Pocket

By Coco Krumme and Mark Armstrong

Saving a story for later can tell us a lot about loyalty, longevity and quality—and it changes the way we think about the most popular stories on the web

—

If we’re to believe Woody Allen, “80 percent of success is showing up.” It takes more than just a great idea: you also need the drive and luck to be in a place where it can be recognized.

Allen is a writer (among other things), so one might wonder if his 80-percent rule applies to writing as well. How much of a writer’s success can be attributed to showing up in the right place at the right time—whether it’s on the front table at Barnes & Noble, the top of the New York Times bestseller list, the first page of Google search rankings, or the top of a person’s Facebook news feed?

And then: What happens after they show up?

Read It Later has a unique dataset to explore these kinds of questions. Nearly Over 4 million users rely on Read It Later when they click the “read later” in their browser, tablet or smartphone—and they come back to our app to dig deeper into the stories they’ve saved, recipes they’ve discovered, or videos their friends have recommended. That means Read It Later users aren’t just drive-by visitors to a piece of content—they’re passionate about it. The content is important enough that they added it to their queue so they wouldn’t miss it.

Earlier this year, our founder Nate Weiner offered a revealing look at how “time-shifting” online content has changed our reading behavior. Here, we take a closer look at the most-read authors on the web.

First, a quick note on the data: We take user privacy very seriously, and all data is anonymized and scrubbed of any identifying information before it is analyzed by our team. For this piece, we examined stories that were saved over a six-month period from May 1 to October 31, 2011. If you see any issues with what we’ve presented here, or have suggestions on what else we could investigate, please drop us a line: trends@readitlater.com.

So… Who Are the ‘Most-Read’ Authors on the Web?

Back to Woody Allen and that remaining 20 percent: Once a reader shows up, does it matter if anyone actually reads what someone has written? Sometimes it seems like what gets read is an afterthought. But often, it’s just too difficult to know what a reader does with a story after they’ve clicked through on the web, or purchased it in a bookstore.

We examined two things: How often our users clicked “read later” on an author’s story, and how often they returned to that story in some fashion.

To give some perspective, Read It Later users clicked “read later” more than 47 million times from May through October of this year. Here’s how that total breaks down over a six-month period:

Who were the most popular writers during this period? A quick glance shows that our readership is highly practical…and (surprise) big on technology.

Here’s the breakdown:

You really like Lifehacker. (Top 10 authors by total saves in Read It Later. “Articles saved” describes total saves of any article by that author, rather than unique articles saved. We looked at all authors with more than 1,000 saves. Note that some sites, including the Economist and Rolling Stone, may not have explicit bylines recognized by Read It Later, making differentiation of individual authors impossible.)

Nine of the Top 10 most-saved authors inside Read It Later are writers for Lifehacker, the must-read blog for lifestyle design. This makes sense, given the useful, evergreen content their team produces every day—it’s a natural for how-to and instructional content that readers might find and want to come back to. Kevin Purdy, Adam Pash, Gina Trapani and the rest lead Read It Later’s most-saved list, followed by a who’s-who of tech blogging—including Jesus Diaz of Gizmodo and MG Siegler of TechCrunch—and big-name writers like Cory Doctorow of Boing Boing and bestselling author Seth Godin. Read It Later users are also an international bunch: Japanese tech writers like Engadget’s Ittousai also made the most-saved list.

While pure popularity is great, we also want to know more about other factors that make content valuable—like longevity and loyalty. It’s not just about how many people clicked the first time—it’s about whether they also came back again.

When Like Means Love…What We Come Back to Read Again and Again

To examine longevity and loyalty, we also looked at the top “return rates” for authors with over 1,000 saves. That is, what was the percentage of Read It Later users who returned to that story in some way? (There’s a lot more to explore around exactly how much content readers consume, or how far down they get in a story, but we’ll examine that in a later post.)

The most interesting thing isn’t just that we found different authors for the top “return rate,” but also different categories of content and types of publishers. Top saves were focused on how-to and tech content (a likely side effect of our largest user group being early tech adopters—because, hey, even at 4 million users, we’re still ‘early’), but the authors with the strongest reader loyalty included writing about sports, general news and gaming.

The magic of Drew Magary.

One constant across both charts: Nick Denton’s Gawker Media properties (Lifehacker, Gizmodo, Deadspin, Gawker) are among the most popular any way you cut it.

The Bloggers vs. The Longform Writers

Daily bloggers are well-represented in the above data, but we also get this question from publishers quite a bit: “How does long-form content perform?” After all, with time-shifting and mobile devices, couch reading has never been better.

