English-Video.net comment policy

The comment field is common to all languages

Let's write in your language and use "Google Translate" together

Please refer to informative community guidelines on TED.com

TED2014

Margaret Gould Stewart: How giant websites design for you (and a billion others, too)

Filmed:

Facebook’s “like” and “share” buttons are seen 22 billion times a day, making them some of the most-viewed design elements ever created. Margaret Gould Stewart, Facebook’s director of product design, outlines three rules for design at such a massive scale—one so big that the tiniest of tweaks can cause global outrage, but also so large that the subtlest of improvements can positively impact the lives of many.

- User experience master
At Facebook (and previously at YouTube), Margaret Gould Stewart designs experiences that touch the lives of a large percentage of the world's population. Full bio

What do you think of when I say the word "design"?
00:12
You probably think of things like this,
00:16
finely crafted objects that you can hold in your hand,
00:18
or maybe logos and posters and maps
00:22
that visually explain things,
00:25
classic icons of timeless design.
00:27
But I'm not here to talk about that kind of design.
00:30
I want to talk about the kind
00:33
that you probably use every day
00:34
and may not give much thought to,
00:36
designs that change all the time
00:38
and that live inside your pocket.
00:40
I'm talking about the design
00:42
of digital experiences
00:44
and specifically the design of systems
00:46
that are so big that their scale
00:49
can be hard to comprehend.
00:51
Consider the fact that Google processes
00:53
over one billion search queries every day,
00:55
that every minute, over 100 hours
00:59
of footage are uploaded to YouTube.
01:01
That's more in a single day
01:03
than all three major U.S. networks broadcast
01:05
in the last five years combined.
01:08
And Facebook transmitting the photos,
01:11
messages and stories
01:13
of over 1.23 billion people.
01:14
That's almost half of the Internet population,
01:17
and a sixth of humanity.
01:20
These are some of the products
01:23
that I've helped design over the course of my career,
01:24
and their scale is so massive
01:27
that they've produced unprecedented
01:29
design challenges.
01:31
But what is really hard
01:33
about designing at scale is this:
01:35
It's hard in part because
01:38
it requires a combination of two things,
01:39
audacity and humility —
01:42
audacity to believe that the thing that you're making
01:45
is something that the entire world wants and needs,
01:49
and humility to understand that as a designer,
01:52
it's not about you or your portfolio,
01:55
it's about the people that you're designing for,
01:58
and how your work just might help them
02:00
live better lives.
02:02
Now, unfortunately, there's no school
02:04
that offers the course Designing for Humanity 101.
02:07
I and the other designers
02:11
who work on these kinds of products
02:12
have had to invent it as we go along,
02:14
and we are teaching ourselves
02:17
the emerging best practices
02:19
of designing at scale,
02:21
and today I'd like share some of the things
02:23
that we've learned over the years.
02:25
Now, the first thing that you need to know
02:27
about designing at scale
02:28
is that the little things really matter.
02:29
Here's a really good example of how
02:32
a very tiny design element can make a big impact.
02:34
The team at Facebook that manages
02:38
the Facebook "Like" button
02:41
decided that it needed to be redesigned.
02:43
The button had kind of gotten out of sync
02:45
with the evolution of our brand
02:48
and it needed to be modernized.
02:49
Now you might think, well, it's a tiny little button,
02:51
it probably is a pretty straightforward,
02:53
easy design assignment, but it wasn't.
02:55
Turns out, there were all kinds of constraints
02:58
for the design of this button.
03:00
You had to work within specific
height and width parameters.
03:02
You had to be careful to make it work
03:05
in a bunch of different languages,
03:07
and be careful about using
fancy gradients or borders
03:09
because it has to degrade gracefully
03:12
in old web browsers.
03:14
The truth is, designing this tiny little button
03:16
was a huge pain in the butt.
03:19
Now, this is the new version of the button,
03:21
and the designer who led this project estimates
03:23
that he spent over 280 hours
03:26
redesigning this button over the course of months.
03:29
Now, why would we spend so much time
03:33
on something so small?
03:35
It's because when you're designing at scale,
03:37
there's no such thing as a small detail.
03:39
This innocent little button
03:42
is seen on average 22 billion times a day
03:44
and on over 7.5 million websites.
03:48
It's one of the single most viewed
design elements ever created.
03:51
Now that's a lot of pressure for a little button
03:55
and the designer behind it,
03:58
but with these kinds of products,
03:59
you need to get even the tiny things right.
04:01
Now, the next thing that you need to understand
04:04
is how to design with data.
04:06
Now, when you're working on products like this,
04:09
you have incredible amounts of information
04:10
about how people are using your product
04:13
that you can then use to influence
