p>=SUMIF($N$2:$N$8708, “pre brk post brk”, $O$2:$O$8708)</p><p>However, looking at average interactions in all four categories for an entire year’s worth of Breaking News Indicator data let the air out of the cannibalization balloon.</p><p>We still saw 2,139.59 interactions per Breaking News link post and average per-post interactions on all link posts at 523.05. These matched earlier data. But now we also have posts immediately before a Breaking News label averaging 536.58 interactions, posts following a Breaking News label averaging 569.60 interactions and our 3 posts serving as both before and after averaging 642.33 interactions. So there doesn’t appear to be any engagement loss in shares with adjacency to a Breaking News post. In fact, all are slightly above average.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rKv34fjr2XzFV8qWOP3xtg.png" /></figure><p>And when I looked closer at the Excel file, there were examples of unengaging Breaking News posts and posts in their proximity which outperformed them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/846/1*JKUgTYB_tJazCuh7hsUINg.gif" /></figure><p>In the end, I couldn’t find any conclusive data to support the cannibalization hypothesis.</p><h4>FINAL VERDICT</h4><p>Does Facebook’s Breaking News Indicator give an engagement boost to shares? Facebook’s tests say it does and it appears to correlate with higher performance in likes, shares, comments, reach and clicks in all of Tribune Publishing’s markets that have it.</p><p>Is there a hidden downside, where it potentially adversely affects other posts? I couldn’t find evidence of it. There are more good days than bad when the label is used. And the shares in the proximity of a Breaking News post perform right around the average.</p><p>Facebook’s Breaking News Indicator appears to function exactly as advertised. It’s a tool publishers can use to help readers quickly identify time-sensitive news and information. And given the delta over typical engagement, readers seem to recognize and respond to it. However, its usage is capped. So if you post dozens of times each day, it’s unlikely to move the engagement meter in a significant way once you start collecting and analyzing months of data.</p><p>That said, there does appear to be measurable gains without any quantifiable risk. Given that Facebook already throttles usage, publishers should use it as aggressively as the news cycle merits.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8559cf5514c3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Twitter length study: Do longer tweets drive more engagement and referral traffic?]]></title>
            <link>https://kurtgessler.medium.com/twitter-length-study-do-longer-tweets-drive-more-engagement-and-referral-traffic-3dd0781363ff?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/3dd0781363ff</guid>
            <category><![CDATA[twitter]]></category>
            <category><![CDATA[social-media]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Wed, 08 Jan 2020 00:22:29 GMT</pubDate>
            <atom:updated>2020-01-08T00:22:29.193Z</atom:updated>
            <content:encoded><![CDATA[<p>Since Twitter’s expansion to 280 characters about <a href="https://techcrunch.com/2017/11/07/twitter-officially-expands-its-character-count-to-280-starting-today/">two years ago</a>, it’s been loosely acknowledged that tweets in the expanded range have somewhat higher engagement.</p><p>Reports have varied on how much of a boost this has provided. Social media optimization platform SocialFlow said <a href="http://www.socialflow.com/users-engage-with-longer-tweets-data-shows/">tweets in the expanded range</a> had double the retweets and likes but a small decline in link clicks. Social media analytics firm Sotrender <a href="https://www.sotrender.com/blog/2018/10/shorter-longer-tweets-one-year-expanded-character-limit-analysis/">also reported</a> tweets in the expanded 141- to 280-character range had higher engagement.</p><p>Twitter verified to me in October the observation that longer tweets anecdotally were more performant than shorter ones. That extra space, they said, provided more room for people to express themselves and be creative.</p><p>In analyzing one year of Twitter Analytics data (23,732 tweets from the Chicago Tribune’s primary <a href="https://twitter.com/chicagotribune">Twitter account</a> between Sept. 1, 2018, and Aug. 31, 2019), we also saw this trend— but it wasn’t that dramatic at first pass.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YxGUjx_TK3kfyTfK-Fcceg.png" /></figure><p>The data show a 40% increase in average retweets and a 34.8% increase in average replies between the 1–140 character range and the 141–280 range. But not only are those percentages somewhat inflated by the low totals, like 4.31 replies, but other categories like url clicks only show a modest 2.6% difference. And really, 5.81 average replies for a long tweet is only .7 replies over the total average of 5.11. The engagement increase provided by the extra length didn’t feel consequential here.</p><p>However, the benefits of a little extra texture in your tweet may be far more profound. When you start to slice the numbers into manageable ranges, not just short and long, a more nuanced pattern emerges.</p><p>The Tribune data suggest that, in general, as the tweets gradually get longer, engagement sharply increases. And at the very longest range, we’ve seen the highest levels of engagement, impressions and link clicks.</p><p>Conversely, the shortest tweets are among least liked, least retweeted and least seen.</p><p>To illustrate this, I divided that Tribune sample of approximately 24,000 tweets into buckets of 20 characters each, reserving a dedicated column for anything 40 or fewer characters. The total count distribution was as follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*SeV62oUBD4fNsuTbjOe6-g.png" /></figure><p>As you can see, we had a limited number of the very longest and very shortest — enough to exclude those categories from the discussion since a single tweet could skew the average when your sample size is 31. So we’ll focus on the average engagement of tweets from 40 characters to 259 characters — effectively any range with at least a 500-tweet sample size (roughly 2% of our dataset).</p><p>Rather than share a screenshot of a conditionally formatted pivot table, I’ll break down each engagement type into its own bar chart to illustrate these findings. In pretty much every case, the difference between a less detailed tweet and a more detailed tweet becomes apparent.</p><p>The most common Twitter engagement, the like, strongly favors the longest tweets. Four of the top five character count categories (120–139, 160–179, 200–219, 220–239 and 240–259) fall into the expanded character count range. And the longest tweets have more than double the likes of the shortest three ranges.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5iexf-g5GliuvXArF2fBSA.png" /></figure><p>Average replies follow a similar but even more obvious pattern. As the character count gets longer, average replies increase. Here we see a 145.45% gap between shortest and longest. Also note that the five longest ranges also have the five highest average replies.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Hxv5pHCcQgote-JRZbSAzw.png" /></figure><p>It’s the same story in retweets. Only we see the broadest engagement gap between the longest and the shortest tweets, 21.32 retweets for the longest range compared with 7.38 for the shortest, about 189%.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*qPvs7DM1j6ht7-9miwDO-Q.png" /></figure><p>Not surprisingly, more engagement like replies and retweets translated into more per-tweet visibility, reflected in an increase in average impressions. The shortest three character count ranges had the lowest average impressions and, again, we see the five longest ranges also have the five highest average impressions. The longest range of tweets (240–259) is 33.86% more viewed than the shortest (40–59), about 6,000 more average impressions per tweet. That total could add up fast.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*YLx8XvT9FH_vR3v3mMe_9g.png" /></figure><p>Link clicks in this method don’t yield that classic Price is Right Cliffhangers pattern, but four of the top five categories (100–119, 140–159, 160–179, 220–239 and 240–259) fall into the expanded character count range. And again, the longest range has the highest average link clicks at 194.72 per tweet.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6sxCM421acAZYCa5n1kDng.png" /></figure><p>Lastly, even profile views — a more frequent path to a new follower than directly from a tweet — significantly favor longer tweets. The longest character count range of tweets have 59.96% more profile clicks than the shortest, from 4.82 to 7.87.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bkv7cbQzgQRoLjrsvr69_Q.png" /></figure><p>The reasons behind these engagement numbers are best illustrated through a few examples of some successful Chicago Tribune tweets — not surprisingly all from the top character count range.</p><h3>Chicago Tribune on Twitter</h3><p>If you hated the ending to HBO&#39;s &quot;Game of Thrones,&quot; don&#39;t worry. Author George R.R. Martin said the show&#39;s ending won&#39;t affect the final two novels. But he&#39;s not worried about fan reaction. &quot;You can&#39;t please everybody, so you&#39;ve got to please yourself.&quot; https://t.co/TGzeHY2JNi</p><p>This 254-character tweet had 1,022,644 impressions, 25,636 link clicks, 1,242 retweets, 153 replies and 4,616 likes.</p><h3>Chicago Tribune on Twitter</h3><p>In 2007, a teenager was photographed in the Chicago Tribune holding a radar gun with a Skokie police officer. Now 26, she is applying to join the Skokie Police Department herself - and recreated the photo with the officer who inspired her. https://t.co/FKeV3ANS3P</p><p>This 241-character tweet had 135,128 impressions, 594 link clicks, 108 retweets, 31 replies and 158 likes.</p><h3>Chicago Tribune on Twitter</h3><p>Can ICE raid private homes? Can people be detained without a deportation order? Can someone seek sanctuary in a church? Questions and answers about what could happen if U.S. Immigration and Customs Enforcement comes to your Chicago neighborhood. https://t.co/h4RxGFA3bZ</p><p>This 247-character tweet had 312,832 impressions, 9,009 link clicks, 868 retweets, 127 replies and 900 likes.</p><p>In all of these cases, each tweet provides more detail than conveyed by just the headline. It’s an extra quote and context in the Game of Throne tweet (mostly George R.R. Martin bracing you for disappointment). The Skokie Police Department radar gun tweet sets up the text and images for a perfect then and now display. And the ICE tweet pulls out some of the specific questions the story answers — a much broader way to tease the article’s content. None of these are ground-breaking tweets. It’s all time-tested social strategy. But you need the extra space to execute these approaches and not just echo the headline already visible in the Twitter card.</p><p>So there’s not a magical threshold of 141 characters. The difference between 125 characters and 150 characters appears to be small. But the difference between 60 characters and 240 characters looks to be statistically significant. Not simply numerically, per se, but informationally. The extra caring and detail — more context, a quote, a fact or great turn of phrase — result in more engagement, more impressions and often more referral traffic.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=3dd0781363ff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Are you posting the same story to Facebook multiple times? You’re probably wasting your time]]></title>
