What makes social media users stop scrolling and stop to watch a video? Brands and agencies the world over would love a definitive answer to that question – and Twitter think they have one.
A new study, conducted on behalf of Twitter by research agency Neuro-Insight, used neuro-science techniques to measure the brain activity of selected Twitter users as they navigated their timelines. Here’s what they discovered…
Brand videos work best
The cynical among you will probably raise an eyebrow at this, but apparently the study determined that video ads on Twitter outperformed ‘online video norms’ – though precisely what data they’ve used to make that comparison they don’t say.
Twitter users ‘personally identify with what they see, take in details at speed and can process it effectively’.
A slightly surprising find is that autoplay videos within the timeline elicit greater memory responses than for videos watched full screen. This is odd, since a larger, more immersive viewing experience ought to create a more lasting expression. According to Twitter, though, the ability to view the associated branding and Tweet copy when watching videos within the timeline is more impactful then when viewing full-screen.
Less is more
The study found that shorter videos are more effective when it comes to mempory encoding. More specifically, videos of 15 seconds or less in length are more likely than videos of 30 seconds to hang around in a user’s mind. Twitter points out that this is opposite to TV, where 30-second videos have apparently proven to be the most effective approach.
This finding makes Twitter – so they say – the perfect platform for video ads given the ‘uniqueness of the Twitter newsfeed environment’.
Another finding: videos don’t require sound, though it certainly helps if someone decides to watch the entire video. The study shows that personal relevance and memory encoding where the same for users watching the first three seconds of a video whether it had sound or not. But if someone watches a video in its entirety, audio will help the video have more of an impact – with dialogue being more effective than music.
First and foremost
The study also finds that the first video watched in a browsing session will have the biggest impact – giving Twitter the perfect chance to promote one of its newest premium products.
The first video watched generates (on average) a 22 per cent uplift across the various key metrics than subsequent videos watched. Want to take advantage of that? Handily, there’s a Twitter product for that. First View, more specifically, which enables brands to buy a top placement for their video content for 24 hours.
The study also touches on time of day, with the morning likely to enhance greater levels of personal relevance and memory encoding. According to Twitter, this is when brands should be sharing ‘tips and useful information’ because users will be more receptive.
So why do people watch videos?
According to the study, there are a series of creative features that will make users more likely to stop scrolling and watch a video:
- Making your story clear from the start is crucial. Apparently an ‘early story arc’ will make your video 58 per cent more likely to be viewed past three seconds.
- It might not be a surprise, but topical videos are 32 per cent more likely to be viewed past three seconds, and are 11 per cent more likely to be watched to the end. As Twitter suggests, ‘Aligning to something culturally relevant or time-specific triggers the brain to respond since there’s a level of familiarity’.
- Get people in your videos: seeing other humans increases the emotional intensity of a video by 133 per cent. This is key because elevated emotion leads to better memory retention.
- Subtitles or captions are also important, making a video 11 per cent more likely to be viewed and 28 per cent more likely to be watched to the end. Of course, if you’re using a sound-off approach you’ll be using text anyway.
To summarise, Twitter suggests that brands tailor their video to suit the timeline. This doesn’t necessarily mean creating a whole new suit of assets, but it is helpfully pointed out that ‘adaptations can be applied’ that will align content with the study’s findings.