Its buttons Facebook "I like" and "Sharing" we see 22 billions of times a day, making them the most frequently appearing design elements ever created.
Η Margaret Gould Stewart, product design director Facebook, summarizes three rules of design on a grand scale - so large that minimal modification can cause global outrage, but also so large that the most subtle improvement can positively affect many people's lives. See the presentation he made at TED Talks
Η translation made by Luca Kaimara, and edited by Chryssa Rapessi.
What do you think when I say the word "design"? You might think things like this, masterpieces that you can keep in your hands, or maybe logos, posters and maps which explain things visually, classic diagrams of timeless design. But I didn't come to talk to you about that kind of planning, but for the species which you probably use everyday and you do not think about it, patterns that are constantly changing and with whom you have very frequent contact. I mean design of digital experiences and especially system design which are so great so that it is difficult to understand their size. Think about Google processing over a billion search requests every day, that every minute, over 100 hours of filming uploaded to YouTube. It is more volume in one day as the three major US networks reported over the past five years all together. And Facebook is transmitting photos, messages and stories from over 1,23 to two people. This is almost half the Internet population and one-sixth of humanity.
1:22 They are some of the products which I helped to design in the course of my career, and their effect is so massive which were design challenges unprecedented. But what is really difficult when planning on such a scale is this: It is partly difficult as it requires a combination of two things, courage and humility - bold, to believe what you are doing is something that the whole world wants and needs, and humility to understand that as a designer, it does not concern you and your portfolio, it's about the people you're designing for, and how your work may help them to make their lives better. Unfortunately, there is no school who teaches the Design Principles for Humanity lesson. Me and the other designers we work on such products we had to invent it along the way, and to be self-taught resulting best practices for large-scale design, and today I would like to share with you what we have learned over these years.
2:26 The first to learn for designing on a large scale is that small things count. Be a really good example how a minimal design element can have a great effect. The Facebook team which manages it button "I like" decided that it should be redesigned. The button started to go wrong with the evolution of our brand and needed to be modernized. You might think it's just a button, so it's probably an easy job for the designer, but it was not. It turned out that there were many limitations to design this button. You had to work with specific parameters of height and width. Be careful to be effective in a bunch of different languages, and use careful gradients and limits because they have to be "erased" beautifully in the old browsers. The truth is, by designing this little button was a huge discomfort.
3:20 This is the new version of the button and the designer who co-ordinated this work calculates that he spent over 280 hours designing the button over and over again for several months. Now, why spend so much time for something so small? Because when you design for a wide scale no detail is insignificant. This innocent little button looks average over 22 billions of times a day and more than 7,5 million sites. It is one of the design elements with the most views. It is very big pressure for a small button and the designer behind it, but with such products, even the details have to be done correctly.
4:03 The next you need to understand is how to design with data. When you work on such products, you have incredible amounts of information about the use of your product that you can use them for your design choices, but you can not just follow the numbers. Let me give you an example so you can understand what I mean. Facebook has long been a tool which allows people to report photos which may violate community standards, such as unwanted mail and abuse. References were made for many photos, but as it turns out, only a small percentage actually violated community standards. Most were ordinary party photos. To give you a specific hypothetical example, let's say my girlfriend, Lora, supposedly, uploads a photo of me from a drunken karaoke night. Definitely hypothetical, I assure you. (Laughs) By the way, you know how some are worried that the boss or their employer will see their unpalatable photos on Facebook; You know how difficult it is to avoid it when you literally work on Facebook? Anyway, there are many such photos which are mistakenly referred to as spasm and insults, and a team engineer had a premonition. He believed something else was happening and he was right, because he has searched many cases discovered that most were from people who were asking to withdraw their own photo. This was a scenario which the group had not taken into account. So they added a new feature which allowed people to ask their friend to remove their photo. But that did not work. Only 20% of people they sent a message to their friend. So the team came back to it. Experts were consulted in conflict resolution. They even studied uniform principles good language, which I did not even know they existed until this investigation was done. And they found something very interesting. They had to go further by facilitating downlink communication. They had to make it easier for the user to express his friend how it made them feel their photo.