For this question, we took a different approach: The above charts are based on pure volume of articles saved, so that rewards the most prolific bloggers and tends to under-recognize those writers who publish longer pieces less often. But that doesn’t mean they are “showing up” any less. Those who channel their energy into fewer, but longer posts, can see the same great payoff.

Let’s compare six very different writers (all of whom we enjoy reading): TechCrunch blogger Alexia Tsotsis, ESPN’s Grantland writers Chuck Klosterman and Tom Bissell, Rolling Stone and New York Magazine writer Vanessa Grigoriadis, and New York Times reporters Jeffrey Gettleman and Susanne Craig.

Tsotsis is a prolific blogger: RIL users saved thousands of different stories with her byline over a 6-month period. Contrast that with the rest of the group: Grigoriadis and Bissell’s output is almost the polar opposite—they don’t publish as often, but when they do, it’s in bursts of epic 4,000-plus-word pieces. (See Grigoriadis’s recent “American Drug Lord in Acapulco” with Mary Cuddehe for Rolling Stone and Bissell’s longform video game review of “L.A. Noire.”)

You can see the differences in story output here:

Different styles, different output.

Tsotsis wins for pure story output, but when you look at average word counts per story, you see the rest of the writers make up ground:

Now let’s examine the “return rate” for each author. What we find is that they’re all much more alike than we first suspected. They’re all well above-average when it comes to readers returning to their work:

Loyalty and longevity pay.

This comparison can be apples-to-oranges: For instance, Alexia’s work is shorter, and there’s a possibility that she’s more likely to be read immediately. But what we’re seeing is an important new way to judge the quality of content, from the standpoint of loyalty to an author and longevity (or long-term usefulness) of their content.

The New York Times and the Best of Both Worlds

Meanwhile, a publisher like The New York Times has embraced both worlds: They have many bloggers on staff, but also investigative and longform reporters who spend months on a project. How do their authors stack up? Let’s look at their most-saved authors inside Read It Later:

Top Times saves: The list includes a cameo from writer Mona Simpson (who was featured for her heartbreaking eulogy for brother Steve Jobs).

Name-brand columnists, tech, politics and media are the big winners here. Columnist and economist Paul Krugman was the most-saved writer at the Times from May through October, followed by his Op-Ed columnist colleagues David Brooks and Thomas Friedman. The next most-saved came from the Times’ blogs: FiveThirtyEight data geek Nate Silver, then four Times tech writers: Nick Bilton, Steve Lohr, David Pogue and Jenna Wortham.

Do the names change when we look at the highest return rates for the Times?

Keeping Track of What Matters

There’s a lot of debate right now on the web about quantity versus quality—tracking quantity has always been easier than understanding the power of quality. We think there’s an important conversation to be had about how we use metrics like longevity and loyalty to better understand what’s important to readers.

Bloggers, long-form writers and how-to columnists all bring unique value to what they publish. They can operate on a different metabolism, so what’s important is whether readers keep showing up to enjoy your work, whatever form it takes.

Woody Allen, by the way, had a 42 percent return rate for his stories in Read It Later. I wonder how his films do?

Infographics by Katie Kosma.

—

Next: Get a sneak peek at Read It Later 3.0, coming soon for iPhone/iPad, Android and Kindle Fire.

Printed media used to allow us to read in the places we found most comfortable.  When you imagine yourself reading the newspaper it’s probably in your favorite chair, at the breakfast table, or at the cafe with an orange mocha frappuccino in your hand.

Unfortunately, as news and media moves online, it moves us away from these places and into our desk chairs.  Even worse, consuming content is no longer on our own schedule.  The flood of content disrupts us all day as if we have an maniacal paperboy throwing new editions on our doorstep every 15 seconds.

However, after studying Read It Later’s own data, it seems that this trend is being reversed.  I’ve found that as devices become more mobile, it’s not only changing where we read, but when.  Today, I’d like to show you some of the data behind this movement.

Today’s data source: 100 million articles saved by Read It Later users across all major web and mobile platforms.

Constant Bombardment

Before looking at when we read, we should look first at when we encounter new content:

This chart shows the number of articles saved to Read It Later each hour (adjusted for timezones).  As you would suspect, it is steady throughout the day as we are bombarded with content every waking hour.

So how are readers dealing with the deluge of information?

Time Shifted Reading

Readers are saving content for a better time.

