Save for Later

Bookmarking Apps and the Wish Economy

In July 2014, Alexis Madrigal reported for NPR on the strange case of Pinterest. The visual bookmarking app “is mostly known as a place people go to find things to buy or make,” he wrote, and — unusually for a high-flying tech company — it had first gained traction among “young women away from the coasts.” But to underestimate it would be a mistake: as “a repository of things that people would like to have or do,” he continued, Pinterest constitutes “a database of intentions.” And assuming that intentions have power and value, it stands to reason that the places where they accumulate do, too.

In presenting Pinterest as a database of intentions, Madrigal reopened a conversation started by John Battelle over a decade earlier, in 2003. At the time, Battelle had Google on the brain; the not-yet-giant was less than a year away from its IPO, and Battelle was in the thick of researching a book about its rise. “The Database of Intentions is simply this,” Battelle wrote:

The aggregate results of every search ever entered, every result list ever tendered, and every path taken as a result.

In 2010, Battelle expanded the notion. “My mistake in 2003 was to assume that the entire Database of Intentions was created through our interactions with traditional web search,” he acknowledged. “I no longer believe this to be true.” In place of this narrow definition, he perceived an ever-evolving field of signals — together constituting “no less than the sum of our economic and cultural potential.”

In his expansion, Battelle noted first four, then five types of signals:

• The Query — what I want
• The Social Graph — who I am, who I know
• The Status Update — what I’m doing, what’s happening
• The Check-In — where I am
• The Purchase — what I buy

Battelle freely acknowledged that his taxonomy of signals was incomplete; he closed his March 2010 update, “The Database of Intentions Is Far Larger Than I Thought,” by affirming that he was “on the lookout for new Signals,” and “quite certain we’re not nearly finished creating them.”

That very same month, a small startup set off on a path that would ultimately prove Battelle’s prediction right. It was in March 2010 that Pinterest launched in closed beta, strengthening a signal that had swirled for years without breaking through: the bookmark.

The Bookmark represents what we wish for. It’s the earliest indicator of intention, and the most vulnerable; by definition, the act of saving something for later means that whatever we hope for hasn’t happened yet. Bookmarks are placeholders for the future. By thumbing through them, we can start to see what might happen next.

Bookmarking tools nominally exist to help people return to where they left off — or, at least, to reassure themselves that they will. Scanning the landscape here in 2015, eight bookmarking apps stand out: Pinterest, Evernote, Pinboard, Product Hunt, Pocket, Instapaper, Wunderlist, and Tumblr.

Pinterest’s value proposition is drawn verbatim from their logged-out Search page. Evernote’s is drawn verbatim from their logged-out homepage, but is one sentence among many; here, I focus on the “Collect” aspect of Evernote’s functionality, rather than its many other use cases. Pinboard’s is drawn verbatim from their logged-out homepage. Product Hunt’s is a paraphrased version of the longer purpose statement on their About page: “Product Hunt surfaces the best new products, every day. It’s a place for product-loving enthusiasts to share and geek out about the latest mobile apps, websites, hardware projects, and tech creations.” Pocket’s is a paraphrased version of their mission statement, drawn from this June 2015 post by CEO Nate Weiner: “To enable people to save the most interesting and important content flowing through their day so they can view it anytime, anywhere.” Instapaper’s is drawn verbatim from their logged-out homepage; only capitalization has been adjusted. Wunderlist’s is drawn from this March 2013 blog post introducing the “Add to Wunderlist” feature; admittedly, it doesn’t fully describe Wunderlist’s overall value proposition, but I’ve highlighted it here to improve comparability. Tumblr’s is a paraphrased version of the value proposition outlined on their About page: “Tumblr lets you effortlessly share anything.”

