A Brief History of the Complete Redesign of Google
Details: http://schedule.sxsw.com/2012/events/event_IAP9720
The designers are embedded deeply into the products. But there isn't a good stage for thinking about the design holistically.
The homepage is iconic so it hasn't changed very much over the years.
There was a redesign in 2007. Most people don't know about it.
The project was called Kanna. Icelandic for to explore, to examine, to investigate.
They looked to lots of design sources for inspiration. They identified over 100 brand attributes, which they then narrowed down to 4 clusters:
Modesty & minimalism
More fun
Organization
Daring
Both redesigns were started by request of the CEO.
They required a lot of introspection on the brand.
Driven by a group of designers that weren't embedded in any product & could look at it with fresh eyes.
In 2011…
Step 1: Strawman
They were asked by the CEO "If you were to redesign Google, what would it look like?" They thought he may be joking. There wasn't more information than that. They decided it would be dumb not to take it seriously.
Wiggins got together some really bold designers and used a sprint to come up with their dream.
They identified 10 or 12 representative screens and went to work on them. It's what they ended up showing to the CEO.
The redesigned screens were not that far off from where they are at today. They presented them to the CEO as before and after screens.
One launched, one didn't. Why?
In 2007, Kanna was a projected presentation. Strawman was printed on 11x17 80lb paper.
Kanna included the design process, whereas Strawman was just the before and after. They did this because the CEO got impatient about seeing the design process.
Before doing the designs, they took a design temperature. They cut out bits and pieces of the apps and design elements and had the CEO put together what he liked.
For Kanna, they presented many options. With Strawman, they presented an opinion.
Timeline
Strawman started in Jan. Larry became CEO and asked for the Google redesign in April. He wanted it launched in the summer.
They panicked but as one panelist said "they were like the dog that caught the car and wasn't sure what to do with it." They all jumped on board and moved forward.
Step 2: Kennedy
The new name of the effort. They compared it to the space age and the launch to the moon.
Google+ jumped on board and were the first ones to launch, even though it was originally designed differently. When it launched, it was given credit for the redesign.
Step 3: The Prototype
They created an HTML prototype to test CSS changes across products. It would give them a feel for how it would play out and how the grid could work across the products.
There were a lot of debates and nervousness internally about the changes.
They decided to focus only on modern browsers and use modern CSS techniques. It made it much easier for them to launch because of that.
There is a 218 ms fade on the buttons because it was someone's birthday.
They also used the prototype to demonstrate the flexibility of providing for different screen resolutions.
The developers ended up using a lot of the CSS as a reference.
Step 4: The Spec
They created a big interactive style guide. It has live prototypes and CSS so the developers can download what they need.
The spec helped the developers deploy the redesign across multiple products w/very little oversight from the design team.
People were generally really into the redesign and excited to implement it.
Step 5: Implementation: Managing Risk and Measuring Success
Google likes data.
Qualitative research. They showed 80 participants a variety of screens which included both the old and the new design for 10 seconds.
They gave them long enough to see it, but not long enough to rationalize why or why not they liked it.
Afterward they gave them a survey asking their opinion.
The new design outperformed the old design on a number of design metrics.
The old design still outperformed in some areas: comprehensive, less lacking, traditional.
Quantitive research
Google search provided platform. They have billions of queries each day. At any given time they are funning 200 experiments.
They shipped it to a small perceptage of users
Change aversion.
Change aversion exists. You'll see a dip when you launch something new. If you did it right, they will come back.
When they finished Gmail, they just switched it on. It became known as Gmailageddon. People erupted with their feedback. On day 1 people freaked out. By the time day 5 came around, if they ended up in the old design, they would complain.
There will always be people who are not happy with the change.
