The Yellow Submarine Reaction™ aka The Tendency To Skip Content

One of my all-time favourite albums is Revolver by The Beatles. I've loved it ever since I heard it on my Dad's record player as a boy. First released in 1966 in LP format - or long player for those who don't remember vinyl - you listened to Revolver in two parts. First you played Side A, then you carefully turned the record over and played Side B. With a good eye and a steady hand you could 'skip' between the individual tracks, but you risked scratching the vinyl in the process.

When cut at 33 rpm most LPs held about 22 minutes of music on each side. Revolver has a total run time of 35 minutes and none of its 14 tracks is more than 3 minutes long. Not keen on Yellow Submarine? Well, it's only 2 minutes and 39 seconds before you get to the goodness of She Said She Said. That is unless you are listening to Revolver on CD. As soon as Ringo drones, "In the town where I was born...", it's difficult to resist the urge to press skip.

Sorry Ringo.

OK, but what has this got to do with e-learning?

This week's E-Learning Heroes Challenge was Audio Interviews with Course Designers. In his introduction, David Anderson listed ten interview questions and suggested we host our audio files in Articulate Rise or upload them to a third party service like Soundcloud. The number of tracks immediately jumped out at me - too many for an 8 Track, but what about a cassette player? Could I tackle this digital project as if it was an analogue one? What about an old-fashioned mix tape?

Analogue vs Digital

Radio buttons are a classic example of a user interface with roots in the pre-digital world. Like the controls of an old radio or television, you can only press one radio button down at a time. Select one and you deactivate the others.

I used this approach in the first version of my 'cassette player', with the added feature that you could not select another track until the current track had finished.

But why did I do this?

• Honestly, it was easier to make my demo work this way. It ensured that only one track played at a time and made it simpler to sync the animated cassette tape with each track. 
• My longest track was only 27 seconds long. Even if the listener had a Yellow Submarine Reaction ™, they wouldn't have to wait for very long until they could select another track.
• The navigation wasn't completely locked down - the listener could still play the ten tracks in any order - but the pared down controls added to the charm of the design. It was a purposefully 'Generation X' take on e-learning.

There's No School Like The Old School...

... as the saying goes. But it turns out that nostalgia is no substitute for functionality. We now consume media very differently than we did in the pre-digital world.

Here are some early reactions to the first version of my demo:

 Why the locked nav Jonathan? Set me free ;) Sorry, I wanted to skip across topics

 I also looked for the skip - but we can’t skip a mix tape!

'Technology doesn't do sentimental,' as Matthew Finnie writes in this article for Wired. Course designers can learn from the way our listening habits have changed - and our attention spans have shortened - as we have moved from vinyl to cassette, compact disc to download and now streaming services. The cost convenience curve has plummeted. Today we expect more content for less time and effort.

Skip the box, skip the media and the album. I just want the song. In fact, I don’t have the patience to download, I’m going to stream it… and so on we go.

To put this into perspective, when music blogger Paul Lamere analysed the 'skip rate' on Spotify he found a:

• 24.14% chance that users will skip a song in the first 5 seconds
• 28.97% chance that users will skip a song in the first 10 seconds
• 35.05% chance that users will skip a song in the first 30 seconds
• 48.6% chance that users will skip a song before it finishes

Suddenly I felt like Ringo. Which one of my tracks was Yellow Submarine?

Every day's a school day...

... as I'm fond of saying. This was a great opportunity to carry out a live A/B test on Twitter on LinkedIn. I re-jigged my demo to unlock the navigation and posted both versions online:

For a while Version A and Version B were neck and neck. Early feedback also reassured me that my first instinct - to lock the audio down - was the correct one:

Well done Jonathan - if this was a learning experience, of course you would lock the audio down - so I voted for A - even though I did try to skip ahead!

Pretty soon though, Version B began to pull ahead in the poll. It was becoming clear that more users preferred having the freedom to skip through the tracks.

The quick to the point statements make it easy to pick A, but B is more applicable to the pull method.

I wondered whether displaying the run time of each track would reduce or increase the impulse to skip. In other words, if you knew the duration of Yellow Submarine would you give Ringo a chance? So I incorporated a timer into Version C of my demo and added this to the A/B test.

Interestingly, this poll suggested that users not only preferred having the freedom to skip media, but also liked to know the duration of media too. However, a 'ticking clock' can be a distraction, as two users highlighted:

I feel like I’m weird but I still kind of like A. The ability to skip around was more fun to play with than listening, and adding the timer just makes my brain skip to the completion and start thinking about the next one.

I like the unlock version best - I appreciated having the option to select in the order of my choosing and thought the clock was distracting. Great job!

Yet other users found the timer reassuring and discouraged them from skipping:

I like to know things. So I like Version C :)

Having the runtime for each track keeps me listening and prevents me from skipping ahead.

Below is a side-by-side video of all three user interfaces. Which do you prefer and why?

Final Thoughts

This limited experiment raised more questions than it answered but underlined the importance of user testing when developing a new e-learning product.

• Try as we might, it's difficult not to project our own preferences onto a project. One person's Yellow Submarine is another person's She Said She Said.
• Easier to build does not mean easier to use. Making life easy for yourself might actually make things more difficult for your users.
• The tendency for users to skip content is strong and will only increase with the growth of streaming and on-demand services. This will continue to influence how users interact with e-learning, for better or for worse.
• However, be cautious about linking course navigation to a media seekbar - this can allow users to fast forward through a course without taking in the content.
• Content is king, the same as it ever was. But we now have a smaller window of time to convince users to watch or listen to media content. Don't hold users hostage with locked navigation; grab their attention with interesting and well presented content instead.