• Pinterest describes itself as both a “visual bookmarking tool” and a “catalog of ideas”; the one assembles the other. The interface presents images in an infinitely-scrolling grid. Boards are public by default, but can be made private.
• Evernote describes itself today as a “workspace” enabling four main types of productivity: writing, collecting, discussing, and presenting. In an earlier arc, Evernote’s ambition was to become an “extension of the mind,” empowering people to “remember everything.” Flexible to a fault, Evernote accepts all kinds of content and offers both visual and textual prisms for retrieving it. Evernote notebooks are private by default, but can be shared.
• Pinboard describes itself as “a fast, no-nonsense bookmarking site for people who value privacy and speed.” Built explicitly as a successor to Delicious — the most prominent of the original social bookmarking services — Pinboard “mostly ditched the ‘social’ aspect.” Its interface is textual and utilitarian: plumbing, not wallpaper. Pinboard bookmarks are public by default, but can easily be made private.
• Product Hunt describes itself as “reddit for products.” Emphasizing the competitive angle of the “hunt” and the social dimension of developing and sharing opinions, Product Hunt may seem like an odd service to describe as a “bookmarking app.” Yet its Collections feature was built in part as a response to a common user request for “a way to bookmark things to try later.” Product Hunt’s interface is oriented toward ordered lists of text, but the browser start page offered by its Chrome extension displays the latest trending products in a visual grid. All hunted items and collections on Product Hunt are public; only messages are private.
• Pocket describes itself as the “world’s leading save-for-later service.” The company initially launched a tool called “Read It Later,” but rebranded it in 2012 as Pocket — “Read It Later not just for pages of text, but videos, images and, in the long term, perhaps much more.” More recently, Pocket’s founder and CEO has described the service as a save button for the internet. With an interface characterized by cheerfully-designed lists of links peppered with thumbnail images, Pocket splits the difference between being textual and visual. Pocket is primarily private, though the app does include a built-in “Send to Friend” feature.
• Instapaper describes itself as a “simple tool for saving web pages to read later on your iPhone, iPad, Android, computer, or Kindle.” Although Pocket and Instapaper have much in common, Instapaper is distinguished by its particular attention to the reading experience. Even so, multimedia has crept in: in 2013, Instapaper added a dedicated video tab, and in September 2015, the team added thumbnail images to articles in the default list layout: “a big departure from the old Instapaper design, which leveraged text heavily, but…a huge improvement that’s been needed for some time now.” Reading lists in Instapaper are private, not public.
• Wunderlist describes itself as “the easiest way to get stuff done.” Defined on Wikipedia as “a cloud-based task management application,” Wunderlist in 2013 declared that it had become “much more than a To-Do List” with the introduction of its Add to Wunderlist feature — a “very clever extension for your Chrome, Firefox and Safari browser [that] allows you to save your favorite content from all over the web to Wunderlist with just one click.” Wunderlist’s interactive elements are primarily textual, but the ability to personalize its background visuals is sold as a Pro feature. To-do lists in Wunderlist are private by default, but can be published.
• Tumblr is “a microblogging platform and social networking website” that prides itself on its customizability. Although expression and connection are the stated goals, its distinctive handling of different content types makes it useful for digital scrapbooking. Tumblr also offers a bookmarklet to enable “quick sharing.” In the main, users consume Tumblr posts via a mixed-media, stacked-stream Dashboard. Yet individual Tumblrs often end up looking more like photo albums; many of the most popular themes feature grid layouts. Tumblrs are public by default, but can be made private.

Mapping these eight bookmarking apps against the key dimensions of medium (image-based vs. text-based) and mode (public vs. private) yields a portrait of the bookmarking landscape in 2015.

While this portrait illustrates what each tool makes possible, it’s missing the texture of real user behavior and real user numbers. On the behavioral side: even though boards on Pinterest are public by default, as of June 2015, almost one in three newly-created boards was a secret wedding board. Although Pinterest doesn’t publicize the overall percentage of public vs. secret boards, given that 30% of all newly-created boards are secret boards within a single category, it seems safe to assume that the balance has tipped toward secrecy since the official rollout of secret boards in February 2014. On the usage side, the numbers vary wildly as well. Case in point: in September 2015, Pinterest announced it had reached the milestone of 100 million monthly active users; in July 2014, Pinboard announced that its monthly active user count stood at 24,000, and had since 2012. What’s possible and what’s popular are two very different propositions.

Not every company highlighted here describes their offering as a “bookmarking app.” Yet the overlap in their value propositions is striking, underlined by the verbs at play: save, gather, and share. The brashness of “sharing” may seem at odds with the delicate act of bookmarking, but the tension largely resolves itself. The thing about sharing is that once you get your hands on a general-purpose tool for collecting and presenting content, it’s entirely possible that the audience you’re most interested in sharing with will be your future self. Whenever a nominally social tool introduces a single-player, “private” mode, that’s a sure sign that the people have spoken. What we wish for exposes us. Many people — perhaps most — just aren’t ready to have their intentions exposed.

