I’m Feeling Lucky, But You’re A Different Story

Many web redesign projects are completed without the user experience (UX) people ever talking to the analytics or social media folks. I always knew this was a bad idea, but never so much as last week. My wife and I were driving somewhere and our conversation turned to technology. When I mentioned the “I’m Feeling Lucky” button on the Google search page, she stopped me, asking “The what button?” I was flummoxed. Here was someone who teaches computing and is a loyal Google user. Yet this ubiquitous button, the yin to the “Google Search” button’s yang, was invisible to her.

This is all too common. People operate under certain habits and expectations. In new situations, such as visiting a new or redesigned web site, mental barriers kick in. The demons that blind us to a site’s “I’m Feeling Lucky” buttons are called cognitive biases. There are lots of these troublemakers, but the culprit here is the Homer Simpson effect.

Its real name is confirmation bias, which is the habit of filtering new information in a way that confirms preconceptions. We see it in politics all the time, as we chuckle knowingly at the news of another family-values politician caught with his pants down (and sometimes hers), or when conservative pundits perceive a newly inaugurated U.S. President as socialist for using the power of federal spending to avert another Great Depression. We look for what we’re expecting to see, whether it’s the silhouette of two faces or of one vase.

I prefer calling the confirmation bias the Homer Simpson effect because I love this example from The Simpsons: It was the mid-90’s and Homer’s TV was on. He had stumbled upon a televised episode of Garrison Keillor’s A Prairie Home Companion. Homer is briefly transfixed by Keillor’s folksy monologue, then suddenly flies into a rage, hammering the TV and yelling, “Be more funny!”

Homer had expected one thing from the television and was confronted with something decidedly different.

How does a “decidedly different” user experience benefit from social media and analytics support during its launch? Here are two ways:

1. The social media team works to add the context necessary for users to interpret the site’s new design and content
2. The analytics team monitors user acceptance and provides insights for potential tweaks

Here’s an example: If Google had launched their search engine today, they might blog about the “I’m Feeling Lucky” button and encourage discussion and reactions on Facebook, Twitter and elsewhere. The launch would be accompanied by real, live intermediaries who would converse about the novel new button.

These conversations would be with (admittedly) only the most plugged-in and social-media-savvy. But that’s okay. In essence, social media discussions would provide another level of beta testing, and serve as grassroots promotion of the new button. Those who are part of the conversation would make it “cool” to know about Google’s audacious new UX.

What Would Google Do NEXT?

In this example, what happens next is measurement. The analytics team would create a new segment to watch, comprised of those who clicked through from social media posts and those who searched about the button before arriving at the new site. (This second group would be identified in the web analytics system. They’re the organic search visitors who used that specific term to click through to the site.)

This small but significant segment would be compared to the majority who didn’t arrive that way, so differences in behavior can be reviewed for insights.

Every new web site has “I’m Feeling Lucky” buttons. These are the features that are new, either to your site or to web browsing in general. If this new UX adds value, go ahead and use it. Just be sure to include them in your social media efforts, and measure scrupulously.

The features that seem as clear as day to you may be completely invisible to many of your best prospects and customers.