Despite its critics, autoplay video is fast becoming ubiquitous across social networks.

Twitter recently announced that all native videos, GIFs and Vines uploaded to the platform would soon autoplay in timelines by default, following the lead of Facebook, which introduced autoplay on desktop and mobile in December 2013.

However, while autoplay may offer benefits in terms of engagement and video views, it poses issues for news outlets publishing stories featuring graphic or upsetting content.

“If we could do it all again… there would have been a discussion about whether or not we should upload the video at all,” explained New York Times social media editor Cynthia Collins, speaking to TheMediaBriefing. 

“But if we decided to upload the video I absolutely would have added a warning.”

The video Collins is referring to was uploaded on July 20 to the Times Facebook Page, which has almost 10 million followers. Part of a series called The Outlaw Ocean by Times journalist Ian Urbina, it showed disturbing footage of several unidentified men being shot and killed at sea.


Screenshot from The New York Times

“We had pretty immediate feedback from users,” said Collins. “One was asking: ‘How do I turn autoplay off? This is offensive, I don’t want to see this,’ and then two [were] criticising us for displaying this video in their feeds on autoplay.”

Although Facebook users can disable autoplay under their account settings, by default videos will play automatically whenever a device is connected to Wi-Fi.

Collins said the video received “dozens” of comments within an hour or two of being posted to Facebook (it was not posted to Twitter).

Not all of the comments were about the autoplay – Collins noted that many were discussing the actual story. However, the Times felt that the criticism was valid and decided to delete the post from its Facebook Page.

“We reviewed the feedback and decided that we likely should, at the very least, have included a warning message within that video post,” said Collins.

“We considered editing the post to add that context, but we just decided it would be best to remove it entirely, delete the video from further distribution, and then post a link to the article with an explanation as to why we deleted it.”

The subsequent post acknowledged that “an auto-play video of this nature may be disturbing to some users” and instead linked off to the story on the Times site, requiring users to make a proactive choice about whether to click through to watch it.

The video itself was also edited to include a two-send title slide warning about the “graphic scenes of violence” at the beginning.


Earlier today on Facebook we posted an auto-play excerpt of a video, linked below. We recognize that an auto-play video…

Posted by The New York Times on Monday, July 20, 2015

Why it matters: Social video is becoming native video

Facebook’s introduction of autoplay for native video came as the network stepped up competition against its main rival in the video-sharing space – YouTube.

The network also tweaked its algorithm to prioritise native video, which now hugely outperforms YouTube on Facebook in terms of uploads, shares and interactions, according to Business Insider.

At the end of Q1 this year, Facebook could boast 4 billion video views every single day, up from three billion in January and one billion back in September 2014.


Screenshots from Business Insider

While YouTube videos are still the most commonly-shared videos on Twitter, native video receives more interactions.

It seems natural that native video will perform better on its home turf than third-party videos which, given the tendency towards autoplay, raises a few new considerations for publishers

The rise of social video

Increasingly, people are getting their daily intake of news from social media.

Coupled with the fact that Facebook’s algorithm prioritises native video, apparently making it 30 per cent of overall newsfeed content, many publishers are now uploading more video to Facebook in a bid to boost engagement and reach new audiences.

Research by NewsWhip, published in May, showed that while links were still the most common post format on publishers’ Facebook Pages, video posts achieved higher than average levels of engagement.

On the BBC News page, for example, videos were getting an average of four times the number of shares as links to the BBC website.

The Al Jazeera offshoot AJ+ recently announced its videos have been viewed more than 430 million times on Facebook over the last 90 days.

While the success of native video on Facebook can’t be completely credited to autoplay – smart production, intriguing content and algorithmic favour also help – it appears to be an effective way of compelling users to stop scrolling through their newsfeed, even just for a few seconds.

During testing of Twitter’s new autoplay feature, users were 2.5 times more likely to prefer autoplay videos over other viewing options, according to a blog post by senior product manager David Regan.

And of course, autoplay offers big benefits for advertisers in terms of engagement and video views.

“It’s just so immediately engaging… when that video starts playing you’re sucked in,” explained Collins. “It does pull you into a story, perhaps sometimes more powerfully than a static image and text.”

Autoplay in news: Best practice

While most news outlets have guidelines in place regarding the publication of graphic or disturbing content on their own sites and channels, it seems not many have considered how these may need to be adapted for social media.

The New York Times is by no means the only news outlet to have been caught out by autoplay video – although it’s difficult to monitor how often this happens as not every organisation is transparent about when posts are deleted.

An NBC regional news channel based in Rhode Island was criticised by Facebook users earlier this year after uploading a video showing a California firefighter falling through the roof of a burning building.

And it’s not just video that’s a problem on social media. Graphic images have the potential to be just as distressing – especially to social media users who haven’t actively sought them out.