If the overlap in the value propositions of these eight apps is striking, the symmetry of their use cases verges on startling. Wherever people gather ideas, the quantity and feasibility of the possibilities they collect will tend toward the fantastical. Anchored in routine events, these wishes blend pragmatism with escapism. Three common bookmarking use cases illustrate the duality of escapism and pragmatism particularly well: read later, recipe collecting, and wedding planning.

For a seemingly esoteric use case, “read later” has a surprising number of proponents. Pocket — originally launched in 2007 as “Read It Later” — and Instapaper are obvious contenders, but the use case crops up in less obvious places, too. On Wunderlist, the app automatically suggests a “Read Later” to-do list for users the first time they attempt to generate a to-do from a link that the service interprets as an article. On Pinterest, the official Pinterest account maintains a lively “Pin now, read later” board, taking the viewer’s assumed curiosity as an opportunity to evangelize the behavior: “Pin any articles that catch your eye to your own ‘Pin now, read later’ board.” Meanwhile, Evernote frames its integration with the Wall Street Journal as a way “to collect WSJ articles for research or reading later.” And in 2010, Apple acknowledged the use case by baking Reading List into Safari itself.

The reality is that most people save more articles than they’ll ever be able to read; Twitter is full of people lamenting the impossibility of reaching “Instapaper Zero,” and exulting in the rare cases that they do. But the most important feature of Read Later tools has never been the resulting queue; it’s the peace of mind that comes from knowing that, once you’ve saved the thing you stopped scrolling for, you’re free to move on.

Weddings are the most elaborate events most people will ever plan. The logistical, emotional, and collaborative pressures at play can quickly spiral into intense overwhelm. Strong systems promise to hold the downward spiral at bay. Among bookmarking apps, Pinterest is the loyal companion of wedding planners everywhere. Yet Pinterest is not alone: because of the inherent complexity and emotional salience of the wedding use case, apps like Evernote can’t resist touting the ways they can ease the process. And although Tumblr isn’t a wedding planning tool per se, the above excerpt from The Everything Mother of the Bride Book illustrates how seamlessly people turn general-purpose tools toward their own ends; here, the author treats Tumblr as Pinterest Lite. Finally, a screenshot from a user blog post shows Wunderlist in action as a wedding planning portal. Though the “Banquet permit(s)” line item may seem cut and dry — hardly wish-worthy — the “Figure out suit” and “Cake” to-do’s reflect the sort of use case Pinterest handles so well: the non-linear process of fulfilling a generic need in a personalized way.

In that sense, a one-word intention — “Cake” — could lead to a thousand rabbit holes; a wordless traversal of everything you’ve ever hoped for. Emerging from the lucid dream of visual search, you see a gridded still life symbolizing the future you didn’t know you wanted.

Whereas weddings happen once or twice in a lifetime, cooking is a daily affair. The promise of a powerful system is that it can turn the chore of menu planning into a reliable pleasure; in the best case, bookmarking apps do exactly that. Pinterest’s utility as a tool for menu planning and menu fantasizing is well-known, but again, it’s not alone: at one point, Evernote considered the use case compelling enough that they released a dedicated app, Evernote Food. (Since discontinued, presumably as part of Evernote’s professional pivot.) And on the occasion of the launch of Wunderlist’s Public Lists feature in August 2014, the introductory blog post singled out the recipe use case specifically. It’s not hard to imagine Product Hunt introducing recipe hunting as a future vertical; everyone eats, and so the topic tends to incite the kind of passionate discussion that service seeks.

Menu planning is fertile ground for fantasy because there are so many do-overs — three meals a day, seven days a week — and because pixels are calorie-free. Eventually, you do have to cook — or, concede to order in. But until then, you can dream.

