"What do you think of when I say the word "design"?You probably think of things like this,finely crafted objects that you can hold in your hand,or maybe logos and posters and mapsthat visually explain things,classic icons of timeless design.But I'm not here to talk about that kind of design.I want to talk about the kindthat you probably use every dayand may not give much thought to,designs that change all the timeand that live inside your pocket.I'm talking about the designof digital experiencesand specifically the design of systemsthat are so big that their scalecan be hard to comprehend.Consider the fact that Google processesover one billion search queries every day,that every minute, over 100 hoursof footage are uploaded to YouTube.That's more in a single daythan all three major U.S. networks broadcastin the last five years combined.And Facebook transmitting the photos,messages and storiesof over 1.23 billion people.That's almost half of the Internet population,and a sixth of humanity.
1:22These are some of the productsthat I've helped design over the course of my career,and their scale is so massivethat they've produced unprecedenteddesign challenges.But what is really hardabout designing at scale is this:It's hard in part becauseit requires a combination of two things,audacity and humility —audacity to believe that the thing that you're makingis something that the entire world wants and needs,and humility to understand that as a designer,it's not about you or your portfolio,it's about the people that you're designing for,and how your work just might help themlive better lives.Now, unfortunately, there's no schoolthat offers the course Designing for Humanity 101.I and the other designerswho work on these kinds of productshave had to invent it as we go along,and we are teaching ourselvesthe emerging best practicesof designing at scale,and today I'd like share some of the thingsthat we've learned over the years.
2:26Now, the first thing that you need to knowabout designing at scaleis that the little things really matter.Here's a really good example of howa very tiny design element can make a big impact.The team at Facebook that managesthe Facebook "Like" buttondecided that it needed to be redesigned.The button had kind of gotten out of syncwith the evolution of our brandand it needed to be modernized.Now you might think, well, it's a tiny little button,it probably is a pretty straightforward,easy design assignment, but it wasn't.Turns out, there were all kinds of constraintsfor the design of this button.You had to work within specific height and width parameters.You had to be careful to make it workin a bunch of different languages,and be careful about using fancy gradients or bordersbecause it has to degrade gracefullyin old web browsers.The truth is, designing this tiny little buttonwas a huge pain in the butt.
3:20Now, this is the new version of the button,and the designer who led this project estimatesthat he spent over 280 hoursredesigning this button over the course of months.Now, why would we spend so much timeon something so small?It's because when you're designing at scale,there's no such thing as a small detail.This innocent little buttonis seen on average 22 billion times a dayand on over 7.5 million websites.It's one of the single most viewed design elements ever created.Now that's a lot of pressure for a little buttonand the designer behind it,but with these kinds of products,you need to get even the tiny things right.
4:03Now, the next thing that you need to understandis how to design with data.Now, when you're working on products like this,you have incredible amounts of informationabout how people are using your productthat you can then use to influenceyour design decisions,but it's not just as simple as following the numbers.Let me give you an exampleso that you can understand what I mean.Facebook has had a tool for a long timethat allowed people to report photosthat may be in violation of our community standards,things like spam and abuse.And there were a ton of photos reported,but as it turns out,only a small percentage were actuallyin violation of those community standards.Most of them were just your typical party photo.Now, to give you a specific hypothetical example,let's say my friend Laura hypotheticallyuploads a picture of mefrom a drunken night of karaoke.This is purely hypothetical, I can assure you.(Laughter)Now, incidentally,you know how some people are kind of worriedthat their boss or employeeis going to discover embarrassing photos of themon Facebook?Do you know how hard that is to avoidwhen you actually work at Facebook?So anyway, there are lots of these photosbeing erroneously reported as spam and abuse,and one of the engineers on the team had a hunch.He really thought there was something else going onand he was right,because when he looked through a bunch of the cases,he found that most of themwere from people who were requestingthe takedown of a photo of themselves.Now this was a scenario that the teamnever even took into account before.So they added a new featurethat allowed people to message their friendto ask them to take the photo down.But it didn't work.Only 20 percent of peoplesent the message to their friend.So the team went back at it.They consulted with experts in conflict resolution.They even studied the universal principles of polite language,which I didn't even actually know existeduntil this research happened.And they found something really interesting.They had to go beyond just helping peopleask their friend to take the photo down.They had to help people express to their friendhow the photo made them feel.