            <link>https://kurtgessler.medium.com/are-you-posting-the-same-story-to-facebook-multiple-times-youre-probably-wasting-your-time-d19264ce1d?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/d19264ce1d</guid>
            <category><![CDATA[social-media-marketing]]></category>
            <category><![CDATA[digital-marketing]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[facebook]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Wed, 14 Nov 2018 14:54:07 GMT</pubDate>
            <atom:updated>2018-11-14T16:59:39.626Z</atom:updated>
            <content:encoded><![CDATA[<p>It’s often called “recirculation” in Facebook parlance, or “strategically reposting” older stories. You know, putting stuff on Facebook that you already published there days, weeks or even months ago.</p><p>The attraction for digital publishers is simple: You boost, in theory, both Facebook exposure and referral traffic with almost no additional cost. It’s a fairly philistine strategy espoused by self-proclaimed brand strategists and vendors of social media tools, usually in pitches about efficiently maximizing audience opportunities.</p><p>The problem? It’s an approach to Facebook that objectively failed at the Chicago Tribune, according to internal analysis. And if you study your own data, you’ll probably find it isn’t working for you, either.</p><p>Here’s a good example of the results we saw when we reposted previously shared stories:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*LmHXHIL0GyNU1dRUYvKAwg.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*RldCSkm6CGwHf1Vhw9D6GA.png" /><figcaption>Second post</figcaption></figure><p>It’s a compelling piece about cotillions in Evanston, one of Chicago’s North Shore suburbs. Each post had a unique approach, with very different text and photos. The first and second posts were separated by two days, one on a weekday and one on a Sunday. In this example, the first post had a reach of 96,703 and the second post had a reach of only 5,614 — even though they linked to the same article. But maybe that first share was just more engaging.</p><p>So here’s another:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/468/1*RckqNNezK7OLIKXIH8j8pg.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/463/1*y61lU98WUlwjoErQj7Om1w.png" /><figcaption>Second post</figcaption></figure><p>Here we have a consumer issue with tech overtones, so we’d expect some success. The two posts have the same photo but different text. They were four days apart. Again the first share reached far more people.</p><p>One more example:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/465/1*FTClU9VRS_MXdNpnUGXcEg.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/465/1*w3O8e9OZWS7wuDOCnKEpWg.png" /><figcaption>Second post</figcaption></figure><p>Here it’s the same photo <em>and</em> the same text. Despite those similarities, we see dramatically different results, with the second share of the same story getting a fraction of the reach and just a single like.</p><p>In looking at a full year of Chicago Tribune Facebook data, we saw this trend repeated hundreds of times. All told, more than 88 percent of second shares of stories published on Facebook as link posts had a lower reach than the first post. And it wasn’t slightly lower. The median loss of reach was more than 70 percent. So if that first share reached 50,000 people, only 15,000 people saw that second share. It’s a trend we had anecdotally observed at the Tribune but never proved out until now.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/459/1*JNocPVMaV73iFQe8hgJmGQ.gif" /><figcaption>First, second and third posts. Average, low and lower.</figcaption></figure><p>Now, there’s a lot of background here that’s critical in understanding the scope of this analysis — which is actually fairly narrow. So get ready for a notebook dump.</p><p>But I want to hit two points before we get too deep in the weeds. The first is that as much as we can control for variables, the quality of a Facebook post is obviously a factor. A really clever second post can outperform a first post. It’s social media. You get points for style. However, it’s clear the trend at the Tribune was biased against that success.</p><p>The second is that there are viable reasons you might want to post the same story link multiple times to Facebook despite the forecast. Maybe it’s important investigative journalism you want seen by as many people as possible, even if the number of people diminishes each time.</p><p>My point in publishing this data isn’t necessarily to discourage you from sharing the same story link multiple times. It’s to give you insight into the results based on what we’ve seen at the Tribune. And I’m hoping it’s specific enough to help anyone replicate this study as well.</p><h3>The Process</h3><p>For the bulk of this analysis, the goal was to measure the outcome of resharing stories. And by stories, I’m specifically referring to link posts to articles on a publisher’s website. For most traditional digital media, this is the majority of Facebook shares.</p><p>We know Facebook’s News Feed algorithm treats post type (photo, video, live video, link) as a signal <a href="https://blog.hootsuite.com/facebook-algorithm/">to some degree</a> in matching audience and content. So comparing a photo post (maybe with a Bitly link to a photo gallery) to that same photo gallery shared as a link post wouldn’t be an apples-to-apples comparison. For the cleanest possible look at how the algorithm reacts to second or third shares of the same link, you must keep post type consistent. In our case, link post to link post.</p><p>To construct a viable dataset, I downloaded all of the 2017 data from CrowdTangle and Facebook Insights (in chunks, as always. Fix this). Turns out, the Tribune had around 10,000 posts in 2017. I then compiled it into one Excel document that included created time, post type, Facebook URL, link URL, engagement, reach, link clicks, message, link text and description.</p><p>The objective was to check for matching link URLs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*lU4_PQl2RKc8wD8LUx5wmw.jpeg" /></figure><p>If a link URL appeared more than once, that means the Chicago Tribune shared that link multiple times. We could then filter by date (to determine order) and post type.</p><p>My first disappointment was that I found many URLs contained UTM codes for tracking, cache busting and navigation. This meant duplicate shares were sorting as unique URLs, such as this <a href="http://www.chicagotribune.com/sports/baseball/cubs/ct-cubs-world-series-steve-bartman-20170731-story.html?s">http://www.chicagotribune.com/sports/baseball/cubs/ct-cubs-world-series-steve-bartman-20170731-story.html?s</a> or <a href="http://www.chicagotribune.com/news/local/politics/ct-donald-trump-voting-commission-illinois-met-0708-20170707-story.html#nt=oft02a-2li3">http://www.chicagotribune.com/news/local/politics/ct-donald-trump-voting-commission-illinois-met-0708-20170707-story.html#nt=oft02a-2li3</a>.</p><p>Newsroom developer Acton Gorton ran a regex script to strip the extraneous data from the URLs. From here, I could sort by URL and find the matches. It turned out, there were a little more than 1,000 instances of the same link being shared in 2017. This was higher than I anticipated.</p><p>After sampling some of this data, I also determined that a few posts in that sample of roughly 1,000 had been created and deleted. The links went to dead pages. Perhaps a second share was removed after someone noticed it had been posted earlier. Those posts, however, still had data, albeit tiny. In response, I had to test each of the 1,000 shared URLs to make sure the second post didn’t underperform that first one because the second only existed for a few minutes. I ended up excluding six pairs of posts where one of the posts was deleted from the final data.</p><p>Then I built two pivot tables. One was around URL, broken down by created time with columns for type by reach. Then I built another around URL, broken down by created time with columns for type by link clicks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OWKrauuiDcx2rB8Emc5tEg.jpeg" /></figure><p>After that, it was just sort, calculate and compile.</p><h3><strong>The Data</strong></h3><p>As Facebook metrics are always in motion, all the data were pulled on Sept. 7, 2018. The tale of the tape is as follows:</p><ul><li>The Chicago Tribune shared 9,699 posts in 2017.</li><li>There were 1,031 total instances of the same link shared in 2017.</li><li>Twelve posts were excluded because either the first or second post was deleted for unknown workflow reasons.</li><li>72 posts were excluded because they didn’t meet our core requirements of having two link posts to calculate between. Maybe the first was a photo post and the second was a video post.</li><li>There were 64 pairs where a native video post followed a link post. I’ll set those aside and get to them later.