Nieman Foundation at Harvard
The past year brought major changes to The New York Times. The social media desk’s editors joined a new department, Audience Development. Our team gained new resources as we combined efforts with colleagues who previously ran the Times’ Facebook page on the business side of the company, formalizing a relationship that had been more casual. Our desk now works alongside teams focused on search engine optimization, community management, newsroom analytics, and growth. We focus on setting standards for the distribution of Times journalism to broader audiences.
On the Times social media desk, we work to maximize the impact of our journalism on Twitter. But we aren’t alone in this work. Our colleagues on copy desks around the newsroom are excellent wordsmiths. And we can often rely on their judgments that what makes good headlines in print or on NYTimes.com will also be powerful on Twitter.
As one example, the story of a security guard’s run-in with the Secret Service during a presidential visit went out over our Twitter feed one evening with a headline that was similar to what appeared in the paper. It was a strong performer:
One Day in an Elevator With Obama, Then Out of a Job http://t.co/C3NM9HooUE
— The New York Times (@nytimes) November 3, 2014
The next day we tried sharing it from @NYTimes again with a guess at more Twitter-friendly language. While the tweet did well, the previous evening’s tweet was much stronger by our measures of clickthrough (as recorded by SocialFlow, like much of the rest of the data referred to in this essay).
A C.D.C. security guard was fired after operating an elevator carrying President Obama http://t.co/D5YqVMdr4T
— The New York Times (@nytimes) November 3, 2014
On a daily basis we publish many articles that need an alternative approach to attract readers that might come to us from social media. But there are also a significant number of instances where we shouldn’t try too hard to write a great tweet when other skilled journalists in our newsroom have already written one in the form of a headline. We’ve learned that lesson on our team, and also work to communicate it to desks around the newsroom that are more actively staffing their own Twitter accounts.
Photographs and other visuals all but took over Twitter’s main stream in 2014. It became not uncommon to see news organizations and even some journalists embed a photo in nearly every tweet. We expanded our use of visuals on @NYTimes, but in a way that served Times journalism and held us to our high standards for photojournalism.
Photos on Twitter are frequently distributed without any context. The 140 characters of space available in a tweet can make it difficult to credit a photographer or provide any information about what is depicted in the photo while also communicating the crux of a news article and including a link back to it. But that doesn’t stop news organizations or Twitter users from casting photos into the void with their tweets, bargaining that the increased engagement that images yield on Twitter make it worthwhile.
At the Times, we have a different set of rules for photography on social media. The two that are most important: images directly uploaded to Twitter must credit the photographer, and not all images used on NYTimes.com can be uploaded to social media because of rights issues. Fortunately, Times staff photographers and contracted freelancers produce a steady supply of imagery on a daily basis that is paired with our reporting online and can help make our Twitter accounts more engaging.
For much of the year, we included credits for photographers in the text of tweets. The results worked in many cases. But it looked clumsy and the space required to make room for the credit often made it difficult to write a better tweet. Moreover, the photos could be removed from the context of the tweet in which they were sent, and the source of the photo would thus be obscured:
This track team is peerless and seemingly ageless (Photo: Doug Mills/NYT) http://t.co/2EWCAtsCA8 pic.twitter.com/drzHgUN2HK
— The New York Times (@nytimes) July 24, 2014
We started working with our photo editors to watermark pictures with a photographer’s credit. An early project that used this approach was “The Way North,” Damien Cave and Todd Heisler’s road trip up I-35 to explore immigration’s impact on middle America. The promo images credited Heisler’s work, and also made clear that the tweets were part of a series:
A journey north to explore how the U.S. has been changed by immigration http://t.co/NU4Xw6E11R pic.twitter.com/3pgSnMXzKP
— The New York Times (@nytimes) May 31, 2014
The workflow of watermarking photos on a custom basis in photo editing software was inefficient. So our Interactive News developers built a watermarking tool that made it easy to add photo credits to any image social media editors wanted to use on Twitter. After testing the tool during New York Fashion Week in September, it became a standard part of our team’s editorial workflow. It preserved the Times pedigree of photos, and the crediting of our photographers became more engaging.
For the first time, more Americans support gun rights over gun control http://t.co/Yjzbo1yufQ pic.twitter.com/LAGoPaFejq
— The New York Times (@nytimes) December 11, 2014
Not every good or successful tweet requires an embedded photo. We counsel Times editors and reporters that using photos on Twitter is subject to editorial discretion and shouldn’t be resorted to reflexively. Sometimes the 22 characters taken up by a photo might be better used to write a more interesting tweet. And sometimes the additional engagement generated by adding a photo — retweets, likes, and replies — yields little additional clickthrough to our journalism on NYTimes.com, ultimately our primary goal.
We have many means to help promote our journalism on Twitter. But the question we always come back to is whether or not deploying certain promotional media is worth the extra effort it sometimes takes.
One example where a judgment call had to be made was the use of video trailers in tweets. Our video and international desks collaborated on a story about a man who survived an Islamic State massacre in Iraq. The video department produced a short video trailer that was embedded in some tweets we sent about the documentary:
The only known survivor of an ISIS massacre tells his story (warning graphic video): http://t.co/EIohT3aEn6 http://t.co/4i39ops934
— The New York Times (@nytimes) September 4, 2014
The tweet with the video trailer performed quite well. But other tweets that used a still image from the video — produced and uploaded with much less effort than the embedded trailer — substantially outperformed that tweet when clickthrough was measured. The difference in performance could possibly be explained by the trailer dissuading potential viewers to click through to the full video.