For example, BBC News received critical comments last month after posting an image to Facebook of a woman in China who fell into an escalator after a panel gave way, and later died.

Although readers may want to be kept informed about news, a simple Twitter search suggests many people have come across autoplay video on social media they would rather not have seen.

Recent research from Eyewitness Media Hub, a non-profit formed to establish best practice for use of eyewitness media in news, suggests many people prefer journalists to explain events which may be distressing, rather than seeing it for themselves.

In a series of focus groups earlier this year around news coverage of the Charlie Hebdo shooting in Paris, two members discussed their experiences of using Facebook, where video was posted showing unarmed police officer Ahmed Merabet shot dead by ISIS militants.

Lindsey: A lot of people who don’t want to see it [graphic/upsetting content] are seeing it anyway because they’re just scrolling through Facebook.
Elaine: Yeah, it’s like your choice is taken away… You can report it [to the social network], but you’ve still seen it.

“I think there is that element of being desensitised in one sense, because the journalists are working with it [graphic content] all day, every day,” Jenni Sargent, director of Eyewitness Media Hub, told TheMediaBriefing.

She believes news outlets also need to be more considerate of the context of consuming news on social media, recognizing that stories need to be packaged differently than they might on a news site or for television.  

Many people use social media to connect with friends and family as well as to follow news.

Coupled with the fact most people access social networks via mobile – a device that tends to feel highly personal due to the amount of time we spend with it – this means unsolicited graphic content may feel even more intrusive to those who don’t wish to see it. Sargent said:

“When you sit down in front of the Six O’Clock News you know you’re about to consume news and you sort of brace yourself in a way. whereas when you’re online scrolling between wedding photos and announcements of new babies you’re not prepared, so it’s even more shocking.”

The best way to warn users about graphic video content on social media is within the wording of the post, said Sargent.

However, outlets must be careful about the wording they use to avoid coming across as clickbait – inviting the digital equivalent of rubbernecking at a traffic accident.

Sargent advises giving as much context to the warning as possible, clearly stating what the video shows using wording such as “this video contains scenes of… or shows the moment of…” to enable viewers to make an informed decision about whether they want to watch it.

Within video, outlets should avoid formatting warnings “like a film trailer,” said Sargent, erring on the side of formal rather than anything which might appear “jaunty,” such as text animation.


Screenshot from The New York Times

However, even a worded warning isn’t all that effective because unless users have turned autoplay off, the video will start to play anyway.

Perhaps then, the onus should be on the likes of Facebook and Twitter to introduce a setting for publishers to disable autoplay for potentially distressing content.

However, as Sargent points out, the thumbnail image is also something journalists should consider.

“That thumbnail should be the graphic content warning if nothing else,” said Sargent. “It shouldn’t be any other part of the video.”

At The New York Times, Collins says the social media team has learned from the comments it received on Facebook.

“Essentially we’re just going to be much more thoughtful about the videos that we upload natively,” she said.

“If we anticipate there could be some disturbing imagery we will always have a discussion as to whether or not we think it’s worth uploading the video to Facebook, or if it’s perhaps better to just link off to the video on our site, so that it would require a bit more of a proactive decision by the user.”

Any videos that could be potentially disturbing will in future be accompanied by an initial title card featuring a warning, added Collins, possibly together with a written caution within the post.

“Overall, I would say we’re happy with our performance of video on Facebook and we can do a lot more with it, but this is just a sensitive area that we need to be careful about.”

Twitter is currently rolling out autoplay on the web and iOS, while Facebook is reportedly testing a continuous autoplay feature – serving up another video after the first one finishes.

And with autoplay being a key feature of both Facebook Instant Articles and Twitter’s forthcoming Project Lightning, it’s a good time for news outlets to consider updating their social media guidelines.

How to turn off autoplay video

Turning off autoplay on social networks will reduce the risk of accidentally seeing videos you would prefer to avoid. On mobile, it also improves battery life and won’t eat into your data plan.

Here’s how to manage autoplay on Twitter, Facebook and Instagram.



On mobile: Settings > Videos and Photos. Toggle Smart Auto-play to off and then select Never Play Videos Automatically.

On the web: Settings > Videos (in the left sidebar). Under Auto-play Videos, select Off from the dropdown.



On mobile: Settings > Video autoplay. Select Never play videos automatically.

On the web: Settings > Account > Content. Untick the autoplay box next to Video Tweets.



Instagram removed the option to disable autoplay in October 2013. By default, the app preloads videos so they start faster whenever possible.

However, you can turn off preloading in your account settings. Go to Mobile Data Use and choose Use Less Data.

While this won’t turn off autoplay completely, videos may take longer to load.

Note that choosing this option won’t affect video preloading when connected to WiFi.