The Digital Think Tank
Over a year after the New York Times Digital Innovation Report, think tanks need to re-evaluate their publishing strategies — and their vision.
In their most basic form, think tanks are part of the information flow in a democratic society, conducting research and analysis, and disseminating their findings and recommendations through publications and live gatherings that allow busy policymakers, advocates, journalists, and average citizens to hear diverse perspectives on important public issues.
— Andrew Selee; Executive Vice President, Woodrow Wilson Center
Think tanks are one piece of the information backbone of democracy. They do the hard work of researching and analyzing data on complex topics that others have neither the time nor the skills to do themselves. They conduct research that takes a longer view than our frequently myopic policy debates, and deliver that research through deep, and deeply text-based, analysis.
But media habits have changed. “Busy policy makers, advocates journalists and average citizens” are often too busy to attend an event or take a deep dive into an academic paper or publication. Serendipitous discovery through a well-curated social stream is just as likely to influence their work as deliberate research. Today, audiences are used to information finding them. It’s a model sprung from digital media, and one that privileges brevity, shareability and a highly visual approach to content.
The transition between these two models has plagued the publishing industry for years, and while think tanks don’t face the same economic pressures as newspapers, they share many of the same goals: more (and more influential) readers, greater engagement from readers and greater share of voice in the conversation. Think tanks that hope to win on those metrics must adapt to the shifting media behaviors of their audience.
It’s been over two years since The New York Times released “Snowfall: The Avalanche at Tunnel Creek,” setting a new bar for long form content; a year since the leak of the Digital Innovation Report, the closest thing we have to a roadmap for how analog publishers must adapt in a digital age. Since its release, new journalistic endeavors like Vox started to crack the code on digitally native publishing, while competitors The Upshot and FiveThirtyEight showed that there is an appetite for content that reorients digital publishing around creative displays of data.
News publishers are starting to make the transition to a digitally native mentality, but innovation within think tanks has been far less robust. In preparation for writing this essay, I conducted a top line audit of the front-end website experience of 25 think tanks, including the top 20 US-based think tanks:
- Only three had experimented with Snow Fall-style presentations of long form reports and data.
- Only six were fully optimized to make use of sharing on social media as a distribution platform.
- Only six had enabled easy sharing or embedding of their data as distinct from the stories and reports in which they were published.
- Only 10 consistently published full reports on their websites as opposed to abstracts supporting downloadable PDF files.
- Only half were certified as “mobile friendly” by Google.
While my research did reveal a few think tanks that are experimenting with the latest trends in digital publishing — and one or two that are well ahead of their peers— we have yet to see the emergence of a true, digitally native think tank.
LongReads, Modularization and Death to the PDF
What experience would a digitally-native think tank deliver to those “busy policy makers, advocates journalists and average citizens?”
Ideally, it would marry the depth and rigor that differentiates think tanks from the shallow reporting and “hot takes” that dominate the news online, but seamlessly integrate highly visual, short form “snippets” that lend themselves to serendipitous discovery and can compete for attention on the social web.
To achieve this think tanks must negotiate two (counter intuitively) complementary, approaches to content: long form and modularization.
Despite the rise in social media and a general trend towards tl;dr culture, long form content is undergoing a renaissance. From Snow Falls to Medium, and from LongReads to Wait But Why, content that asks for a deeper dive from the reader, but delivers on a high quality design and engaging reading experience, is finding an audience, especially among people within high-influence, information-driven occupations that are the core audience of think tanks.
It’s a format ideally suited to the content that think tanks are most valued for, but requires more than making reports available for download. The most successful long form content is written with the reader in mind (not fellow academics or researchers, as many policy papers are), and it invests as much time in design and the reading experience as it does the copy and research, incorporating a mix of creative layouts, motion graphics, photo and video to deliver an experience that brings the data to life and captures a reader’s attention.