</li></ul><p>Once all that was categorized and grouped, that left us with 424 post pairings where a first, second or third share of a link post could be identified.</p><p>Of those 424 total Chicago Tribune link post pairings in 2017, where a link post was followed by another link post, 88.21 percent of second shares had lower reach than the first.</p><p>In the 88.21 percent of cases where the second link post underperformed the first, the median percentage loss of reach was 70.74 percent with a median decline of 29,515 reach and an average decline of 60,726.26 reach.</p><p>In the 11.79 percent of cases where the second link post outperformed the initial post, the median reach gain was 53.98 percent with a smaller median gain of 6,702 reach and an average reach gain of 13,857.88. When the second share won, the victory was smaller.</p><p>Drilling down further, all the second and third link posts the Chicago Tribune shared combined for a median reach of 12,603 and an average reach of 18,113.54. This is significantly below the median and average monthly per post reach for <em>every</em> month for all of 2017. This suggests that not only did the second post usually underperform the first post, but it had less potential than even a typical post.</p><p>In practice, it looked like this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*VwQcp20WD4UCADzg2JKIOg.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*_L0_nYMDusSOjcrGf9Qerg.png" /><figcaption>Second post</figcaption></figure><p>Or this:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*IDYqictFVUNOKkEg1K1GNw.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/455/1*2O2v_QrJtkyivkcNRkkc3Q.png" /><figcaption>Second post</figcaption></figure><p>So, if your goal is simply audience and exposure, you’re probably better off picking something new rather than re-sharing something old, provided you have choices.</p><h3><strong>Video as a Second Post</strong></h3><p>Separately, I mentioned there were 64 instances where a native video post followed a link post. In 62.5% of cases, that native video second post with a Bitly link to the same story actually outperformed in reach the original link post. That reverses the trend even if the dataset is smaller.</p><p>Here’s an example:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/461/1*Gwt7P99t9ReS3LHByISLow.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/460/1*g9am0SsokUZr8nqQCrQzhQ.png" /><figcaption>Second post</figcaption></figure><p>But, in this same sample of 64 pairs, the link posts averaged 1,765.71 link clicks while native videos averaged only 582.95 link clicks. As an aside, native photos as a second post averaged 212.1 link clicks.</p><p>As such, posting a native video as the second share strongly helped avoid the reach and engagement loss, but those videos got far fewer link clicks — as videos do. So even if you mitigate reach loss with native video second postings, your referral potential, measured by average link clicks, is diminished by 66.98 percent, based on this sample.</p><h3><strong>The takeaway</strong></h3><p>Now, why posts behave this way is not something Facebook publicly discusses. We know, loosely, when someone you follow makes a post, an initial determination is made based on many factors like topic and post type to determine if you’re likely to interact with that share. If it’s optimal, you see it. The algorithm tries to find the best initial audience based on an individual’s Facebook behavior. It’s person centered not publisher centered.</p><p>So if you’re a person who always interacts with Cubs videos and the Chicago Tribune, which you follow, posts a Cubs video, you’re more likely to see that post. Once there’s interactions, that share could be expanded to more people based on comments, likes etc.</p><p>If a publisher shares the same link a second time, Facebook, having already tried to find the optimal matches, tries to find the second best audience. In our example above, this might be a Chicago sports fan not just a Cubs fan. This, in theory, lowers the ceiling. This also debunks any thought that there’s a safe window after which you can repost for maximum exposure.</p><p>So what do you do with all this?</p><p>Based on the Tribune’s data, if your goal is just referral traffic, <a href="http://www.niemanlab.org/2018/08/facebooks-message-to-media-we-are-not-interested-in-talking-to-you-about-your-traffic-that-is-the-old-world-and-there-is-no-going-back/">something Facebook doesn’t much care about</a>, maybe ease up on spamming the same link. The data-smart strategy would be prioritize unshared content over previously shared content.</p><p>If your goal is exposure (read: reach), native video second posts appear like an effective option, for now, to mitigate that loss. Native video second posts might be a good way to keep a hot topic — or one you want to be hot — in front of Facebook users. But you almost certainly won’t get the same level of referral traffic. And let’s be honest, many stories aren’t effectively told through video.</p><p>Lastly, despite a sizeable decline in reach, you could establish a certain threshold where you would share a link post a second time because even a 70-percent decline in reach is still above the median reach of a typical post. Just don’t be surprised when your Facebook engagement metrics wither like Chicago’s dreams of securing Amazon’s #HQ2 — or second posts about that very topic.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/466/1*O6cJX3RNEdO-2ajsQZSYvw.png" /><figcaption>First post</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/458/1*6uIlYJ-9GjotTZtAdUu3_A.png" /><figcaption>Second post</figcaption></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d19264ce1d" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Have Twitter users stopped clicking on links?]]></title>
            <link>https://kurtgessler.medium.com/have-twitter-users-stopped-clicking-on-links-79d4cfb88660?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/79d4cfb88660</guid>
            <category><![CDATA[analytics]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[twitter]]></category>
            <category><![CDATA[metrics]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Tue, 04 Sep 2018 14:13:18 GMT</pubDate>
            <atom:updated>2018-09-04T14:13:18.910Z</atom:updated>
            <content:encoded><![CDATA[<p>Responses that start with “saved you a click” or “I didn’t click the link, but …” are ubiquitous on social media, especially Twitter.</p><p>But more and more, some analytics show it’s becoming less a punchline and more a reality at a time when publishers aren’t as concerned with soft social metrics like engagement and are more focused on direct referral traffic.</p><p>Those link clicks can translate into subscriptions, aka direct consumer revenue.</p><p>While Twitter is neither Google nor Facebook for the Tribune, it remains a top 5 referrer source, and it’s something that we can control. It outstripped email alert link clicks 7-to-1 at the beginning of 2017. For all its trolls and troubles, it makes a difference, audience wise.</p><p>Unfortunately, the number of monthly link clicks on the Chicago Tribune’s primary Twitter account, <a href="https://twitter.com/chicagotribune">@chicagotribune</a>, have declined 40.88% from January 2017 to July 2018, according to Twitter Analytics data. And that’s after a rebound.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0HqyDH1RVwFF6nMZaZWCaQ.png" /></figure><p>That’s a big decline in a fairly compressed window of time.</p><p>Now, there’s one obvious factor, but then it gets a little strange, so bear with me.</p><p>First and foremost, the Tribune’s methods and overall <a href="https://medium.com/@kurtgessler/twitter-tips-for-media-institutional-accounts-63bea2deec03">Twitter strategy</a> have been consistent. It’s mostly the same great team giving you some of the best Twitter on Twitter.</p><h3>Chicago Tribune on Twitter</h3><p>Wisconsin man gets a pay raise https://t.co/aTVKCcWQza</p><p>But, over time, the frequency has lagged a bit. The Tribune is tweeting fewer times each day.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5PJPLLBfTczoLRwiIRCpQA.png" /></figure><p>Obviously tweeting 10 fewer times a day has an impact on referral traffic over time.</p><p>But, I would expect other factors to mitigate this. For example, we’ve added 200,000 followers, even with the impact of the recent <a href="https://gizmodo.com/twitter-is-suspending-more-than-one-million-accounts-pe-1827409235">Twitter bot purge</a>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*USaz61CAaLuAVFMG5n4mRg.png" /></figure><p>That growth should raise the potential audience for each tweet.</p><p>And, if you look at overall tweet interactions — the total number of retweets and likes on a tweet via <a href="https://www.crowdtangle.com/resources/glossary">CrowdTangle</a> — it’s been volatile but certainly not in freefall.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WHTIwFM_oThx1QfLTv8pdw.png" /></figure><p>Monthly interactions and link clicks are clearly walking separate paths.</p><p>Now, factoring in that follower growth, it’s not surprising that both average and median per tweet impressions — the number of the times the tweet has been seen — have been flat to up. More followers; more potential impressions. And since last September, it’s been just gains.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WIXwq3po6yXKwypCvZQ9kg.png" /></figure><p>And again, I might expect an increase in engagement like link clicks. Not for every tweet, mind you, but en masse over time.</p><p>But this is where it gets strange. Despite average and median per tweet impressions rising since last September, link clicks continue to trend down. Even as followers have sharply increased and impressions are drifting up, the average and median link clicks per tweet have fallen. A lot.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-3-A1YuBn_QxTKFLjSFW6Q.png" /></figure><p>There’s your freefall.</p><p>In January 2017, a tweet sent from <a href="http://twitter.com/chicagotribune">@chicagotribune</a> had an average of 193.71 link clicks and a median of 116 link clicks. We tweeted something; more than 100 people clicked on the link to watch/read it. (And I still prefer medians in social media strategy to attenuate the outliers.) By July of 2018, that same tweet had an average of 137.68 link clicks and a median of 68. That’s a tremendous loss of per-tweet efficacy — and overall audience.</p><p>So, while the decline in daily tweets plays a role in overall monthly link click decline, the per-tweet loss is a far more damning influence (10 lost tweets x 116 link clicks vs. 62 tweets x a 48 median link-click loss). But what I can’t figure out is why that’s happening.</p><p>Tweeting less won’t impact the result of each individual tweet. And followers are up and interactions and impressions are steady to up. I would assume impressions (potential audience) and link clicks would follow similar arcs. But those last two charts are far from mirror images.</p><p>So are we just in the TL;DR era of Twitter? Is any other publishers seeing this per tweet loss in median and/or average link clicks? Because beyond just the loss in overall referrer traffic, we’re all just a little worse off if no one clicks the link anymore.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=79d4cfb88660" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Journalists buying followers on social media isn’t trivial. It’s unethical and it matters.]]></title>