04:15
your design decisions,
04:17
but it's not just as simple as following the numbers.
04:18
Let me give you an example
04:21
so that you can understand what I mean.
04:22
Facebook has had a tool for a long time
04:25
that allowed people to report photos
04:27
that may be in violation of our community standards,
04:29
things like spam and abuse.
04:32
And there were a ton of photos reported,
04:34
but as it turns out,
04:36
only a small percentage were actually
04:38
in violation of those community standards.
04:40
Most of them were just your typical party photo.
04:43
Now, to give you a specific hypothetical example,
04:45
let's say my friend Laura hypothetically
04:48
uploads a picture of me
04:51
from a drunken night of karaoke.
04:52
This is purely hypothetical, I can assure you.
04:55
(Laughter)
04:59
Now, incidentally,
05:00
you know how some people are kind of worried
05:02
that their boss or employee
05:03
is going to discover embarrassing photos of them
05:05
on Facebook?
05:07
Do you know how hard that is to avoid
05:08
when you actually work at Facebook?
05:10
So anyway, there are lots of these photos
05:14
being erroneously reported as spam and abuse,
05:16
and one of the engineers on the team had a hunch.
05:19
He really thought there was something else going on
05:22
and he was right,
05:24
because when he looked
through a bunch of the cases,
05:25
he found that most of them
05:27
were from people who were requesting
05:29
the takedown of a photo of themselves.
05:31
Now this was a scenario that the team
05:34
never even took into account before.
05:36
So they added a new feature
05:38
that allowed people to message their friend
05:40
to ask them to take the photo down.
05:42
But it didn't work.
05:44
Only 20 percent of people
05:46
sent the message to their friend.
05:47
So the team went back at it.
05:49
They consulted with experts in conflict resolution.
05:51
They even studied the universal principles
05:55
of polite language,
05:58
which I didn't even actually know existed
05:59
until this research happened.
06:01
And they found something really interesting.
06:03
They had to go beyond just helping people
06:05
ask their friend to take the photo down.
06:07
They had to help people express to their friend
06:09
how the photo made them feel.
06:12
Here's how the experience works today.
06:14
So I find this hypothetical photo of myself,
06:17
and it's not spam, it's not abuse,
06:20
but I really wish it weren't on the site.
06:23
So I report it and I say,
06:25
"I'm in this photo and I don't like it,"
06:28
and then we dig deeper.
06:30
Why don't you like this photo of yourself?
06:34
And I select "It's embarrassing."
06:36
And then I'm encouraged to message my friend,
06:39
but here's the critical difference.
06:42
I'm provided specific suggested language
06:44
that helps me communicate to Laura
06:48
how the photo makes me feel.
06:50
Now the team found that this relatively small change
06:52
had a huge impact.
06:55
Before, only 20 percent of people
06:57
were sending the message,
06:59
and now 60 percent were,
07:00
and surveys showed that people
07:02
on both sides of the conversation
07:04
felt better as a result.
07:06
That same survey showed
07:08
that 90 percent of your friends
07:09
want to know if they've done something to upset you.
07:12
Now I don't know who the other 10 percent are,
07:14
but maybe that's where our "Unfriend" feature
07:16
can come in handy.
07:18
So as you can see,
07:20
these decisions are highly nuanced.
07:21
Of course we use a lot of data
07:24
to inform our decisions,
07:26
but we also rely very heavily on iteration,
07:27
research, testing, intuition, human empathy.
07:31
It's both art and science.
07:35
Now, sometimes the designers
who work on these products
07:37
are called "data-driven,"
07:39
which is a term that totally drives us bonkers.
07:41
The fact is, it would be irresponsible of us
07:44
not to rigorously test our designs
07:47
when so many people are counting on us
07:49
to get it right,
07:51
but data analytics
07:53
will never be a substitute for design intuition.
07:55
Data can help you make a good design great,
07:58
but it will never made a bad design good.
08:01
The next thing that you need
to understand as a principle
08:05
is that when you introduce change,
08:08
you need to do it extraordinarily carefully.
08:09
Now I often have joked that
08:12
I spend almost as much time
08:14
designing the introduction of change
08:16
as I do the change itself,
08:18
and I'm sure that we can all relate to that
08:20
when something that we use a lot changes
08:22
and then we have to adjust.
08:24
The fact is, people can become
08:26
very efficient at using bad design,
08:29
and so even if the change is
good for them in the long run,
08:32
it's still incredibly frustrating when it happens,
08:34
and this is particularly true
08:37
with user-generated content platforms,
08:39
because people can rightfully
claim a sense of ownership.
08:42
It is, after all, their content.
08:45
Now, years ago, when I was working at YouTube,
08:49
we were looking for ways to
08:51
encourage more people to rate videos,
08:53
and it was interesting because
when we looked into the data,
08:56
we found that almost everyone was exclusively using
08:58
the highest five-star rating,
