Hello there! I’m at SXSW (South By Southwest) 2012 this weekend, going to the interactive conference. I’m also on a panel called Binary Bitches: Keeping Open Source Open to Women with RIT professor Andrea Hickerson and RIT design thinking lecturer Xanthe Matychak. If you are at SXSW and interested in free / open source software, you should drop by and say hi. 🙂 It’s tomorrow at 5.
Anyway, I tend to take copious notes when I go to talks. I thought they might be useful to a wider audience, so here you go.
Design from the Gut: Dangerous or Differentiator?
Session Description:
(From the SXSW 2012 schedule)
The internet is a never-ending data source. Through it we are able to monitor visitor activity, study traffic patterns, and use these analytics to help guide users in the directions we want. Usability testing gives us behavioral information which can either affirm design decisions or inform necessary changes. Research and analytics go a long way in selling a creative direction to clients who are focused on engaging with their customers and in how marketing dollars will impact their bottom line.
But what about a designer’s instinct—that moment when a designer just knows what they’re building is right? When and how do their years of professional experience, inspirational collections, and life observations become deciding factors?
Learn from a panel of design veterans, with experience that ranges from client services to product development, about past experiences and their personal stance on the subject.
Panelists:
- Jane Leibrock User Experience Researcher, Facebook
- Laurel Hechanova Designer & Illustrator, Apocalypse OK
- Naz Hamid Founder of Weightshift, design studio in San Francisco
- Phil Coffman Creative Director at Element (in Austin)
- Bill Couch Software Engineer, Twitter (formerly of USA Today)
Session Twitter Feeds:
(Attendees could ask questions via Twitter)
- @dftg on Twitter
- #SXfromthegut
- #fromthegut
Session website:
The panelists set up a website:
http://designfromthegut.com/
Notes:
What is designing from the gut?
Naz >
Watched Ed Speakerman talk at TED. Sometimes he design typefaces – sometimes he’ll design the whole thing out, no research… then backtrack, tell the story of why he did it that way, after the fact. At the time I’ve done that in projects before – you just design something based on what you’ve done or what you feel for the client / project at ahnd. There’s other factors that go into why sometimes that tends to happen – quick turnaround or small budget or budgets that don’t allow for longer user research period – but I think that’s pretty common. You just know it. You feel it in your bones, that you designed something great. That’s what I think designing for the gut it.
Laurel >
Gut is a weird concept if you’re a professional, because you keep up with best practices, and your peers in the industry. Gut is smarter than a lot of people’s research, so designing from Eric Speakerman’s gut is more informed than I think maybe… “Gut” for him has a different connotation.
Naz >
Gut is really experience. It’s informed by everything you’ve done.
(Phil) So if you’re just getting started, you’re a beginner, what do you do? Things you’ve seen and been influenced by, that steered you towards being a designer – what do you do?
Laurel >
It’s like learning an instrument. You start listening to the bands you like, listen to their songs… Initially maybe you mimic your hero’s designs a little to closely; over time you develop your own style. Listening to your gut while you’re still learning is probably a good way to balance it out.
(Phil) Analytical data / usability research vs the gut. Does one ever trump the other? Does this change during the phases of the project? Is there a logical time one makes more sense than another?
Bill >
Definitely times where one is better than the other. More and more devices are coming out now, technology changing so fast. He was at USA Today, was fortunate to work on the iphone/ipad apps. It did well, they wanted to be involved in ipad after iphone – contacted apple and got in touch with them about creating an app for ipad. But we had never seen or touched or played with one – designing for a device you’ve never worked with before is difficult. We had the devices, but they were chained into a windowless section of the office, could not take it into any context the devices would be used in. We had to make paper prototypes, discuss theoretical scenarios, had to explore knowing the features that existed. We spent a lot of time thinking about and iterating on actual prototypes for the device. Would you simulate a page turn or some other interaction to allow for navigation? It was a really crazy moment of experimentation but it was all based on what we thought made sense. We had no idea what anybody else was going to do…. I believe everyone was in a similar position that we were. We could see how it all played out after the device came out. Definitely want to focus on gut vs research depending on how much you know about the tech.