6:13Here's how the experience works today.So I find this hypothetical photo of myself,and it's not spam, it's not abuse,but I really wish it weren't on the site.So I report it and I say,"I'm in this photo and I don't like it,"and then we dig deeper.Why don't you like this photo of yourself?And I select "It's embarrassing."And then I'm encouraged to message my friend,but here's the critical difference.I'm provided specific suggested languagethat helps me communicate to Laurahow the photo makes me feel.Now the team found that this relatively small changehad a huge impact.Before, only 20 percent of peoplewere sending the message,and now 60 percent were,and surveys showed that peopleon both sides of the conversationfelt better as a result.That same survey showedthat 90 percent of your friendswant to know if they've done something to upset you.Now I don't know who the other 10 percent are,but maybe that's where our "Unfriend" featurecan come in handy.
7:19So as you can see,these decisions are highly nuanced.Of course we use a lot of datato inform our decisions,but we also rely very heavily on iteration,research, testing, intuition, human empathy.It's both art and science.Now, sometimes the designers who work on these productsare called "data-driven,"which is a term that totally drives us bonkers.The fact is, it would be irresponsible of usnot to rigorously test our designswhen so many people are counting on usto get it right,but data analytics will never be a substitute for design intuition.Data can help you make a good design great,but it will never made a bad design good.
8:04The next thing that you need to understand as a principleis that when you introduce change,you need to do it extraordinarily carefully.Now I often have joked thatI spend almost as much timedesigning the introduction of changeas I do the change itself,and I'm sure that we can all relate to thatwhen something that we use a lot changesand then we have to adjust.The fact is, people can becomevery efficient at using bad design,and so even if the change is good for them in the long run,it's still incredibly frustrating when it happens,and this is particularly truewith user-generated content platforms,because people can rightfully claim a sense of ownership.It is, after all, their content.
8:48Now, years ago, when I was working at YouTube,we were looking for ways toencourage more people to rate videos,and it was interesting because when we looked into the data,we found that almost everyone was exclusively usingthe highest five-star rating,a handful of people were usingthe lowest one-star,and virtually no onewas using two, three or four stars.So we decided to simplifyinto an up-down kind of voting binary model.It's going to be much easier for people to engage with.But people were very attachedto the five-star rating system.Video creators really loved their ratings.Millions and millions of peoplewere accustomed to the old design.So in order to help peopleprepare themselves for changeand acclimate to the new design more quickly,we actually published the data graphsharing with the communitythe rationale for what we were going to do,and it even engaged the larger industryin a conversation, which resulted inmy favorite TechCrunch headline of all time:"YouTube Comes to a 5-Star Realization:Its Ratings Are Useless."
9:54Now, it's impossible to completely avoidchange aversion when you're making changesto products that so many people use.Even though we tried to do all the right things,we still received our customary floodof video protests and angry emailsand even a package that had to be scanned by security,but we have to rememberpeople care intensely about this stuff,and it's because these products, this work,really, really matters to them.
10:22Now, we know that we have to be carefulabout paying attention to the details,we have to be cognizant about how we use datain our design process,and we have to introduce changevery, very carefully.Now, these things are all really useful.They're good best practices for designing at scale.But they don't mean anythingif you don't understand somethingmuch more fundamental.You have to understand who you are designing for.
10:50Now, when you set a goal to designfor the entire human race,and you start to engage in that goal in earnest,at some point you run into the wallsof the bubble that you're living in.Now, in San Francisco, we get a little miffedwhen we hit a dead cell zonebecause we can't use our phones to navigateto the new hipster coffee shop.But what if you had to drive four hoursto charge your phonebecause you had no reliable source of electricity?What if you had no access to public libraries?What if your country had no free press?What would these products start to mean to you?This is what Google, YouTube and Facebooklook like to most of the world,and it's what they'll look liketo most of the next five billion peopleto come online.Designing for low-end cell phonesis not glamorous design work,but if you want to design for the whole world,you have to design for where people are,and not where you are.
11:47So how do we keep this big, big picture in mind?We try to travel outside of our bubble to see, hearand understand the people we're designing for.We use our products in non-English languagesto make sure that they work just as well.And we try to use one of these phones from time to timeto keep in touch with their reality.
12:06So what does it mean to design at a global scale?It means difficult and sometimes exasperating work to try to improve and evolve products.Finding the audacity and the humility to do right by themcan be pretty exhausting,and the humility part,it's a little tough on the design ego.Because these products are always changing,everything that I've designed in my careeris pretty much gone,and everything that I will design will fade away.But here's what remains:the never-ending thrillof being a part of something that is so big,you can hardly get your head around it,and the promise that it just might change the world.
12:50(Applause) 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.
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.