            <link>https://kurtgessler.medium.com/journalists-buying-followers-on-social-media-isnt-trivial-it-s-unethical-and-it-matters-22072944b702?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/22072944b702</guid>
            <category><![CDATA[social-media]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Thu, 01 Feb 2018 15:29:59 GMT</pubDate>
            <atom:updated>2018-02-01T15:29:59.938Z</atom:updated>
            <content:encoded><![CDATA[<p>The most surprising wrinkle in the New York Times’ piece <a href="https://www.nytimes.com/interactive/2018/01/27/technology/social-media-bots.html">The Follower Factory</a> wasn’t the existence of this dark industry. And it wasn’t that some in the media paid for followers on Twitter, Facebook and Instagram. Without doubt, the part that grabbed me the most was the apathy some took toward this practice.</p><p>In the immediate murky backwash, the Sun-Times <a href="http://www.chicagotribune.com/news/chicagoinc/ct-met-richard-roeper-twitter-0130-chicago-inc-20180129-story.html">swiftly suspended columnist Richard Roeper</a>, one of the individuals referenced directly in the story.</p><blockquote>“We became aware over the weekend of issues relating to Rich Roeper’s Twitter account,” Sun-Times editor Chris Fusco said via email. “We’re investigating these issues. We will not be publishing any reviews or columns by Rich until this investigation is complete.”</blockquote><p>With the Sun-Times investigation ongoing, we should reserve full judgment. We have the NYT’s reporting tying Roeper to <a href="https://devumi.com/">Devumi</a>, a company built around “accelerating your social growth.” He was the lead name under NYT’s “Media” list.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/664/1*-xdKrRFZB6NDlNxLxYTymA.png" /></figure><p>You also can look at tools like Twitter Audit, which <a href="https://www.twitteraudit.com/richardroeper">suggests Roeper</a> has more than 125,000 fake followers of his 224,000 total.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/503/1*a5Pm9fL6w5GiC9dI0yLudw.png" /><figcaption>“Thumbs up” for 128,653 fake users.</figcaption></figure><p>Roeper <a href="https://www.washingtonpost.com/news/morning-mix/wp/2018/01/30/chicago-sun-times-sidelines-film-critic-richard-roeper-for-allegedly-buying-twitter-followers/?utm_term=.b6bb5acd8055">has not commented</a> on the issue.</p><p>But the issue really isn’t about a single columnist like Roeper. The central question many in the media was grappling with was simple: Just how big a deal was this? Does it matter that Richard Roeper might have paid a company for followers like <a href="https://twitter.com/Sidy06266084">@Sidy06266084</a> or <a href="https://twitter.com/fdsfdsf30499949">@fdsfdsf30499949</a>?</p><p>Buying followers isn’t specifically addressed in many social media policies and guidelines. I suspect <a href="https://www.thedailybeast.com/the-hill-bans-staff-from-buying-twitter-followers">it will be more common now</a>, especially with the Sun-Times setting the precedent that it’s a suspension-worthy, investigation-worthy offense. The closest that the Chicago Tribune might currently have is this:</p><blockquote>“Editorial and news departments also may work with marketing, promotion, creative, circulation or other departments to improve readership and financial success, but they should never do anything that might jeopardize the integrity of the news report.”</blockquote><p>So, does buying followers jeopardize the integrity of the news report? Some said no, and briskly waved off concerns.</p><h3>Neil Steinberg on Twitter</h3><p>Honestly, I viewed buying followers as like plastic surgery -- a self-enhancement I wouldn&#39;t do, myself, but can&#39;t condemn in others. I don&#39;t do it because it&#39;s sad, and because I read &quot;Death in Venice&quot; the scene where Gustaf von Auschenbach has his hair dyed. Not a moral thing. https://t.co/b4cBqRziwQ</p><h3>Tony Galati on Twitter</h3><p>@NeilSteinberg @richardroeper Seems like harmless oneupsmanship.</p><h3>pantagrapher on Twitter</h3><p>I agree. I get why it&#39;s lame; I don&#39;t get why it&#39;s unethical. https://t.co/DFRllfbTcU</p><p>Personally, I can think of four somewhat interwoven reasons why I find it troubling, why it does undercut your journalistic integrity.</p><h4>Journalists are in the truth business</h4><p>You are being inherently dishonest in buying followers. You’re trying to deceive someone, whether it’s the public — “Look at all the people that follow me. You should, too” — or your boss — “Look at how I’ve grown my social media footprint. I deserve a raise or a contract extension.” No one buys followers just because it’s fun. There’s an end game. Some level of purposeful deceit is part of that action. Without doubt, it’s damn hard work to build any kind of social following. It’s also hard work to tell a good story or make a great photo. The person who would take a shortcut and buy fans might be a person willing to take other shortcuts.</p><h4>It leaves a stain on more than just you</h4><p>Speaking of integrity, how do you think your employer looks when it’s revealed that you bought those followers? It doesn’t just impugn your reputation, it affects your company’s brand as well. Was your employer neglectful in not monitoring your social media behavior? Or, even worse, did it encourage its journalists to be deceitful in manner? Did this person’s coworkers also buy their Facebook friends? Trust is a fragile commodity.</p><h4>It’s no different than falsifying your resume</h4><p>When I hire people, I look at their social media accounts. Their activity. Their fluency with the language and platform. The engagement on their pages. And yes, even their followers. Let’s say the Tribune were hiring a Bears beat writer and we had two equal candidates, say Jane and Jen. Jane had 500,000 Twitter, Facebook and Instagram followers and Jen had 500. Could that be a point of differentiation? Unquestionably, yes. There’s pressure in 2018 to have a robust social media following. It equates to tips and trust and traffic. That pressure, whether in Hollywood or the media, is why services like Devumi exist.</p><p>But in the end, buying followers is like lying on your resume. You’re giving the impression you have a stronger, more loyal audience than you’ve actually earned. You’re implying that you worked harder than your peers and cut through the social media noise and created a relationship with people.</p><p>But just like padding a resume, it never ends well. Fake followers don’t equate to higher engagement, no more than putting Python on my Linkedin page makes me a web developer.</p><h4>You’re supporting a shady industry based on trickery</h4><p>If you pay someone $20 for 5,000 Twitter followers on the social media black market, where is your money going? What industry are you furthering with your investment? Let’s find out.</p><p><a href="http://buytwitterfollowersreview.org/">Buy Twitter Followers Reviews</a> is a clearinghouse of many sites like Devumi, which happens to be ranked №1. At №3 is <a href="https://coincrack.com/buy-twitter-followers/?aff=UA-43616961-8">CoinCrack,</a> where 1,000 Twitter followers will cost you $9. CoinCrack’s motto is refreshingly honest — to help you “appear massively popular online.” Note: Just “appear” not “become.” Oh, but they’re having problems I see.</p><blockquote>Twitter has been causing quite a bit of trouble lately, expect delays. New Twitter Follower orders have been disabled until things cool down.</blockquote><p><a href="http://buycheapfollowerslikes.org/?ref=1">CheapFollowersLikes.org</a> also seems knowledgeable, especially in the benefits of a strong Instagram strategy.</p><blockquote>It always will be good to your business if you have and buy instagram followers. To get a lot of boons of using instagram you should need to get lots of instagram followers.</blockquote><p>Yup. All of that seems totally legit and reputable and something that I’d want to affiliate myself with professionally.</p><p>So is buying followers “harmless one-upsmanship” or “plastic surgery” for platforms? Absolutely not. On its surface, it lacks the basic integrity of the profession. It’s dishonesty that benefits the journalist and no one else. The intent is to deceive. And it serves to commoditize what should be a genuine relationship between a journalist and the public.</p><p>But hey, if none of that matters, I’m sure there’s a batch of Sidy06266084s ripe and ready for purchase.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=22072944b702" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Facebook’s algorithm isn’t surfacing one-third of our posts. And it’s getting worse]]></title>
            <link>https://kurtgessler.medium.com/facebooks-algorithm-isn-t-surfacing-one-third-of-our-posts-and-it-s-getting-worse-68e37ee025a3?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/68e37ee025a3</guid>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[content-marketing]]></category>
            <category><![CDATA[social-media]]></category>
            <category><![CDATA[publishing]]></category>
            <category><![CDATA[facebook]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Tue, 18 Apr 2017 02:49:32 GMT</pubDate>