Jane >
People think research comes in later on. I try to get UX research involved early on. Early formative research in coming up with ideas for products, before designing or coding. I tend to do a lot up front… first or second iteration is when people are itching to do usability test… right of the top of my head, my gut, what i know about ux… usability best practices – we could probably shop something around internally and make it better and clean it up before bringing users into the lab. Iterate before you bring users in and do formal research.
(Naz) There’s different types of research, low level, just checking in on things with a small group. Even over skype – it’s like a gut check – what do you think of this? How does that scale? How do you make that scale?
Naz>
When you take… say a photoshop comp… hey Bill, check this out. Phil, check this out. You’re doing a redesign and you want to get feedback from your best friends / colleagues – is that enough? Does that scale? Does it have to go on bigger than that? What is the sample size that is useful?
Phil >
I do that all the time. I don’t give context. I show them a jpg – what do you think? I built this core – your trusted friends – cull it down to people who you know are going to be on the same level you’re at, same kind of journey in professional career… or they provide really valuable feedback. I’ve never really felt the need to go beyond my core circle. I’m not working for a massive product company. I work for a small shop with 4 people. I’ve never sat in on a usability test in a room where I’ve showed my work and had someone do it. I see the value in it but my projects never had the budget or time to fit that in. Relying on my core knit group of folks up to this point has always been okay.
(Naz) How do you do this [user research] without budget?
Jane >
Refining design consulting other designs… getting near the end of the process, you need to bring in users. Make sure it makes sense for their context / life experience. If you do one round of usability with 5 or 6 people, you’ll catch the major problems right them. It’s time to fix those problems, then do a second round.
Laurel >
There’s a danger in asking your friends, if they are in your industry. They’ll catch kerning issues, surface-level design flaws. You should ask your grand-dad, the guy washing your car… get new perspectives [for the bigger picture].
Bill >
As a gut check we’ll do a handful of tests at Twitter to see what makes sense, what doesn’t. We’ll release something to employees only to get a sense for whether or not it works, how it feels. We extend out beyond that… as many people as you need to discover issues.
Jane >
You could test just 2-3 people, and if everyone got that wrong, or if nobody saw that button, you’ll know.
Laurel >
When the experience is the entire product, testing and research more important. If it’s a means to an end, ecommerce or selling t-shirts… the t-shirt is the product. The website is part of the experience but testing that experience is maybe less important than if you’re trying to build a to-do app, or the app interaction is going to be more difficult than interacting at storefront.
(Naz) If you don’t design from the gut at all, is it still possible to create something beautiful?
Jane >
I don’t think that’s possible
Laurel >
You’re not a machine. Your decisions… everything is tainted… you are a filter for whatever you decide.
Phil >
I could have all this data, but I really don’t have the time to cull through it. You have to make decisions on the fly. You are relying on your gut when you do that. The creative process is not linear, it ebbs and flows, so you have to hit a rhythm and start relying on your experience… maybe you have to pull in the thing you know worked in the last project. Ut can be a time issue… something flares up and you’ve got to put out a fire. You come back – sitting down and going through the data, it’s very valuable… time in my day really impacts how I actually end up doing what I do.
(Laurel) Is it possible to be irresponsible through overanalyzing? Blow the budget, etc.?
Phil >
Yes, data can keep coming. You could keep going through the data… could continue to analyze and talk away through it.
Jane >
it’s also possible to keep iterating based on data limitlessly. at facebook we are very analytical and data driven. we make a small change, look at metrics. if you are constantly doing little tweaks only, you’re not doing a full design…
Bill >
Even small percentages, when you work at a large scale, small changes results in a lot of people being affected. When we try to increase the numbers of users in Twitter, for example, we change something in the sign up flow that improves it 2%: that is thousands or tens of thousands of users.