On the other end of the spectrum is modularization — breaking content down into the smallest component parts that can engage a reader: a tweet, a Facebook post, a photo etc. Modularization involves the process of taking content and refashioning it for a short form experience, either as a deliberate promotional strategy, or to enable readers to share and spread content themselves. Modularization can be applied to whole stories and topics, or individual data points (charts, graphs, gifs, etc.) that are shared on social media. Tim O’Reilly has a fantastic piece on modularization and government — which he calls “the architecture of participation” — that 100% applies to the work of think tanks.
Sitting on a mountain of data, expertise and substantive analysis during a time of increased complexity and interest in economic policy and foreign affairs, think tanks are naturally poised to take advantage of these dual trends in publishing. There’s only one problem.
Most think tank websites and digital departments are not architected in a way that enables them to take advantage of trends in long form or modularized content.
Nowhere is this more apparent than the prevalence of PDF files on think tank websites. Every website I surveyed in the research for this article had a significant portion of their content locked up in PDF files, a format that actively impedes their goal of informing the public and driving conversation. PDF files often go unindexed by search engines (limiting discovery), are difficult to share and all but impossible to read on mobile devices. As a result, PDFs are highly unlikely to be read by any audience, at any time, on any device.
In an analysis of their own PDF reports conducted last year, the World Bank found that 1/3 of all reports published in PDF format by the bank had never been downloaded. Fewer than 13% had been downloaded more than 250 times, making PDF files — the dominant format think tanks use to present information — almost wholly ineffective in helping think tanks achieve their goals.
Which isn’t to suggest that there is no digital innovation happening at think tanks. Think tanks almost universally hopped on the blog wagon in the late ‘aughts, and many have experimented with one-off interactive tools for data display. The Brookings Institution in particular has invested heavily in exploring new publishing models, most significantly through the Brookings Essays, a content series modeled on The New York Times Snow Fall experiments that has delivered increased engagement with site visitors. The Economic Policy Institute and Pew Research both upload data onto their websites as unique, shareable content pieces.
Roughly half of all think tanks I surveyed had experimented in some notable way with long form content or short form social sharing. Despite a few notable efforts, great examples of think tanks adopting a truly digital-first mentality are few and far between. And no think tank has rebuilt their website (let along their whole digital ecosystem) to construct a digital-first experience that leverages the unique advantages of their content, while taking into account changes in how information flows or how media is consumed by their target audience.
The Digitally Native Think Tank
How can think tanks catch up to their more advanced peers innovating in the realm of policy and data journalism? There’s no one answer — in large part because the answer for anyone one think tank should be tailored to the resources, goals and assets at their disposal — but there are overarching principles and tactics that all think tanks should consider as starting points.
To Go Long Form, Invest In Your CMS
Well-designed, visual long form content creation is resource intensive. While The Brookings Essays are delivering from an engagement standpoint, they reportedly require months of staff time to research and write, and (presumably) more in development costs/time to produce. Investing that much time and energy on one piece — which may or may not resonate with the news cycle when it goes live — is akin to chasing viral videos, you might get lucky, or you might just have a very nicely done content piece with middling performance.
To fully unlock the value of this format, think tanks need to take the 2–3 month production cycle of The Brookings Essay and shrink it down, making long form a replicable content format that can scale across multiple reports and publications. In the words of Kevin Delaney, Editor-in-Chief of Quartz, “I’d rather have a Snow Fall builder than a Snow Fall.” Snow Fall builders require investing in a custom content management system.
“I’d rather have a Snow Fall builder than a Snow Fall.”
When Ezra Klein left The Washington Post to start Vox, he cited The Post’s archaic content management system (and conversely Vox Media’s innovative CMS, Chorus), as a primary reason for leaving. A customized CMS is an expensive proposition, but think tanks don’t have to go it alone, and they’re not starting from scratch. Vox posts their code on Github, as does FOLD, a new project of the MIT Media Lab that uses technology similar to Vox’s card stacks to create interactive, long form content pieces. Both could serve as a starting point for think tanks looking to change the technology underlying their publishing process.