            <atom:updated>2017-04-19T16:48:41.244Z</atom:updated>
            <content:encoded><![CDATA[<p>Starting in January of this year, we at the Chicago Tribune started to anecdotally see a fairly significant change in our post reach.</p><p>We weren’t seeing a huge difference in post consumption or daily average reach, but we were just seeing more misses than hits. At the Tribune, we have a fairly stable and predictable audience. We had around a half million fans at the end of March and have seen slow but steady growth in the last year. Most Facebook posts fell into the 25,000 to 50,000 reach range — with a few big successes and few spectacular failures each day, usually based on the quality of the content or the quality and creativity of the share.</p><p>But starting earlier this year, we started to see far more misses. And not reaches in the low 20,000&#39;s but 4,000 reach or 6,000 reach. Digital Editor <a href="https://medium.com/@RandiMShaffer">Randi Shaffer</a> was one of the first to notice.</p><p>Some of our data looked fairly normal. Our average daily organic reach looks volatile but familiar. We used average daily because Facebook doesn’t give you monthly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9ba_B-vNm4XC7ORApms91g.png" /></figure><p>But, our per post metrics were falling. Our average organic post reach was on the decline, but not that far off a prior low of mid-2016.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*k3oyIJCWsj8trNXWzNJMMQ.png" /></figure><p>Still, averages can be strongly affected by outliers. It was our median organic post reach that was far more revealing. Yes, we were on a downward trend — and that trend put us at record lows. The middle was falling precipitously. Our median post was being seen by fewer and fewer people, organically.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wf-DW58zaRofbNKdeGweqw.png" /></figure><p>To try to get a handle on the nature of the issue, I decided to download 15 months of Facebook post data — no small feat since Insights only lets you grab 500 posts at a time. Then I merged the 30 csv files and sorted the organic post data into four buckets: less than 10,000 reach, 10,001–25,000, 25,001-50,000 and 50,001+. The results literally stunned me. The number of posts that fell into the category of lowest efficacy—the ones seen by the fewest number of people — was skyrocketing.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gd-dvIcWq6UehdUrEH3uLQ.png" /></figure><p>In December of 2016, we had only 8 posts with 10,000 reach or less. In January of 2017, that had grown to 80. In February, 159. And in March, a ridiculous 242 posts were seen by fewer than 10,000 people. And while late 2016 saw record lows in that lowest quartile, that 242 is far above any prior month in our dataset. And we were seeing a steady decrease in that 25,001 to 50,000 quartile. That had gone from 248 in January 2016 to 141 in March 2017.</p><p>What did this mean? In baseball terms, we were hitting far fewer doubles and we were striking out 1 every 3 times at the plate. Four months earlier, we struck out 1 of every 90 at-bats.</p><p>And it was happening despite solid growth on our Facebook page — which, logically, would translate to increased reach.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bcaT-53Pg9APo-skn9spYQ.png" /></figure><p>Yet many of our posts were seeing diminished organic reach despite picking up more than 130,000 Facebook fans.</p><p>So why could this be happening? Let’s look at some of the possible factors.</p><p><strong>Increase in post frequency</strong></p><p>One possible explanation is that during this period we slightly increased the number of posts per day. However, the rise was quite small and doesn’t coincide with the recent trend.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*t9FDW6mXHm3Epny8IYvgdQ.png" /></figure><p>We went from roughly 20 posts per day to 24 posts per day. And however small that may look per day, that number adds up over time. In this timeframe, that represented a 25 percent increase in monthly posts.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*G-lexTZ6F-q-j4Rp_s3DRg.png" /></figure><p>However, Facebook’s formal guidance is roughly 24 to 48 posts per day. And we looked at the cadence of 23 other newspaper Facebook pages and determined we were in the mid- to low-end. For the last six months, the Atlanta Journal-Constitution (661,551 fans) posted 74 times per day and the Houston Chronicle (385,086 fans) 58. Big thanks to <a href="https://twitter.com/wolfeelizabeth?lang=en">Digital Editor Elizabeth Wolfe</a> for crunching these numbers with CrowdTangle to prove our normalcy.</p><p>While it may be unlikely that frequency is our culprit, it’s a change in behavior we have to consider.</p><p>And moving ahead, this may be more of an issue. NewsWhip in its dive into three years of social data <a href="https://www.newswhip.com/2017/03/facebook-pages-engagements-per-post/">directly recommended</a> quality over quantity in its five primary takeaways: <em>“Post fewer but quality content that delivers value to your audience’s lives.”</em></p><p><strong>Content type mix</strong></p><p>In addition to frequency, Facebook also has its optimal content mix — 50 percent links, 25 percent video and 25 percent photos. To be honest, we don’t come close to this. But neither does any media organization comparable to the Chicago Tribune. Of those same 23 newspaper Facebook pages we looked at, most were posting majority link content types in the last 6 months. USA Today had the lowest percentage of links at 78.14% and the Sun-Times the most at 99.28%. But in general, most newspapers were sharing links on their primary Facebook page around 90% to 95% of the time. So the Tribune, at 97.2%, was slightly high but not atypical.</p><p>And given that our content mix hasn’t changed substantially in the last 15 months, this would have to be a somewhat remote possibility.</p><p>Still, NewsWhip <a href="https://www.newswhip.com/2016/05/facebook-engagement-restructure/">reported last year</a> that links were seeing less engagement even as video was on the rise, however artificial. And we’re posting mostly links.</p><p><strong>Facebook Instant Articles</strong></p><p>Another factor could be our lack of Instant Article adoption. Our parent company has been circumspect in regards to Instant Articles. We’ve been testing Instant Articles, but we have yet to deploy them in Chicago. Given the <a href="https://media.fb.com/2017/03/09/instant-articles-more-monetization-control-and-flexibility/">changes announced last month</a>, giving publishers a bit more control, I’m hopeful this will change.</p><p>But again, since we’ve not changed posting formats, this would seem an unlikely factor … unless the algorithm changed.</p><p><strong>So what about that Facebook algorithm …</strong></p><p>What I inevitably come back to is that something changed on Facebook’s side of the equation.</p><p>The last algo update mentioned in <a href="https://newsroom.fb.com/news/">Facebook’s newsroom blog</a> was “<a href="https://newsroom.fb.com/news/2017/01/news-feed-fyi-new-signals-to-show-you-more-authentic-and-timely-stories/">New Signals to Show You More Authentic and Timely Stories</a>” on Jan. 31. This is around a month after we identified our shift, but still curiously close. Of course Facebook didn’t foresee any hiccups: “We anticipate that most Pages won’t see any significant changes to their distribution in News Feed.”</p><p>Given that I hope the Tribune passes muster for “authentic,” let’s focus on the second half. What exact signals are being used to determine what “timely” means? It purports to favor topics that are being discussed in real time.</p><blockquote>For example, if your favorite soccer team just won a game, we might show you posts about the game higher up in News Feed because people are talking about it more broadly on Facebook.</blockquote><p>So exactly how much reach lift is conferred by dovetailing with Facebook-defined trending topics? Conversely, does this punish topics not being discussed? Or did the Facebook real-time algorithm become more like Instagram’s, prioritizing content based on the volume of immediate comments, shares, likes and reactions—and squelching posts that are initially ignored? That could be “timely.”</p><p>Beyond that, the usual <a href="http://wallaroomedia.com/facebook-newsfeed-algorithm-change-history/">sites that track Facebook changes</a> haven’t noted anything else that would account for our massive shifts in audience.</p><p>The <a href="https://newsroom.fb.com/news/2016/06/building-a-better-news-feed-for-you/">Friends and Family newsfeed change</a> in mid-2016 didn’t seem to strongly affect the Tribune. And Nieman <a href="http://www.niemanlab.org/2016/08/that-friends-and-family-facebook-algorithm-change-doesnt-seem-to-be-hurting-traffic-to-news-sites/">reported in mid-August</a> that the Friends and Family change seemed to have little impact on all publishers. <a href="http://www.theverge.com/2017/3/28/15081398/facebook-stories-snapchat-camera-direct">Ephemeral</a> also is months late. Maybe it’s a little of everything.</p><p>So here we are. The data show that we are having the fewest number of our most successful posts and the most of our least successful at a time when our strategy hasn’t significantly changed and our fans have grown.</p><p>So, is anyone else experiencing this situation, and if so, does anyone know why and how to compensate? Because if 1 of 3 Facebook posts isn’t going to be surfaced by the algorithm to a significant degree, that would change how we play the game.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=68e37ee025a3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[13 more before-after examples of engaging headlines]]></title>
            <link>https://kurtgessler.medium.com/13-more-before-after-examples-of-engaging-headlines-6c77d1f3d2fa?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/6c77d1f3d2fa</guid>
            <category><![CDATA[content-marketing]]></category>
            <category><![CDATA[seo]]></category>
            <category><![CDATA[medium]]></category>
            <category><![CDATA[marketing]]></category>
            <category><![CDATA[headlines]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Sun, 16 Apr 2017 05:07:21 GMT</pubDate>
            <atom:updated>2017-04-16T05:07:21.935Z</atom:updated>