If you ignore accessibility and screen readers… a small percentage of the total, but still a large number of people.
Scale comes into this.
Laurel >
While doing research for this – I think where to draw the line for research is kind of a gut decision.
(Naz) With endless data is there a point at which you have to pick a data point or to find a new data set?
Phil >
I have a lot of inner turmoil. Endless loop even if I’m designing from the gut. I can be fickle. Just pull out of it, look at the data, look at what the client’s trying to accomplish, look at some other products or companies that did something similar – look at the numbers, look at the data… use that as a factor. It’s such a personal and an emotional thing… unless you’re more stable than I am…
Another question – how much can… if you haven’t done the cold hard studies… if you’re looking at other sites – a client that says ‘it’s just like pinterest!’ if you use what they have done as your research – is that a good or bad thing?
(Phil) Is it a good thing or bad thing to study other sites if you don’t have research?
Jane >
That can work well if you know their users are similar to your users. Pinterest has a really neat interface, if you only want to copy it because its cool, and you don’t consider that it doesn’t work for your users, that’s a problem.
Naz >
Who in this room has heard ‘apple does it’ ‘facebook does it’ ‘twitter does it’ in a meeting?
It’s a form from taking someone else’s user research… looking at what their users do, but the audiences may not be the same.
Bill >
I wonder if you ride on what someone else has done – was it their intuition, or research? You can sometimes come into an echo chamber … oh, this place did it, we’ll do it. I’ve noticed – any interaction facebook puts in the iphone app… most installed app on the iphone… their menu to the left that slides other – now other apps have copied that. several iterations ago – used to be a sliderbar at the time, between news feed, sections, list… we were loking how to make our sections of paper scale, so we used that interaction since we thought it was a similar userbase.. thousands of people who would use both products. If we use that, people will be familiar with it because they used it before.
You know people are familiar because they’ve done it in other apps / websites so you use the same interaction pattern. Scrolling, parallax, is a very common trend. Is it something people know how to interact with from the get go…? Just because someone big did it, other people do it. Maybe not because research shows it makes sense.
(Bill) If you’re using other people’s research – how do you know if it’s based on research or intuition?
Naz >
I think you’ll see their success. You’ll see their results.
Bill >
But a lot of theses sites or apps iterate on these things. Whether or not it worked for them, they may be looking at alternatives to improve it or stay fresh. Did it work for them or not if they drop it?
(Naz) Gut design leads to innovation. You don’t want to do what everyone else is doing. How do you sell that to a team or client when other patterns are successful?
Phil >
It’s hard.
(Naz) There’s a point at which designers are going to be, ‘Well, I think it should just be this way.’ Fight it? Pull back?
Phil >
It’s different depending if you’re doing it internally vs selling to a client.
Internally… in some respects, if you’re doing it internally with your team, there’s a level of respect and you’ve worked together. If I’ve pitched an idea i think is new or fresh, I have a pre-existing relationship, seems like a good idea, we have a good history. You have to provide… you can’t just go in there and say ‘we need to do it this way.’
You still have to provide some basis for your decision. I think – even though you think it’s just from the gut – maybe you were looking at another site, got this idea, followed it this way.
I think it’s easier to sell internally. you could say ‘let’s just try it’ maybe do a quick prototype, see how it behaves.
With a client, it’s completely different. I used to run into the problem of going into a talk with a client, walk them top to bottom in a comp, explain all the cool design things I did in the comp – client doesn’t care. they want to know how it will effect their bottom line and what they are trying to accomplish. Without cold hard facts and data… you have to phrase it and word it in a way they will sponsor it. you gotta sell it… a lot of it at that point – a lot of what I’ve got here is what U know feels right. You can’t just go in there and say that. You have to know how to read the person…
(Naz) Is there a matter of… trust? What if the trust isn’t there, or what if they trust you and it doesn’t work, how do you recover? How do you build the trust in the first place?