09:02
a handful of people were using
09:04
the lowest one-star,
09:05
and virtually no one
09:07
was using two, three or four stars.
09:09
So we decided to simplify
09:11
into an up-down kind of voting binary model.
09:13
It's going to be much easier
for people to engage with.
09:16
But people were very attached
09:19
to the five-star rating system.
09:22
Video creators really loved their ratings.
09:23
Millions and millions of people
09:26
were accustomed to the old design.
09:27
So in order to help people
09:29
prepare themselves for change
09:31
and acclimate to the new design more quickly,
09:33
we actually published the data graph
09:35
sharing with the community
09:38
the rationale for what we were going to do,
09:39
and it even engaged the larger industry
09:41
in a conversation, which resulted in
09:44
my favorite TechCrunch headline of all time:
09:46
"YouTube Comes to a 5-Star Realization:
09:49
Its Ratings Are Useless."
09:52
Now, it's impossible to completely avoid
09:55
change aversion when you're making changes
09:58
to products that so many people use.
10:00
Even though we tried to do all the right things,
10:02
we still received our customary flood
10:04
of video protests and angry emails
10:06
and even a package that had
to be scanned by security,
10:09
but we have to remember
10:13
people care intensely about this stuff,
10:15
and it's because these products, this work,
10:18
really, really matters to them.
10:21
Now, we know that we have to be careful
10:23
about paying attention to the details,
10:26
we have to be cognizant about how we use data
10:28
in our design process,
10:31
and we have to introduce change
10:33
very, very carefully.
10:34
Now, these things are all really useful.
10:36
They're good best practices for designing at scale.
10:38
But they don't mean anything
10:41
if you don't understand something
10:43
much more fundamental.
10:45
You have to understand who you are designing for.
10:46
Now, when you set a goal to design
10:51
for the entire human race,
10:53
and you start to engage in that goal in earnest,
10:55
at some point you run into the walls
10:58
of the bubble that you're living in.
11:00
Now, in San Francisco, we get a little miffed
11:02
when we hit a dead cell zone
11:05
because we can't use our phones to navigate
11:06
to the new hipster coffee shop.
11:08
But what if you had to drive four hours
11:11
to charge your phone
11:14
because you had no reliable source of electricity?
11:15
What if you had no access to public libraries?
11:18
What if your country had no free press?
11:22
What would these products start to mean to you?
11:25
This is what Google, YouTube and Facebook
11:28
look like to most of the world,
11:31
and it's what they'll look like
11:33
to most of the next five billion people
11:34
to come online.
11:36
Designing for low-end cell phones
11:38
is not glamorous design work,
11:40
but if you want to design for the whole world,
11:42
you have to design for where people are,
11:44
and not where you are.
11:46
So how do we keep this big, big picture in mind?
11:48
We try to travel outside of our bubble to see, hear
11:51
and understand the people we're designing for.
11:54
We use our products in non-English languages
11:57
to make sure that they work just as well.
11:59
And we try to use one of these
phones from time to time
12:01
to keep in touch with their reality.
12:04
So what does it mean to design at a global scale?
12:07
It means difficult and sometimes exasperating work
12:12
to try to improve and evolve products.
12:15
Finding the audacity and the
humility to do right by them
12:19
can be pretty exhausting,
12:22
and the humility part,
12:24
it's a little tough on the design ego.
12:25
Because these products are always changing,
12:27
everything that I've designed in my career
12:30
is pretty much gone,
12:32
and everything that I will design will fade away.
12:34
But here's what remains:
12:37
the never-ending thrill
12:39
of being a part of something that is so big,
12:41
you can hardly get your head around it,
12:43
and the promise that it just might change the world.
12:46
Thank you.
12:49
(Applause)
12:51

▲Back to top

About the speaker:

Margaret Gould Stewart - User experience master
At Facebook (and previously at YouTube), Margaret Gould Stewart designs experiences that touch the lives of a large percentage of the world's population.

Why you should listen

Margaret Gould Stewart has spent her career asking, “How do we design user experiences that change the world in fundamental ways?” It's a powerful question that has led her to manage user experiences for six of the ten most visited websites in the world, including Facebook, where she serves as Director of Product Design.

Before joining Facebook, Margaret managed the User Experience Team for YouTube, where she oversaw the largest redesign in the company's history, including the YouTube player page. She came to YouTube after two years leading Search and Consumer Products UX at Google. She approaches her work with a combined appreciation for timeless great design and transient digital technologies, and always with the end goal of improving people's lives. As she says: "Design is creativity in service of others."

More profile about the speaker
Margaret Gould Stewart | Speaker | TED.com