Vox’s corporate structure is also instructive. Vox is an entity of Vox Media, under which multiple publishing brands — SB Nation, Vox, Eater, etc. — live. The Chorus platform, including all innovations, are developed for all the Vox brands based on the best practices and innovations uncovered by individual publications. In the past, like-minded think tanks have come together to produce costly interactive experiences. If costs proved prohibitive, ideologically similar think tanks could join together to build and maintain the backbone of an advanced CMS that serves the needs of all, while allowing for individual customizations as the CMS is implemented on individual websites.
CMS Investment = Content Modularization
Long form content and modularization are complementary, because it is in the modularization of content that long form becomes scalable. The key lies in creating a system that allows all content types — photos, slide shows, videos, document embeds, social call-outs, etc. — to be created individually and snapped into a larger template like legos (this is basically what is happening with FOLD).
Like Lego, the CMS should be customized such that any non-technical user can create these modular pieces with no developer assistance and only minimal design assistance after the initial production of visual assets. In fact, it’s preferable if the CMS automates entirely as many of these features as possible. Done right, automation can ensure consistent delivery of a quality experience across devices and platforms with staff of lower technical proficiency.
A few ways that modularization could be implemented to the benefit of think tanks include:
- Card Stacks As The New Policy Brief: Cards are transforming the web, and utilizing card stacks to track stories over time and organize information is one of Vox’s major innovations. Think tanks could take a page from Vox and turn card stacks into the new policy briefs and fact sheets, making them more readable on mobile devices, embeddable within larger stories, and easily updated with new research.
- Every Data Point is a Potential Content Module: Well-designed data is gold on social media. Make all graphics and data points into modular elements that are embeddable within long form content or third party platforms, shareable on social channels, and searchable. These are the lego bricks that fill your content frame. A few think tanks, like Pew and EPI, are already exploring this. (Update: A week after this was published, Quartz announced the launch of Atlas, a tool to do precisely this.)
- Mine Your Long Form Video for Content: Video is a largely untapped resource for think tanks. Almost all think tanks hold regular events. These are often live-streamed or available for viewing in their totality. Very few people want watch a 1.5 hour live stream or archived video. Event footage is not just a final product, its also raw material. Mine that footage to create short videos that highlight the most shareable moments.
- Enable Deep Linking within Content: Even as you modularize and adapt content into shorter form moments, there are ways to make your long form content more useful. Enable deep linking to specific time codes within longer videos. You can do the same for text content (See every paragraph written on Jay Rosen’s Press Think).
Implementing these features can maximize the return on your investment in time and resources producing long form, Snow Fall-style pieces. All think tanks should consider the types of content they regularly produce and how modularization can create a more engaging on-page experience for the reader and wring more worth out of each piece of content.
Optimize for Social to Drive Discovery
Modularization can help make rich-media long form content scaleable, but it also needs to be shareable. Think tanks often measure success by the volume of news coverage they attain. In digital media, social conversation is the new press hit — all content elements need to be optimized for social.
- Optimize Out-Bound Link Shares: It takes a tailored approach to build a referral loop between your readers and their social networks. All social platforms have specific content formats optimized to facilitate this kind of interaction. Customized share copy and imagery for every platform you use, and make sure every site contributor should be able to control the appearance of these link shares within the CMS, when “core” content is loaded into the system.
- Package Content for Social-Only: Not every piece of content must — or should — live on your website. Think tanks need to get on the distributed content bandwagon by packaging content specifically to perform well on social — either as a means of driving traffic or for pure on-platform engagement. In this latter category, Vox scored big with their packaging of an exclusive interview with President Obama, and Facebook’s new Instant Articles format (currently limited to a few news publishers), indicates that this is a publishing strategy that’s just getting started.
