Inside User Research at YouTube

If you think about watching a video online, it may seem pretty easy. A player, a play/pause button and some content. Done. But what about if the video is being played on a mobile phone? Or on a big screen? What if it's being viewed in Nairobi? Or Shanghai? Now let's say it's being viewed by someone who wants to share her thoughts on the video and by someone who wants to do nothing more than watch more videos. Before you know it, watching a video becomes more complicated than you realize.

Enter user research. While far from providing all the answers, it can help illuminate how the site is actually used -- as opposed to guessing how it might be used or assuming the user is just like the people designing the site.

So what exactly is user research like at YouTube? Sometimes it means letting users design their ideal experience. For example, last year we used a method called FIDO (first utilized by Fidelity Investments) where we cut out different elements of various video sites, stuck them on magnets, and had users arrange their ideal organization of the elements (see below for an example). Other times we use a more standard research method called a usability study, which entails seeing whether a user can or can't complete certain standard site tasks in a usability lab.

Sometimes having users come into labs is not enough, though; we want to understand how users use YouTube in their context, in their living room, with their laptop on their lap, sprawled out on the couch. In this case we might have field studies where we interview users in their homes. In addition to such qualitative research, we look closely at the behavior of millions of users through traffic analysis and try to understand what users think of the site by deploying thousands of surveys.

We still have a lot left to learn about how people use YouTube, but some things have become clear. One of the most important findings has to do with the difference between the large group of users who are on YouTube simply to watch videos and a smaller, but very important, group of more engaged users -- often uploaders. The latter group will, unsurprisingly, care about details like how to make communication with their audience easier and more effective, how to grow their audience, and even how to make money on YouTube. The former, on the other hand, want as simple of an interface as possible: "Just let me watch the video, please!" You can see this difference in the results of the FIDO experiment we mentioned earlier -- note how differently each kind of user arranged features on the pageand the sheer number of elements in the scenario on the right:

To make matters more complicated, not everyone fits nicely into one of the two aforementioned categories -- for instance, there are users who like to watch videos, but they might also occasionally comment or favorite. Their ideal experience likely looks like something in-between the two examples pictured above.

It's not always easy to know what the best balance is for everyone, but we are committed to working towards figuring it out. One thing we know for sure is that at the end of the day, we need to build a product that is easy to use and understand. If it becomes too complicated or cluttered, we'll all need to step back and think again.

If you have any thoughts on this, you can let us know here and see what a user research survey looks like if you haven't encountered one yet. In addition, if you'd like to participate in any upcoming research we have, please fill out this form.