What's with the hype about videos in Social Media?

Videos are currently being hyped in connection with social media. Look at your social media accounts and you will see that videos have become a dominant form of content. And apparently, the supply is matched by a corresponding demand. A company like Facebook would not place so much emphasis on videos if it hadn't come to the conclusion that this kind of content is particularly popular with users.


Indeed, a change in user behaviour seems to favor videos. The following insights draw on the Video Effects 2016 (Burda) study which, in turn, is based on a survey of the BurdaForward Panel (for the panel's socio-demographics cf.[1]). The term "video" is interpreted in a broad sense including films and series. Accordingly, the study summarizes video sharing-websites, media archives and streaming services among the providers. Here are some of the results:

• The willingness to watch videos has increased from 66 % (2015) to 86 % (2016).
• 49 % watch videos every day (2015: 35 %). 
• Videos are consumed most frequently at home (88 %). 
• Only 3 % use videos at school, university or work. 
• About 42 % watch videos up to 10 minutes in length. 
• 70 % watch videos on their notebooks or laptops, 57 % on their smartphones.
• Heavy video users prefer to watch them on the smartphone.

So, videos have indeed become increasingly popular. The usage is intended for private rather than educational or professional purposes which may be due to the fact that the range of videos available for educational or professional purposes is simply not large enough. Furthermore, successful videos must work on mobile devices because they have outstripped stationary devices at home too.

What motivates video consumption? 

Nearly eight out of ten respondents want to inform themselves by means of video and only every second person watches videos for entertainment. I would have seen the relationship the other way around.

Reasons for video consumption
... for information (78%)   ... for learning (35%)   ... because it's fun (50%)   ... as a pastime (50%)   ... to relax (45%)   ... because friends recommended or send me the video (28 %)


In terms of content, music, news and movies lead the ranking:

Information/Learning	Entertainment
news from politics (39 %)  instructions (e. g. computer programs, cooking, cosmetics, DIY) (29 %)  gaming videos (e. g. previews, tests, solutions) (14 %)  product tests (20 %)  household tips (14 %)	music (45 %)  movies (35 %)  documentaries (29 %)  series/soaps (29 %)  sports (19 %)  fashion (9 %)

The classification "information/learning vs. entertainment" does not appear in the study, and is, of course, worth discussing. Sports and documentaries, for example, can - depending on one's interests - also be used for information or learning purposes. But it does reflect the two broad reasons for video usage - information and entertainment - quite well and is consistent with the theory of uses and gratifications. [2]

Expanding the category "instructions (29 %)" further, the following picture emerges:

Drill-down: "instructions"
computer (hardware and software) utorials (57 %)  cooking (55 %) DIY (42 %)  handicrafts (28 %) gardening (22 %) cosmetics / styling (20 %)  photography (20 %) sewing (12 %) other (8 %)

A provisional conclusion can already be drawn here: Perhaps with the exception of computer/software tutorials, videos are mainly watched to see how things are done. Most probably, this how-to-approach also applies to computer hardware tutorials and software handling. What one doesn't find on the list is abstract information. And there's probably a reason for that: there's nothing to see.

Get to the point! 

It depends on the length of the video whether it is watched until the end. 75 % stop a video if it is too long (2015: 47 %). Unfortunately, the study does not include data on when a video is too long. Another source [3] (and probably in the context of advertising videos) has these following numbers:

Loss of viewers depending on video duration
after 10 seconds: 10 % after 20 seconds 20 %  after 30 seconds 34 %  after 60 seconds 54 %  after 2 min 76 %  after 3 min 83 %  after 5 min 90 %

After 60 seconds, half the audience has gone! The diagram describes the decreasing marginal utility of each further second. Unfortunately, the source doesn't give more detailed information about these numbers. My hunch is that they are applicable to videos with less than 5 minutes in length and that different "laws" are applicable to series or movies (binge-watching!). But the direction is clear. The viewer decides during the first minute whether she/he feels informed, entertained or bored.

Takeaways

• Videos do the trick if there is really something to see. Do you read instruction booklets? Or do you like following the dots, numbers, and arrows in graphical setup instruction? Probably not. A video shows how it is done - if necessary, repeatedly. The viewer does not have to translate an abstract description into real-life sequences of action. Thus, the video has a relieving effect on cognitive load. 
• What is true with texts is even more important for videos: Get to the point. A video is a grab bag for the viewer. She/he has to open it to see what's inside. A long but well-structured text, on the other hand, has a big advantage. It can be scanned to find the right place to get started. In principle, this is also possible with a video, e. g. with a transcription. But in this case, many viewers will stick to the text right away. 
• Make sure that you get the viewer through the first minute. If you want to inform or instruct the viewer open the grab bag for her/him. If you want to entertain the viewer create the necessary suspense as soon as possible. Whenever possible tell an engaging  story.

______________________________________________________________________________

[1] Burda (2016) Video Effects (2016). Link
[2]  Ruggiero, T. E. (2000). Uses and gratifications theory in the 21st century. Mass communication & society, 3(1), 3-37.  Link
[2] Optimal video length on the Internet (2015?) Link