I would like to bring attention to a hot topic around the Liferay offices lately, and one that you should definitely consider for the websites that you are working on: behavioral analytics. I'd like to explain this with a small example that you might be able to relate to:
Let's all start with a lovely place that I call Cup Island:
I took this photo in the Liferay LA office last Friday morning. This is one of the small tables downstairs near our lounge/reading area. You can see this table immediately when you enter our North entrance. It is usually dotted with different mugs and glasses throughout the day. This is Cup Island.
As I was taking this photo, Shannon from our Marketing department said to me, "why are people leaving their used cups around? Someone should probably put them away."
And to someone who notices Cup Island for the first time, this is the natural response. In fact, I have left a cup here before, and had it disappear into the hands of the facilities cleaning team (which they kindly gave back to me after panicking over its disappearance).
But I explained to Shannon the purpose of Cup Island. "Don't worry, these cups aren't left here to die," (I'm paraphrasing a little bit), "this is a temporarily holding place until their owners come back for them."
You see, the reason that Shannon didn't realize this is because she is a woman. Wait - I don't mean this in a bad way - what I mean is, Cup Island only exists near the men's restroom.
So that is the purpose of Cup Island - a convenient place to put cups when using the restroom. To other people, this might only be Decorative Plant Island. It really depends on why you use the table.
So why use it to hold cups? Let's understand the audience. This restroom is near the kitchen, the only kitchen at our office, and the only place to get drinks from. And more than half of the employees at Liferay don't work on this side of the building, but instead work either upstairs or in another wing.
So when an engineer who sits upstairs needs to use the restroom, they are probably thinking, "I should also bring my cup with me so I can visit the kitchen in one trip." And right before they go into the restroom, they see this table, leave their cup, and pick it up when they exit the restroom.
So how can we improve this situation? Well, it depends on what we want to achieve. We cannot ask employees to stop drinking water, or to stop using the restroom (some of you may be thinking that asking for one may lead to the other anyway, but after a while our employees would cease to exist).
If our goal is to stop people from leaving cups on the island, we could try providing larger cups, so refills might not be as necessary. Or we could build a kitcken that is closer to the employees (or move employees closer to the kitchen). Or we could simply just remove the table. Employees may leave their cups further away, or just take two trips from their desks, which might cost them convenience and time.
If our goal is to make Cup Island easier to use, we could remove the huge plant and make more cup space. Or we could turn it into a Super Cup Island, with more capacity and a handy labeling system so people never get their cups confused.
Whatever we choose, these needs still exist: (1) everyone needs to drink water, and (2) everyone needs to use the restroom. But behavioral analytics tells us that the kitchen or bathroom may be too far away.
Learning to improve websites based on user behavior
For the redesign of liferay.com, we are trying to study where people are leaving their "cups" on our website. We know that there are community members, enterprise customers, trainees, partners, event goers, internal staff, liferay newbies, and evaluators - each of us enjoying the site, or fighting with it. We are studying how people use the site, where they are clicking, and trying to figure out why. If we can really understand our users, only then can we improve their experience. We're playing around with Crazy Egg, Google Analytics, and some other marketing tools. We're suggesting to the core product team that it would be killer to bake web analytics directly into Liferay Portal. Hey, it would make my job a lot easier.
Admittedly, this is not a new subject at all. Admittedly, Liferay has lots of room for improvement (the webteam would know, as we are working on our instance of 6.1 day in and day out). For Portal 6.2, we've started doing in-house user testing - I know this because I was just tested last week! And we've begun to make incremental improvements to the website. I already have examples of some small victories we've had lately, but I will save that for the next post.
We do have some lovely UI and content changes on the way. Just to prove how serious I am: the code is already committed and was approved by Brian Chan himself (well, he approves everyone's code). Stay tuned, same web time, same web channel.