6:13 Here's how the experience works today. So I find my hypothetical photo and it's not a spam, it's not abuse, but I really do not want it published. So I report and I say, "I'm in this picture and I don't like it" and then we look deeper. Why do not you like your photo? And I choose "It puts me in a mess." Then encourage me to send a message to my friend, but here is the main difference. I'm given specific suggested expressions which help me transfer to Lora how it makes me feel the photo. The team found that this relatively small change had a tremendous impact. Before, only 20% of people sent the message, and now they did the 60%, and statistics showed that the communicators on both sides they eventually felt better. The same research has shown that 90% of your friends they want to know if they did something that upset you. I do not know the view of the rest of 10% but the "Delete Friend" button can serve you there.
7:19 As you can see these decisions are extremely subtle. Of course we use a lot of data to inform our decisions, but we also rely heavily on it repetition, experimentation, intuition, human consciousness. It is at the same time art and science. Sometimes the designers working on these products are called "data pawns" which is a term that annoys us. The fact is that it would be irresponsible on our behalf not to scrutinize our plans when so many people wait to do it properly, but the analysis of the data it will never replace design intuition. The data may help a good plan to become great, but they will never do a bad plan.
8:04 The next you have to understand as the beginning is that when you make a change, it has to be done with great care. I often say joking that I spend so much time designing the introduction of change, as long as I dedicate myself to the change itself, and I think we have all felt it, when something we use a lot changes and then we have to adjust. The fact is that people handle poorly even poorly designed products, and although their change is in the long run, they are incredibly unhappy when it happens, and that is especially true with data platforms produced by users, because people rightly demand a sense of ownership. It is also their own material.
8:48 Years ago, when I was working on YouTube, we were looking for ways Encourage people to rate videos, and it was interesting because when we saw the data we found out that almost everyone was using it exclusively the five-star rating, few people used the minimum of a star, and almost nobody chose two, three or four stars. So we decided to simplify it in a binary voting pattern you choose above or below. It will be much easier to handle. But people were tied with the five-star system. Producers of the videos worshiped the scoring. Millions of people were used to the old design. So to help people to prepare for change and acclimate faster to the new design, we published the data chart to share with the community the reasoning of the project and also introduced other businesses, and my favorite headline arose of the TechCrunch Tech News website: "YouTube concludes for the 5 star system: His rating is useless. "
9:54 It is impossible to completely avoid the aversion to change when you make changes in products that are used so much. We tried to do the right thing but we again accepted the usual storm from protest videos and outraged emails even a parcel that security had to check, but we have to remember that people care about these things, because these products, this job, it really means a lot to them.
10:22 Now we know we need to pay close attention in examining the details, to be aware of how we use the data in the design process, and we must promote change too carefully. These things are really useful. They are very good design practice on a large scale. But they mean nothing if we do not understand something much more basic. You need to know who you are planning.
10:50 So when you have a goal to draw for all mankind, and you begin to work seriously towards this goal, sometime you hit the walls of the bubble you live in. In San Francisco we are a bit disturbed when it does not catch the cell phone and the navigator does not work to find the way for the new trendy café. But what if you had to drive four hours για να φορτίσετε το phone your why do not you have a reliable source of energy? If you did not have access to public libraries? What if your country did not have freedom of press? What then would these products mean to you? Let 's look like Google, YouTube and Facebook in the world, and so they will look to the next five billion people which will be connected to the internet. Design for cheap mobiles not impressive design work, but if you want to plan for the whole world, you have to draw with the view of the world, not with your own view.
11:47 So how do we keep the general picture in our minds? We're trying to get out of the bubble to be able to understand those for whom we are planning. We use our products in non-English languages to make sure they are working properly. And we try to use such a phone at times to be in touch with their own reality.
12:06 So what does it mean to plan for use on a global scale? It means difficult and sometimes debilitating work try to improve and evolve products. Finding courage and humility to do the right thing, it is usually exhausting, and the part of humility, it's a bit heavy for design egotism. As these products are constantly changing, everything I designed in my career have more or less disappeared, and everything I will draw at some point will go out. But what remains: the endless enthusiasm to belong to something so big, that you can hardly see it altogether, and the promise that it might even change the world.
12:48 Thanks.