            <content:encoded><![CDATA[<p>As the response was fairly positive to my original Medium post on <a href="https://medium.com/@kurtgessler/18-tips-for-writing-engaging-headlines-27-makeovers-that-saved-stories-from-extinction-55b8e73b84a2">headline engagement and best practices</a>, I wanted to share a few more examples before I moved on to other topics.</p><p>After the first piece, I received a lot of nice comments, the great folks at MediaShift <a href="http://mediashift.org/2016/11/18-tips-writing-engaging-headlines-27-makeovers-saved-stories-extinction/">republished the piece</a> and I even got a chance to ramble on about headlines to the awesome staff at the <a href="http://pilotonline.com/">Virginian-Pilot</a> in Norfolk, Va. As an aside, I now have a pretty slick Google Slides version of my thoughts on this topic!</p><p>But as I continued to talk about headlines, I kept finding new and better case studies. So what follows really is more of an addendum to the first post, a final notebook dump of sorts, where we at the Chicago Tribune took digitally deficient headlines and really focused on drawing out the compelling aspects.</p><p>As before, in all of these examples, we at least doubled realtime homepage engagement (click-through rate) after we made the change. We based this on Chartbeat’s heads-up display. So if 50 people were clicking on a headline before, at least 100 were after. Visually, that looks like this …</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/561/1*Q73L-k2DhRrEfDLRgHt3YA.png" /></figure><p>But this is only one measuring stick. More compelling heds are more sharable on social as well.</p><p>Why the before-after format? While every story’s unique, I think there’s more to be learned from studying empirically successful headlines than in just reading tips and theories. Plus, as I said, I already wrote <a href="https://medium.com/@kurtgessler/18-tips-for-writing-engaging-headlines-27-makeovers-that-saved-stories-from-extinction-55b8e73b84a2">that</a>. So, I hope there’s a little added value in these 13 new examples.</p><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6xfDJ_OnSqSaXop_TghzIA.png" /></figure><p>This one is particularly galling in that that headline is vague, a pun and also dead wrong, as the story specifically says they’re NOT falling down.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/841/1*CvrS6VFAJugwmMY6nYFcNg.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ifP8ORgFuCG6L_4z3LbSSg.png" /></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ohqsu29TDYMQmWArGUmL7Q.png" /><figcaption>The sports version of the “Rain didn’t dampen the spirits of fest-goers” headline.</figcaption></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/659/1*dvs_QEF8sbgw6HTMYOnI-g.png" /><figcaption>Added details and, most importantly, scope.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1002/1*mXxjy6cE5CQvupTnBHi9nw.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/276/1*yUOvikL2IM-TgU0ADkFRyQ.png" /><figcaption>Removed arcane “land bank” reference, added purpose.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/987/1*SOK2ej9k4fsOmh6uGVr1Zw.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/896/1*utTJmgBi8irEcGAXADTnWQ.jpeg" /><figcaption>Drew out the conflict a bit with stronger language and simplified the issue.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gPnMucszSv7PW4u7rGHOrg.png" /><figcaption>This is a perfectly good headline — detail rich (price, geography) with a hint of controversy.</figcaption></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/271/1*9B7WLz6Fl_1iWnr9S31vsQ.png" /><figcaption>But context made all the difference.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_Nv2gHjSlhGzdvTaXhjuPA.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/969/1*BvsYKNnGWDitaUlmLoD_Eg.png" /><figcaption>Added specifics, removed the lesser-known term “equity grant” and made it more conversational.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rBMwoyRNVpxNvRQqWlZwlw.png" /><figcaption>Focused a lot of attention on the preference angle.</figcaption></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*KwE6lH_cuLaf5XW7vSomiQ.png" /><figcaption>Attempted to bring out the exclusive nature of this story while adding context. This isn’t just any jail.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*viEMItUxq0_IGWV33JJyXA.png" /></figure><p>After, part 1:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/647/1*gfRbeLAIuQqTrRxsv9wAfA.png" /><figcaption>Started to hone in on the better part of the story.</figcaption></figure><p>After, part 2:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*aX5PtEypbJN_ZNjAHp8Qew.png" /><figcaption>And we’re all in on the most compelling aspect, as determined by our readers.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Em-cluvX3iZxuRaeEQTjDw.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RNEW0ic6G1_FfnACK7khcA.png" /><figcaption>Removed unnecessary last names and focused on cleanly telling the narrative.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3gPomxsBHA8fP7lpMubhaw.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/544/1*IrFQkh3r88_bEZmdfpidkA.png" /><figcaption>Here’s another case where the first headline was fine. But through testing, this quote proved far more engaging.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Lz0wd35NRPNKrOlSL1VsPA.png" /><figcaption>Starts with Illinois geologists. In fact, sand is the third noun.</figcaption></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/923/1*fuyeHl7_9VmGXuSRs97K-A.png" /><figcaption>The initial headline was detail rich but actually less clear. This is a story about missing sand. We used a question hed to hammer that home.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*rmUGkzb-abFUseN0-9bUdQ.png" /></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*3oOWy7c892qalXZ9vpkPVQ.png" /><figcaption>Details. Details. Details.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NDk-8g6AnZRV-zHXfBIeEw.png" /><figcaption>Original Associated Press headline.</figcaption></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tSZjetjdATNshGXbS61j8Q.png" /><figcaption>Snaps right to the most controversial statement made at the event — the one that got all the attention.</figcaption></figure><p>More H/T’s to my great Tribune team, including Jeff Cercone, Kyle Betts and Aly Brumback and Alex Parker.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6c77d1f3d2fa" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[18 tips for writing engaging headlines + 27 makeovers that saved stories from extinction]]></title>
            <link>https://kurtgessler.medium.com/18-tips-for-writing-engaging-headlines-27-makeovers-that-saved-stories-from-extinction-55b8e73b84a2?source=rss-dd9f4cc96152------2</link>
            <guid isPermaLink="false">https://medium.com/p/55b8e73b84a2</guid>
            <category><![CDATA[headlines]]></category>
            <category><![CDATA[seo]]></category>
            <category><![CDATA[content-strategy]]></category>
            <category><![CDATA[blogging]]></category>
            <category><![CDATA[content-marketing]]></category>
            <dc:creator><![CDATA[Kurt Gessler]]></dc:creator>
            <pubDate>Mon, 28 Nov 2016 18:15:45 GMT</pubDate>
            <atom:updated>2017-02-17T02:01:09.833Z</atom:updated>
            <content:encoded><![CDATA[<p>Let’s start with the obvious: Most news sites are in headline crisis.</p><p>It’s truly a crisis because your headline is the first — and maybe the best — shot you have at attracting a reader.</p><p>Yet many headlines look like these:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N6ElI4C4SE5zkvDvqTvdYQ.jpeg" /></figure><p>Honestly, based on those examples, it’s hard to argue against Will Oremus’ <a href="http://www.slate.com/blogs/moneybox/2016/03/11/newspapers_are_still_putting_boring_headlines_on_amazing_stories_like_the.html">case in Slate</a> that “local media routinely fails to emphasize the most interesting aspect of amazing news stories.”</p><blockquote>“(Media organizations are) laboring under the mistaken assumption that the role of a headline is to summarize a few relevant facts from a story that people will read regardless. That assumption is a relic of a time when people’s choice of news sources was constrained by the few newspapers or magazines they happened to subscribe to, or the few channels available in their basic cable package. Even then, I’m certain that fascinating stories went unread thanks to dreary headlines. Editors just didn’t have the analytics tools to measure it.”</blockquote><p>And newspaper web sites seem to be the epicenter of this epidemic. Now, there are several understandable reasons behind this. First, being understated is baked into the newspaper workflow because you can rely on the full context of a printed page. In a newspaper, a story might get a subhead, pullquote, graphic and photo. Online, you don’t get that sumptuous setting. Offsite, the source may be harder to discern, which in many cases suggests geographic (Boston Globe, Houston Chronicle) or topical context. And looking at just a headline and link in an RSS feed won’t tell you what section, giving readers a clue that “Brady” in the headline means Tom not Jim or Bunch.</p><p>Online, your product is unbundled. You get 10 words. Or 8. Or maybe 13, like I used above, to market your work. Digital success is like selling a newspaper story by story rather than day by day or week by week. And in selling that day’s paper, by subscription or newsstand, there’s just less urgency to make the headline awesome on that 150-word story buried at the bottom of page 11. Sections and geographic centers all are comfortable assumptions you can’t make in digital headlines. You must have a certain sense of desperation in writing web headlines, like those eight words are the difference between that column’s or blog’s life or death. Mostly, because it is. You aren’t owed readership. Your headline helps earn it — along with a handful of other factors like author and brand.</p><p>So any strategy involving growing and sustaining digital audience must incorporate excellence in headline writing. Must.</p><p>A few points to clarify here as we begin. I’ll be discussing writing for readers here, not for search engine optimization. That will be a consideration at times, but mostly we’re talking about people creating headlines for people.