Once you’ve optimized your process for customized out-bound social promotion, consider A/B testing the content itself. Platforms like Share Progress allow anyone to A/B test the images and text for out-bound social, while also tracking 2nd and 3rd generation shares. Other platforms, like Naytev, allow you to A/B test your own social posts, delivering intelligence that can guide how you spend your paid budget.
Have An Audience Development Strategy
A great on-page reading/viewing experience, modular content creation that works at scale, serendipitous discovery through optimized social sharing — these things only gets you so far. In the past, your audience may have sought you out. Now, journalists are on deadline, and often responsible for churning out multiple blog posts per day on top of their old workload. Legislative aides or advocates may only seek you out when they have a specific need.
Just like The New York Times and other publications, think tanks must invest resources in an audience development strategy to ensure that the investment in content creation bears fruit. Audience development includes:
- Targeted (and retargeted) paid media. Every think tank needs a dedicated paid media budget to build an audience. Used in combination, keyword/interest targeting, custom audiences that leverage your email newsletter and event registration lists, and retargeting of site visitors through tracking pixels can form the basis of a highly targeted digital audience that cares about your work and will spread your research and policy prescriptions on social media. Paid promotion will ensure that these audiences actually see your content, while some smart data collection, will allow you to grow and refine that audience over time.
- Social listening. A social listening strategy will help you identify opportunities to engage influencers and join the online conversation. When conversations begin trending, you can work with your own experts and staff to engage that conversation directly, and incorporate new hashtags and keywords into your paid media targeting, contributing to the targeted growth of your audience.
- SEO. Yes, SEO. Even in the age of social discovery, it pays to cover the basics. Perform a good old fashioned SEO analysis of your site, learn where you rank, put together a plan to improve, and assign someone to enforce good SEO practices in all the content you produce. People are searching for the content you create. Make sure they can find it (and then fold them into your audience development with tracking pixels so you can build a relationship with that reader beyond the first visit).
A Change in Vision — Not Just Tactics
Think tanks are already making enormous investments in content creation and distribution. I’ve outlined the basics for adapting that process to the digital age, but deploying these changes is just the beginning. The reason think tanks need this counsel at all is because, as my initial research for this piece indicated, they fell behind — and are falling more behind every day.
In just the last few months, Buzzfeed announced a new custom platform for measuring social sharing, while Facebook continued to change the news distribution game with more tweaks to its algorithm. New metrics of success like engaged time continue to gain traction, while PEW’s annual report confirmed the ongoing demise of traditional journalism and the further advance of mobile as the dominant digital platform.
Think tanks need to do more than adopt a few new tactics, or initiate a twice-a-decade web redesign. They need to change the way they think about their own work, and their relationship to digital. They need to recognize that they are publishers, just like The New York Times, The Atlantic, Vox, and any number of the institutions struggling to reinvent themselves to accommodate a changed media landscape.
Reinvention is going to be an iterative, evolutionary process involving not just changes to tactics, but a rethinking of strategy, workflow and staffing. Reinvention will require fundamentally rethinking how think tanks deliver on their mission. To go back to where we started:
[T]hink tanks are part of the information flow in a democratic society, conducting research and analysis, and disseminating their findings and recommendations through publications and live gatherings that allow busy policymakers, advocates, journalists, and average citizens to hear diverse perspectives on important public issues.
In the past 10 years, almost every word in that sentence has undergone a radical transformation provoked by the emergence of digital media — with one major exception: “think tanks.” The challenge that think tanks now face might be rephrased accordingly:
Think tanks need to adapt how they contribute to the flow of information in a democratic society, find new ways of conducting and presenting their findings in digital formats, and discover new distribution strategies to ensure that policymakers, journalists and average citizens are exposed to a diversity of opinions on important public policy issues.
That will require not just adaptation to the media landscape as it exists today, but the vision, the will, the discipline and the tools that allow for continued adaptation — and hopefully innovation — into the future.
