DISCLAIMER: i've never really had any professional training in user interface design... and i'm sure that's obvious from my blog fonts & colors. However, i started designing & developing applications over 25 years ago, back in high-school when i developed some payroll reports one summer for my mom's company. Since 1991 or so, around when Object Builder and Visual Basic came out, most apps i've built or worked with have been in windows-based or browser-based graphical UI environments. And for the last 5-8 years or so, most startups & applications i've been involved and investing in are based on social networks & platforms. So whether or not i have any fucking clue what i'm doing, let's just say i've seen my fair share of designing visual user interfaces & interacting with social platforms.
If there's one thing i've learned from all that geeking around, it's that UI typically works best when it's butt-simple. As a famous PayPal colleague of mine once stated succinctly:
"Users are Stupid... give them something to click on."
Wise words. You'll do well not to forget them, Young Jedi.
However it's not always enough to simply give people underlined text links to click on... rather, it's important to have strong associated visual cues that encourage users to take action. Sometimes that can be as simple as just creating a beveled, slightly rounded, 3D-looking button with a color offset and some text that identifies a Call-To-Action (CTA). Many people have become trained thru years of working with operating systems to click on things that look like buttons. particularly BIG buttons. particularly BIG buttons with pictures or icons. they are almost irresistible. Go on. Click it! you know you want to!
But -- and i do mean butt -- even big buttons with big graphics aren't always going to capture user attention. What has gradually happened over the past 10 years is that online consumer interfaces have started to zero in on basic human behaviors, recognition systems, and patterns. many of those offline interactions start with the simplest of human interactions -- looking at someone's face.
In fact, you could argue that much of the online experience these days is less about reading text, and a lot more about looking at faces, icons, and other visual representations of people.
"The FACES, You Idiot... It's ALL About The F**KING FACES!"
About 2-3 weeks ago, i was meeting with a startup about some new widget workflow they were creating. The company shall remain nameless, but they were trying to develop a new UI around people exploring other people. However, the interface was still mostly comprised of text explanations and directives. Although it was being built as a social platform, they were missing the social part.
And of course this cuts right to the heart of the matter.
As the startup founder and i were discussing alternatives, i happened to check my inbox (because i think it had been at least 2-3 minutes since my last fix), and i noticed a new LinkedIn Network Updates email -- who got a new job, who made a new connection, etc. And i thought it was curious that there weren't any visual images in the email.
Now believe me, i am a HUGE FAN of LinkedIn, and i'm friends of a lot of folks over there. And they are making a LOT of money, and are very likely to be an IPO candidate way before Facebook. But i *do* think they have done some things fundamentally wrong over the years, and possibly the prime evidence of that is their very late adoption of user photos almost 4 years after launching and the overall sparse use of faces within a social networking platform. LinkedIn is about *professional* social networking, but i don't think being professional means you have to be boring or not engaged. in any case, LinkedIn is doing just fine but i'd strongly suggest they are missing out by not making their platform more visual, more social, more viral, more engaged, and more group-oriented. but perhaps that's a longer discussion for another post. i'm still hoping they get there soon.
there was at least one other example of FAIL that i can't remember that pushed me over the edge, and i finally had to scream and just tweet out:
for many folks, the state of the art in consumer internet development is to design big beautiful fonts and flash animations and green-blue web 2.0 colors. and for Google (at least until recently with Wave), the most advanced design elements they seem to come up with were still almostly exclusively text-driven.
IMHO, the two popular companies that have actually done the best job with user interface design optimization are YouTube and Facebook. and of those two, the one that has CLEARLY done the best job around UI for social platforms has been Facebook.
YouTube iterated furiously over the years to optimize UI elements & buttons, and while not always a success from a monetization standpoint, sheer usage & adoption has been nothing short of astonishing. Obviously they were doing something right. there was an incredible amount of iterative development, and an emphasis on UI element overlays on videos, and embeds on other sites, all of which led to massive distribution, engagement, retention, etc. the thing i thought was always smart about YouTube was that the interface was largely about browse-selecting one image from among several options. it was mostly a picture-driven UI. and guess what? many of those pictures are of people's faces.
Facebook -- hey, wonder whether *they* thought much about Faces? -- was started almost exclusively as a collection game around visual images. as it grew & morphed, the profile pages & pictures remained hugely important, however the subsequent innovations around NewsFeed -- again, extremely visual, lots of faces -- and also the overall Facebook Platform, and associated user invitation dialogs, all created interesting & engaging ways for users to click around, into, and on top of interesting visual data (usually faces). they continue to find interesting ways to personalize the user experience, and i don't see them slowing down at all.
that said, i do think there hasn't been much innovation around the user invitation dialogs, and that this is a place where i think a lot of interesting work could be done to a) select LESS # of people, b) who i CARE about more, c) make the faces BIGGER, and d) constrained by the CONTEXT of the current conversation keywords. altho the dialog below is about faces, it's still too many for me.
interestingly, Google Wave takes a MUCH different approach and utilizes a very graphic / picture-rich environment, and a relatively complex UI. while i haven't played with it yet, it does look pretty cool. still, i wonder if this isn't about 3 years late in the making... Google seems to be bragging about how the product has been in the making since 2007 or even 2004, but that seems like a bit of FAIL there... i mean, that's either an engineering fail for taking several years to get a product out the door (which still isn't live yet), or a management fail for not pushing them to get it done sooner. anyway, i guess i shouldn't shit on their innovation even if it's a little johnny-come-lately, but i would have to say it seems more like a reaction to the recent News feed & activity stream innovation that's been happening at Facebook & Twitter than original thinking. but maybe i'm being too harsh. what i will say is that gmail / gTalk already has some useful features based on simply associating faces with email & IM conversations. however, i'll also say that when i used to work with Jawed Karim at PayPal, before he went off to create YouTube he was showing me a visual IM client way back in 2004. perhaps the interesting & minimalist elements of Wave are derived from this very basic association... and if so then i'll be eager to see what they've put together when it ships. in any case, it's clear that Wave is emphasizing faces in a way that has never before been done at Google.
Twitter is another company that i might suggest has done some basic things right with face representations, altho the UI is rather spartan (unless you're using a Twitter client of one sort or other). and before Twitter, i thought MyBlogLog had done some interesting things with faces as well.
As i almost always suggest, the reference presentation for many of these ideas is something called "Putting the Fun in Functional" by Amy Jo Kim, Shufflebrain. AJ does a great job of explaining why the basic human psychology around collection behavior are well-suited for "collecting faces", and how this makes it easier for people to take the actions you want them to.
Brad Feld (& others) recently picked up my tweet and wrote his own piece on "all about the faces", however i think in that post he was discussing a somewhat different subject around whether faces or avatar icons work better in certain social environments. i think he also comes to the conclusion that faces are important, altho i also think warm, fuzzy, familiar avatar icons can also function in positive, user-engaging ways just like faces can.
anyway, this post has now gone on quite long enough, but the primary point i was trying to make is that many consumer internet applications can improve their UI & conversion by simply:
1) Keeping the overall UI simple
2) Use big, high-contrast 3D buttons with simple call-to-action text
3) Use large graphics & icons, and in particular use hi-quality zoomed-in faces
and of course, all of this should be done with rapid, iterative, a/b testing to measure results.
alright this really hasn't hit all the points i wanted to cover yet, but i'll stop here.
at some point in the future i'd like to talk more about how to improve invitation UI dialogs by filtering the available people & friends to focus on smaller subsets with larger facial representations. in other words, less people, who i care more about, with bigger pictures.
or then again, maybe just a big-ass picture of someone's butt would work just as well.