</p><p>The primary metric we’ll use is homepage click-through rate — the rate in which people loyal to the brand select (or don’t) a piece of content from a field of competing stories. I’m not going to focus on social analytics, engaged time, quality clicks etc. Those can be strongly influenced by platform, design or story length. Engagement time is the job of the content and the UX. The headline just gets people there. That’s our mission here.</p><p>And lastly, I’m not trying to sell anything. There are many great tools to leverage realtime metrics and A/B test headlines, like <a href="https://chartbeat.com/publishing/headline-optimization/">Chartbeat</a>, <a href="https://www.taboola.com/increase-engagement">Taboola</a> and <a href="https://www.optimizely.com/resources/ab-testing-tool/">Optimizely</a>. But all of those are still based on crafting good headlines. You don’t learn much A/B testing garbage.</p><p>So we’ll start with the 18 tips on writing effective headlines to help you create those engaging options. Then we’ll go through 27 examples of stories on ChicagoTribune.com that were first published with vague or print-centric headlines. I’ve picked these 27 examples because, in all those cases, we at least doubled realtime homepage engagement (click-through rate) after we made the change. We based this on Chartbeat’s heads-up display. So if 20 people were clicking on a headline before we reworked it, at least 40 were after. Most exceeded that threshold. These 27 cases were game-changers not just a nominal improvement. Hopefully, between the tips and practical lessons in the before-and-after examples, there are a few takeaways.</p><p><strong>1. Be clear and focused, first and foremost.</strong> Directly tell the reader why they should care or how this affects his or her life. There is nothing more important than clarity. Nothing. Straight headlines aren’t dull — they are lean and effective, elegant in their own right.</p><p>Part of being clear, at times, is stating what the writer and editor believe to be obvious. “Homeowners face tax hike under mayor’s budget plan” makes sense if your readers come from a single town. But how likely is that? “Chicago homeowners face tax hike under Mayor Emanuel’s budget plan” is not only more clear, but also better for SEO. And considering nearly of <a href="http://gawker.com/5783852/nearly-one-third-of-americans-dont-know-who-the-vice-president-is">third of Americans</a> couldn’t name the vice president, don’t assume they know their mayor or wouldn’t benefit from knowing exactly what city we’re talking about.</p><p><strong>2. Don’t get too cute or punny.</strong> Let me underscore that: Most pun headlines suck. Why? You assume the reader gets the joke. You assume there are no language barriers. You assume originality, that your pun hasn’t been used dozens of times already. And worst, you assume there’s nothing more interesting in the story.</p><p>Here’s a few examples of pun headlines where the joke is the focus rather than the story. And when your story centers around child molestation, like the Hastert piece, an attempt at clever wordplay only trivializes a serious matter.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*jvs_ljT_BYr10Fj0zDGzgw.jpeg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/568/1*rlYpziamYnakbHGI_5OWtQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*J02ZFIZetOl17nF91DFWLQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/483/1*pEEzezPTj1ZwbY9_EZqKbw.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZP9wQPm7rtXoVr2Mvz0tUg.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DPiYMkAwB9sIhNnEcqeI6Q.png" /></figure><p>What does an officer being “bullish” have to do with an arrest? And Pleez, sadly, has <a href="https://www.yelp.com/biz/pleez-highland-park">already closed</a>. I blame the imperceptible headline.</p><p><strong>3. Write it so it reads fast.</strong> The first few words are critical; don’t make a reader hunt to find key points. Front-load them into your headline. This also helps with SEO.</p><blockquote>“Billy Bush, embroiled in Trump tape scandal, is suspended from ‘Today’ show” —<a href="http://www.latimes.com/entertainment/envelope/cotown/la-et-st-bush-trump-today-20161008-snap-story.html"> Los Angeles Times</a>.</blockquote><blockquote>“Florence Henderson, TV’s Carol Brady, dies at 82” —<a href="http://www.cnn.com/2016/11/25/entertainment/florence-henderson-obit/index.html"> CNN</a>.</blockquote><p>Don’t start with “‘Brady Bunch’ star” or “TV’s Carol Brady.” Start with the name of the recognizable actress or disgraced celebrity.</p><p>For the same reasons, unless it’s absolutely critical, don’t start a headline with the attribution. “Police confirm serial killer escaped from jail” is less effective than “Serial killer escaped from jail, police confirm.” Key details come first.</p><p><strong>4. If the deadline isn’t now, write multiple headlines.</strong> Challenge yourself. Upworthy espouses <a href="https://www.theguardian.com/media/2014/mar/16/upworthy-website-generation-y-awful-headlines">25 discrete headlines</a> per story. That feels excessive, but it does suggest you shouldn’t stop after your first headline. I find writing at least three to be extremely helpful.</p><p><strong>5. Write with an active voice.</strong></p><p><strong>6. Favor proper nouns</strong> (names, teams, locations). This also helps with SEO. But also realize that sometimes being vague may be beneficial. Is the incident more important than the locale? Is the law more important than who proposed it? Both approaches have their place. Know when to use them.</p><p><strong>7. Is the writer important?</strong> Consider including her or him.</p><p><strong>8. Headlines with eight words or more perform best</strong>, says <a href="http://www.outbrain.com/blog/how-headline-length-impacts-engagement">Outbrain</a>. Now here you may be at odds with SEO best practices. We’ve also seen SEO benefit to keeping headlines generally between 40 characters and 77 characters. And <a href="https://moz.com/blog/title-tag-length-guidelines-2016-edition">Moz recommends</a> headlines of around 60 characters to avoid truncation in search results pages. <a href="https://www.searchenginejournal.com/google-titles-and-descriptions-2016/163812/">Others</a> say 70 characters. The answer, in my mind, is to just write your best headline being mindful of where readers may not see the rest. A great headline that goes a little long is better than an SEO perfect headline that’s revolting.</p><p><strong>9. Ask a critical question in your headline.</strong></p><p><strong>10. Use descriptive and evocative words, terms that inspire or amuse or anger.</strong> A University of Pennsylvania <a href="http://journals.ama.org/doi/abs/10.1509/jmr.10.0353">study</a> suggests <a href="http://www.newyorker.com/online/blogs/elements/2014/01/the-six-things-that-make-stories-go-viral-will-amaze-and-maybe-infuriate-you.html?utm_source=www&amp;utm_medium=tw&amp;utm_campaign=20140121&amp;mobify=0">eliciting an emotional response</a> is the critical reason content goes viral. Chose words with power.</p><p><strong>11. Pique a reader’s curiosity — but don’t be click-baity. </strong>Again, clarity first. You erode brand trust and confidence if you’re just teasing and never delivering. Don’t play the guess-what-band-is-coming-to-town game. Be a compelling and reliable source of information.</p><p><strong>12.</strong> How-to’s, lists and tips can make compelling number-based headlines (“8 headline mistakes you can avoid” or “How you can avoid these 8 headline mistakes”). But <strong>the content must support the headline.</strong> You can’t use this list-style of headline on a narrative story. Headline and story must be in concert not conflict.</p><p><strong>13. Have a great quote? Use it. </strong>If you’re stuck, this is usually the shortest path to a unique headline.</p><p><strong>14. Don’t limit yourself to the story’s lede</strong> (beginning). The headline has to promote the most-engaging element, whether that’s a number, quote, fact, thought, topic, opinion etc. Maybe that’s 10 paragraphs in.</p><p><strong>15. Only use the most common acronyms.</strong> FBI? sure. SA’s office? Probably not. However, you should use abbreviations like “feds”. Headlines should be approachable and conversational.</p><p><strong>16. Some words just suck.</strong> Don’t use those. Here’s a few weak headline words I hate. All of them are hard to read, weak, vague or jargony. Post more in the comments and I’ll add them.</p><blockquote>Parcel<br>Zoning<br>Ordinance<br>Figure, as a noun for an individual<br>Individual<br>Incident<br>Issue, as a noun<br>Protocol<br>Flap, as a noun<br>Fans, as a verb<br>TIF<br>Announce (this is almost always extraneous) <br>Welcomes<br>Spat<br>formally/officially<br>Fiscal<br>Appropriate (verb)<br>Bolsters<br>Spate<br>Decry<br>Nigh, as in “the time is …”<br>Incentive, where “taxpayer money/funds” would be more clear<br>Bemoan</blockquote><p><strong>17. Focus on impact and implications.</strong> Why is this story important to people? I like a quote from co-worker Mark Jacob, metro editor at the Chicago Tribune: “Tell people why the story is about them.”</p><p>Structures like the ones below are becoming a little threadbare, but at their core they aim to be direct and useful, albeit with a touch of drama.</p><p>Here’s the truth about XXX<br>This is what XXX means to you<br>Experts offer advice on XXX<br>XXX things you need to know about YYY<br>This is the true story of XXX<br>XXX is a problem. This is the solution<br>XXX happened. Then YYY did.</p><blockquote>She called the man who sexually assaulted her a rapist. Then he sued her for defamation. —<a href="https://www.washingtonpost.com/news/post-nation/wp/2016/10/03/i-felt-re-victimized-woman-sued-for-referring-to-the-man-who-sexually-assaulted-her-as-a-rapist/?tid=sm_tw&amp;utm_term=.a690432ec3b1"> Washington Post</a></blockquote><blockquote>Looks like the 2nd debate had about 20% fewer viewers than the 1st. Here are the reasons why. —<a href="http://money.cnn.com/2016/10/10/media/clinton-trump-debate-nielsen-twitter-facebook/index.html"> CNN</a></blockquote><p>Hard to argue those aren’t compelling reads.</p><p><strong>18.</strong> But the most important tip isn’t arcane or gimmicky. It’s obvious, but it’s often ignored. <strong>Make <em>absolutely sure</em> the most interesting, most important part of the story is conveyed in your headline.