Beyond the general-purpose bookmarking apps, there’s at least one more recipe-saving contender worth mentioning: NYT Cooking, an app that distills 17,000+ recipes drawn from the New York Times into an interface designed for exploration and action. (Built to sync with — you guessed it — Evernote.) NYT Cooking’s gridded layout of recipe tiles bears more than a passing resemblance to Pinterest. It’s the framing of the app that sets it apart: when you arrive at NYT Cooking, you know exactly what to do. Whereas general-purpose apps sell prospective users on all the possibilities — flailing around to find the use case that sticks —purpose-built apps succeed by painting a picture of a single desirable destination, then lighting the way.

Should bookmarking be considered an activity in its own right, or merely a feature? On the one hand, most of the bookmarking apps discussed here have modest user bases. On the other hand, Pinterest’s 100 million monthly active users confound any claim that bookmarking is exclusively a niche behavior. Whatever the ultimate answer, one thing is clear: as a feature, bookmarking is ubiquitous — once you know where to look.

A quick flip through my phone revealed bookmarking buttons at every turn:

• The New York Times allows for whisking links into other apps via the Share sheet, but it reserves pride of place for an in-app “Save” feature.
• Medium presents a bookmarking affordance alongside its “Recommend” button: one for me, one for you.
• Twitter favorites are notoriously — and somewhat infuriatingly — flexible, but bookmarking tweets with links to read later is one common use case.
• YouTube lets users save any video to “Watch Later” in a special playlist.
• Amazon has long offered a “Wish List” feature, which fits the bookmarking paradigm perfectly.
• Foursquare gives users the chance to “Save” any destination of interest.
• In July 2014, Facebook launched a “Save” feature — an outgrowth of their 2012 acquisition of read-it-later startup Spool. Saved items can be filtered automatically by content type — currently broken down into Links, Videos, Places, Music, Books, Movies, TV Shows, and Events.

The list of bookmarking affordances goes on and on. But why is it so long? Even taking the fundamental utility of bookmarking as a given, bookmarking behavior is surely a small fraction of consumption behavior. And if that’s the case, then why do apps uniformly prioritize the development of bookmarking features? Because the urge to enrich the Database of Intentions is irresistible.

For apps, bookmarks provide a way to capture a pure intention; pure signal. It’s because bookmarks are low-commitment that companies love them. Everyone knows that commitment is hard to come by. But hope springs eternal.

Wishes are vulnerable. They’re vulnerable to the whims of fate and the failings of willpower. But they’re also vulnerable to judgment. Every company that introduces private bookmarking — as an app or as a feature — knows the bargain it’s striking: the more protection I give, the more signal I’ll get. In fact, Evan Sharp — one of Pinterest’s co-founders — explicitly acknowledged this tradeoff in his 2014 interview with then-Atlantic journalist, Alexis Madrigal:

The things people do on Pinterest are so precious: this is what I want, this is what I think I want. They’re not ever sure, they’re feeling their way through. It’s a very weird emotional state. It’s this very beautiful thing to me. And if you start loading it with commercial pressure, it could really ruin the core of the experience.

Sharp’s observation calls to mind a similar impression taken from an entirely different domain, articulated by playwright Annie Baker and shared with me by Jenna Wortham:

The way human beings speak is so heartbreaking to me — we never sound the way we want to sound. We’re always stopping ourselves in mid–sentence because we’re so terrified of saying the wrong thing.…We’re all sort of quietly suffering as we go about our days, trying and failing to communicate to other people what we want and what we believe.

To wish is to “feel or express a strong desire or hope for something that cannot or probably will not happen.” To bookmark is a tentative act, verging on fatalistic; there are no guarantees. And that’s why, compared to the other signals that comprise the Database of Intentions — The Query, The Social Graph, The Status Update, The Check-In, and The Purchase — The Bookmark can be frustrating. The Social Graph and The Purchase map the past; The Query, The Status Update, and The Check-In map the present. They’re accurate. They’re reliable. They’re sensible. They’re also limited. Past is almost always prelude, but it doesn’t capture the ways in which we’re willing to change.

By contrast, the future is fractal. And in their capacity to structure abundance, bookmarks are, too. The wish economy matters because it catalogs all of the possibilities we’re open to. In aggregate, the incorporation of bookmarks into the Database of Intentions gives companies new tools for persuasion. But at a personal level, the tender utility of bookmarking remains.

Here is a list that helps me know myself.

Out of everything I could have saved, I chose this.