The only known survivor of an ISIS massacre in Iraq tells his story (warning: graphic video) http://t.co/BA4n35tdEg pic.twitter.com/TY1YwP2bs8
— The New York Times (@nytimes) September 4, 2014
There could be important journalistic reasons to use a video trailer, an animated gif, a specially produced image card, or other multimedia to promote a story on Twitter. Some of these methods may also be useful on other social media platforms like Facebook or Instagram. But sometimes a simpler, less-labor-intensive approach yields a strong — or stronger — performance.
Times journalists take great pride in the work that we do to report the news and tell stories. In addition to the skilled journalism that goes into the work, there’s a prodigious vocabulary that goes into describing what is special and unique about the finished product. And sometimes all the effort to describe our hustle can get in the way of telling potential readers about the great story they’re about to be exposed to.
A powerful entry from the Times’s Op-Doc series about a man going blind was illustrative of this problem and its solution. The first impulse in tweeting about the story was to highlight the series it was a part of, and the power of the audio diaries that were the basis of the video:
Op-Docs: A short documentary on a writer who became completely blind and chronicled the experience in an audio diary http://t.co/R9knPDvDMJ
— The New York Times (@nytimes) January 17, 2014
As we “peacocked” the story — displaying its fancy plumage — it underperformed on Twitter. So we tried another approach that focused on enticing readers with key details of the story. In this instance, we made no mention of the multimedia or the series it was a part of:
As he lost his eyesight, he knew he couldn't let blindness destroy him http://t.co/kWMQzX8Jtd
— The New York Times (@nytimes) January 19, 2014
The clickthrough to the documentary doubled. The story itself resonated more than its constituent parts. When we set out to promote other major works of journalism, we focus our efforts on the resonance of the story our journalists are telling, and not the ways in which they tell it.
On a number of occasions this year, @NYTimes saw major spikes in engagement from Twitter users. But that wasn’t always good news. In some instances, consumers of Times journalism were angered by articles or columns written by our journalists. They often directed their ire more at the institution than they did at the individual journalists who had written the items that caused offense.
When other types of companies face a maelstrom of outrage on social media, they tend to use their social media platforms to respond to people who have been angered, seeing them in part as customers or potential customers. But as a news organization, we expect @NYTimes to remain above the fray to a certain degree, delivering our journalism and not getting caught in the middle of how it is received.
It’s difficult to imagine the Times as an institution responding to individual Twitter users like an airline might respond to upset passengers. But finding a suitable way to recognize and engage sincere criticism of our journalism that reaches us via social media would be a suitable use of such platforms. In 2015, we need to spend more time thinking about ways to be responsive to readership that comes to us from social media.
For all of our efforts to plan ahead and maximize the audience around enterprise journalism, Twitter users continued to come to @NYTimes for our coverage of unfolding news events that we don’t always anticipate. The shooting of Michael Brown in Ferguson and its aftermath; the rise of ISIS in Syria and Iraq; the deaths of Robin Williams, Joan Rivers, and Philip Seymour Hoffman; Michael Sam coming out; Ebola’s spread through West African countries and beyond; and more. Large numbers of Twitter users who follow @NYTimes were ongoing and substantial audiences for tweets about these news events. The news we broke about an autopsy of Michael Brown was the most clicked on @NYTimes tweet we’ve recorded.
Breaking News: Autopsy Shows Michael Brown Was Struck at Least 6 Timeshttp://t.co/T17F2bwd0y
— The New York Times (@nytimes) August 18, 2014
Our readiness to cover these events as they emerged helped maximize the impact of Times journalism on @NYTimes and across Twitter more broadly. But beyond @NYTimes, we found effective coverage of major news events to be one of the most important means of raising the follower counts for other Twitter accounts.
Several desks in the Times newsroom used Twitter to thoroughly cover anticipated news events like Hollywood’s awards season, the World Cup in Brazil and the events leading up to and following the grand jury’s decision on the Michael Brown case. In each instance, the Twitter accounts that provided wall-to-wall coverage of these events gained followers at rates that could probably only be matched by buying them. It demonstrated around our newsroom that investing effort in using social media platforms to share news had a measurable impact in growing the audience for a desk’s journalism.
This essay has been pretty serious. But the Times social media desk tried to have some fun on @NYTimes and other accounts this year. As we explained last January, the tweets we wrote that produce the “ZOMG!” moments aren’t always our greatest hits in terms of clickthrough or engagement. But finding the right opportunities to veer away from our institutional voice helps leaven our daily coverage of news and creates different kinds of memorable moments that readers enjoy.
Any newspaper’s copy chief or front page designer might tell you the same thing about the moments when they stick out their ink-stained necks. For that reason, I’ll close this year with several examples where we sought to strike that balance:
Autocorrect has come a long wa http://t.co/64Tt1RvUVt
— The New York Times (@nytimes) June 6, 2014
On Taylor Swift, a NYT dance critic has a take take take take take http://t.co/PjMjWqlXRw
— The New York Times (@nytimes) August 20, 2014
A brief history of kissing in movies http://t.co/a7uClvbUat pic.twitter.com/HE19lusyii
— The New York Times (@nytimes) December 11, 2014
It's Mike D. with the master plan … for a townhouse in Brooklyn: http://t.co/aDWhbHwIH7
— NYT Metro (@NYTMetro) September 15, 2014
SOME CONTRACTORS IN NEW YORK CITY ARE EXPERIMENTING WITH NEW, QUIETER ELECTRIC JACKHAMMERS http://t.co/DWrALoEKO8
— The New York Times (@nytimes) December 20, 2014
We hope that 2015 yields more memorable moments produced by the journalists who help run the Times’s social media platforms.
Thanks to Hanna Ingber, Sona Patel, Daniel Victor, Talya Minsberg, Karen Workman, Justin Bank, and Cynthia Collins for their contributions.
Michael Roston is senior staff editor for social media at The New York Times.