</strong> And that usually takes time and effort. There’s no way to write a great headline by just skimming the story.</p><p>This <a href="http://graphics.chicagotribune.com/features/returning-to-the-prairie/">project</a> with amazing visuals went live on the Tribune with this headline: “Returning to the prairie.”</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*LXwNlFroQdUsBc3O_FJDTg.jpeg" /></figure><p>It doesn’t get much more vague than that. I knew it would do poorly, and in fact Chartbeat showed that it was being ignored by most of our readers. Realtime audience metrics helps a lot to identify weak headlines. So right away I knew it needed an intervention. It took me 10+ minutes, but I settled on “Man sold home, spent last decade to create 60-acre prairie garden.” And I played up the geography, Chicago’s affluent North Shore, in the readout. I probably should have put North Shore in the headline anyway.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/446/1*Gr9tAzkwQOXr1HIkDXI5lQ.jpeg" /></figure><p>A sense of scale. Hint of commitment through time. Clarity. The change had immediate impact and became the second-most read story of the day.</p><p>I didn’t change a photo or re-write the piece. I simply adjusted how it was being marketed to the audience. That’s the power of the headline.</p><p>Here’s 26 more before-and-after examples, all with lessons based on those tips.</p><p><a href="http://www.chicagotribune.com/dining/chi-topquality-japanese-wagyu-featured-at-roka-akor-20141202-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*unI9pdzOACbHgJWApQzZRQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1000/1*aC1kvw5ZJyeGu6PSr0mc9w.jpeg" /><figcaption><em>Removed jargon. Chose geography over specificity of a single restaurant.</em></figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/718/1*BsuggSw2Hmmj0Oa5KYqZoQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/718/1*w0ea13bMsOn4BVMOyMS1JA.jpeg" /><figcaption>Replaced vague “think twice” with specifics of why you should care.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GK6yfVw-vjgyY3c95M5GSQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ktclbo8Dm6PBn22Y5OrxcQ.jpeg" /><figcaption>Added powerful narrative of the crime and its geography.</figcaption></figure><p><a href="http://www.chicagotribune.com/news/local/breaking/ct-lake-michigan-asian-carp-met-20160822-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*NE2jLvb-7VS19UXsVP05Vg.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xBzz82Lb41dhPcKSE8GqPA.jpeg" /><figcaption>Replaced vague “multiple tools” with exotic specifics. The story is about those details.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*n921WXEgq9c49wtcy3KpoA.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/632/1*1kH4AJ03GoFOpfwPJMviyw.png" /><figcaption>Replaced vague “exultant” with examples. People we’re just happy, they wee crying.</figcaption></figure><p><a href="http://www.chicagotribune.com/news/local/breaking/chi-chicago-crime-violence-murder-buried-southwest-side20131006-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/974/1*udDi1ualT6ERU1hg-nlMUg.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/998/1*4MDZrCdJ1LJJAKivTMR4sQ.jpeg" /><figcaption>Removed geography and focused on nature of crime and relationship of subjects.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*ZU-gHEouzO4AJChUD2ZZiA.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7vQUvd3oYW1IjVmhZUy8Uw.png" /><figcaption>Removed jargon like “Big Coal” and focused on consumer angle. Clarified geography as Illinois.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*r_-etXeCA0Se9bVaRuMuNw.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/414/1*OOqO4Oz2YEZqjB-XopmGvQ.jpeg" /><figcaption>The nature of recipe is far more compelling than “limited edition” implies.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/610/1*7OlMBJobIh_6sROhy4lYXQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/540/1*mCuE7Q3ioDAED2t5eEg8jg.jpeg" /><figcaption>Added clarity and details. I mean, millions of things are made from scratch in Chicago.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*InBZ3_99wwKFyDBijnsUjA.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Ik_VULNW-MvzPQBPSpJjMg.jpeg" /><figcaption>Tried to add scale and intrigue by replacing vague phrase “old-school effort.” In retrospect, even this could be better, focusing on the technical challenges.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eCdzUNijIw0oX5thA5zBeA.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*6kam_8j8h_H2I2zuPuGOSQ.jpeg" /><figcaption>Added cost, narrowed geography and used a powerful quote.</figcaption></figure><p><a href="http://www.chicagotribune.com/ct-jeb-bush-presidential-bid-20160925-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/881/1*qiIumrZa0WJurfWt0HXSbw.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/955/1*MceECgG4i0I_kr8fR7ovYg.jpeg" /><figcaption>Used quote and more evocative language.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1001/1*SFM-EXTku1yHGr-B1Q1E5Q.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/863/1*S38FKEbApJxJQUyVHqJpCg.jpeg" /><figcaption>Narrowed geography to not needlessly scare parents of other Chicago schools.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*JYIaO5KpYjomigS4ansD5Q.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-XmnR9M08CvMPPuqzS2O1A.png" /><figcaption>Added scale and removed odd phrase “lobster roll addiction.” That’s not a thing.</figcaption></figure><p><a href="http://www.chicagotribune.com/news/local/breaking/chi-arcadio-davila-charged-with-murder-ricky-pike-20130305-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1010/1*ueNgOlonO1KHRfp6ozkr7g.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1006/1*UItIhYi4L22cgQbycNA11w.jpeg" /><figcaption>Added crime, victim and geography. The “gift to mom” was the arrests in her son’s murder, which isn’t very clear.</figcaption></figure><p><a href="http://www.chicagotribune.com/suburbs/hinsdale/ct-nik-wallenda-incline-students-met-20141106-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/992/1*MNgitsuM11cwV2WzdbPSjQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/990/1*f5ShiBgnbivy_mWVhjnS3g.jpeg" /><figcaption>Added humor and a little specificity since “complicated” is soft.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/904/1*DuZLL0Auf51DmTWs3_FSlQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/947/1*vmXeWm087QdUCna-SmEvrg.jpeg" /><figcaption>Focused on a unique element of story.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/569/1*C9UCp_v8P3KNbXgcHE-fng.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/298/1*V9CSCazyBKbwl8VMnYskYA.png" /><figcaption>Removed pun and focused on consumer novelty of the suit.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/878/1*lbU9cJQKJjfGyGEi0_-6Cw.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/450/1*Q_DQvN0UT6XKzxG7Cy6fzA.png" /><figcaption>Added specificity and scope.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1rPtCDzGhAVc_liPeJO2ug.png" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*9zSs5YziIwsf-Y_z7tw7IQ.png" /><figcaption>Added key institution, purpose and profile of the effort.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1021/1*0HyI7el9hpOeYDybUZQJqQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1010/1*VLrVD-DROESjXSry97WfaQ.jpeg" /><figcaption>Opted to remove names lesser-known athletes (vague can be good) and focused on the nature of the allegations.</figcaption></figure><p><a href="http://www.chicagotribune.com/news/local/breaking/ct-back-amusement-taxes-music-bars-20160822-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*V6oyOuAnu91NSkY3HhTW6Q.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*eUC1QoNCUIyuoMG-foz-2A.jpeg" /><figcaption>First headline was clear and solid. However, it lacked the highest-level implication.</figcaption></figure><p><a href="http://www.chicagotribune.com/news/local/breaking/chi-trump-expected-to-testify-in-chicago-about-condo-deal-20130514-story.html">Before:</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*2NNeg9ze2MJkjDKiHkOErA.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1020/1*oPlDrV26hm_XoxKtnil-1g.jpeg" /><figcaption>Pulled out attention-grabbing quote as focus. Probably could have kept “Chicago.”</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/751/1*uIeMCoTww9Cqy4F350cKcQ.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/559/1*XKQ8GiqdOgosqTmY8rhJTg.jpeg" /><figcaption>Removed “cha-ching” phrase and focused clearly on the consumer aspect.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/349/1*X0bAER8uweRVSR4-ZCIlPg.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/349/1*9X1DKzqkhjStIhTWkGpvew.jpeg" /><figcaption>Initial headline lacked clarity and detail.</figcaption></figure><p>Before:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/880/1*K6BiSCltCrBWZ7KQVhY2qg.jpeg" /></figure><p>After:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/332/1*ua_Dy9ZL62SaJKNnBW3hIA.jpeg" /><figcaption>Spelled out what vague phrase “difficult questions” meant.</figcaption></figure><p>Now, if your site publishes 12 articles and columns, it’s probably easy to apply these standards to all your content. However, in many cases it’s simply not feasible to A/B test 200 headlines a day. One viable game plan would be to focus on key stories, those on a home page or at the top of the app. Or focus on the stories that are most important to your mission. You also could look at realtime metrics to see what’s under performing at large. Another process we at the Chicago Tribune took was to analyze what types of content our readers most valued, so that when a story from a traditionally popular area was under performing, we tried a new headline rather than assume the story was the problem.</p><p>And, of course, sometimes you only have to look at a headline like “Cheater, cheater” or “Putting Chicago’s violence on the table” to determine a course correction is needed. And once you rescue those stories from bad headlines, hey, feel free to be a little “braggadocious.”</p><p><em>H/T to Jim Jaworski, Tom Palmer and Joe Ruppel for some of the examples.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=55b8e73b84a2" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>