People are busy.  It’s unrealistic that we are going to consume all of this content the exact moment we discover it.  So we leave dozens of tabs open, we email ourselves links, or we use Read It Later to hold on to the content until we are ready to consume it.

Let’s take a look at when different users get around to reading the content they save during the day:

Computer Users

This graph shows the number of articles read each hour by Read It Later users on their computer.

Compared to the times articles are saved, you can see that a significant amount of content was shifted towards the end of a user’s day (6PM – 9PM).  The graph is not as flat either.  Where the number of saves remains fairly constant between 8AM – 4PM, the number of reads grows more sharply until noon and then begins to fall off until after work.

Overall though, this graph isn’t a dramatic departure from the times we are saving content.  It seems that while on a computer, we are more susceptible to discovering additional content throughout the day.

iPhone Users

This is where it gets interesting.  You’ll note four major spikes when most of the reading on an iPhone is done:

• 6am – Early morning, breakfast
• 9am – The morning commute, start of the work day
• 5pm – 6pm – End of the work day and the commute home
• 8pm – 10pm – Couch time, prime time, bed time

In reality, this really is a graph of whitespace time. Whitespace is the time between A and B. It’s the time on the subway or bus. It’s the time standing in line. It’s a spare moment.

It is during these moments between tasks and locations that people reach for their phone. These are perfect times to knock an item or two off of your reading list.  By saving content for later, readers are able to consume content during the voids in their day without interrupting the day’s normal flow.

iPad Users

The graph of when users are reading on the iPad shows the biggest time for reading: personal prime time.

This is generally the most relaxing time of day.  After a long day, work is done, dinner is resting in your belly and there is nothing left to do but put your feet up and relax.  This time slot is the same one coveted by television.  When the majority of people are consuming content it seems perfectly natural that people would use this time to do their reading as well.

Not surprising, if you look back at the graphs for computer and iPhone reading, you’ll see spikes during this same time (8 – 10 PM) appear on all graphs.

How the iPad is Changing Online Reading

While many have speculated that the iPad is going to replace printed newspaper and magazines, it is already changing the way we read online content too.

The newspaper/magazine paradigm suits tablets extremely well.  They are portable while still being large enough to make reading enjoyable.

As I started by saying, the newspaper’s portability allowed us to read in the places we found comfortable.  More importantly, it let us read the day’s content on our own schedule.

The iPad is doing the same.

Look what happens if we rerun the number of articles read on a computer but only for users who own an iPad:

Aside from a quick lunch hour at their desk, iPad owners are no longer doing the majority of their reading on their computers.  They are saving it for their personal prime time, when they can relax comfortably, iPad in hand and burn through the content they found during the day.

What This Says About the Future of Reading Content Online

When a reader is given a choice about how to consume their content, a major shift in behavior occurs.  They no longer consume the majority of their content during the day, on their computer.  Instead they shift that content to prime time and onto a device better suited for consumption.

Initially, it appears that the devices users prefer for reading are mobile devices, most notably the iPad.  It’s the iPad leading the jailbreak from consuming content in our desk chairs.

As better mobile experiences become more accessible to more readers, this movement will continue to grow.  Readers want to consume content in a comfortable place, on their own time and mobile devices are making it possible for readers to take control once more.

Over the past 3 years, Read It Later has grown to be one of the world’s largest digital reading platforms.  Users have saved over 100 million articles and spent a combined 2.5 million hours reading last year in the iPhone and iPad apps alone.

This has created a pile of data.

As I explored this data in an effort to better understand how users were using RIL, I discovered a lot of really interesting insights into how the web, mobile and content spaces are changing.  Recent changes in these spaces are frequently dissected in the press, but often with little data. With access to such a tremendous dataset, I wanted to start providing that data and hopefully start answering the questions we all have.

After a lot of digging, typing and regret over closing my eyes through a few math classes, I’d like to introduce the new Trends series.  Today’s inaugural post is the one closest to Read It Later’s heart and attempts to answer the simple question: When are people reading?

You can follow along to future posts by subscribing or following @ReadItLater on Twitter.

If you have any questions you are dying to have answered or ideas for future posts, I’d love to hear from you: trends@getpocket.com

——-

A Quick Word on User Privacy

User privacy is incredibly important to me.  I want to make it very clear how user data is being used here (or more appropriately how it is NOT being used).  Trends only looks at generalized patterns regarding reading and never at any specific user’s data.  No individual user or reading list is viewed to compile these posts.  No specific user will ever be highlighted in one of these posts.  If you have any concerns about privacy, please read RIL’s privacy policy and do not hesitate to contact me.