Laurel >
Hopefully before you signed on to work with the client, you established that you were on the same page, you understand their objectives, you agree with their objectives, so that passion is there…
(Naz) What if it changes?
Laurel >
The design approach – give it to me, I want it – this is not the right way to approach it. You’re going to have to come up with reasons that would sell someone on it. If you lost your client’s trust – this is more like PR.. you need to talk to them. Figure out why you lost the trust in the first place. You may have undermined your foundation for doing design work totally.
Bill >
Internally, on trust, at twitter… a good question. I think… it’s certainly something as designs and concepts are developed and iterated on by our design team. people see them, they work with them, they find they work or don’t work. It’s how you develop that process. Do they take feedback? Do they work well with you? You establish a relationship that way. the relationship that is established becomes strong… you need the experience with them to build trust. A lot of times too, work will speak for itself. Also, do new iterations for new products… redesigned the Twitter site last December – you have to understand and develop that trust. iIhave a lot of trust in the design team and what they are going to do. You just establish a rapport there, I guess.
(Naz) When the data feels incorrect, or the gut feels incorrect, or the two don’t coincide. One or the other doesn’t match up. What do you do?
Jane >
What do you mean the data isn’t correct?
Naz >
Your perception of the data – small sample size, not enough data – making assumption about certain things, whether it’s the gut or data. you have two things at odds with each other.
Jane >
I think when data comes in, in some form, and it seems at odds with your gut, I don’t think the best reaction is to make a change based on the data. I think it’s a moment where you are missing some information. If you don’t know what that data means, go back and try to do more research. It’s a “reassess” moment, not make a change moment.
If you had a product prototyped, showed it to some users, and what came back was people that you thought would really like it didn’t like it: I think it is a signal that you don’t understand your users very well, so you should step back and go learn more about them before proceeding with the design.
Bill >
Relationship between designing from the gut, validating hypothesis, and actually shipping something.
For the sake of iteration, we’ll go with this, release it, see what happens. The web isn’t permanent. We’ll settle the disagreement that way. If you are in a position in a place where there is a lot of iteration, you can use that as a way to validate it.
We released the redesign last Dec of the site. A lot of intuition and gut for how to conceptualize the site for a really wide audience. Some of that works well for some people, for others it doesn’t. We used what came back from that to iterate. Make some of those bigger leaps and trust that things will work out for there to be any progress.
So that is where I think the line should be drawn, gut or research.
Jane >
Also the matter… you can test as much as you want with a large number of users, but you really don’t know what happens when it scales, so you just try it.
Bill >
We’ll use gut and imtuition for big conceptual stuff. We’ll use data for where to put timestamps, color, surface-level design things. Those are things where analytics come into play – that is fine-tuning… sometimes a composition looks good, makes sense… over sustained periods of time, larger companies like Facebook and Twitter, people who have used and passionate about the project, feel passionate against the change. Over the time of use… they’ll like it over time. If it doesn’t happen that way, you’ll seem long term trends, and you know something is wrong.
Phil >
Jane & Bill work for Facebook and Twitter. You are working on a product that you work on every single day, day in, day out. you have that laser focus…
Myself, and Laurel, we work on many different projects, can change widely depending on clients.
(Phil) Laurel, do you find that you have to rely on gut more because you don’t know what is coming next? Or are there enough patterns that there is always going to be something familiar, no matter what the product?
Laurel >
I never reinvent the wheel… all web apps. I’m not in entirely unfamiliar territory with each product. In that sense… no gut thing to the question. Nice thing about having different clients all the time is that you get to step outside of the role that they are immersed in. For example, I have two clients who are architects. I love architecture, but I’m not steeped in it, so the things they assume everybody knows, I get to tell them no – not everyone knows what that is. Not being familiar can mean you can give good feedback.
Naz >
You know, if you’ve got the sort of relationship where hopefully, let’s do this, this way, make sure whatever we are going to do is going to stand up to your measure of success for the project. There is a point at which you have to say, let’s dig deep, can we do the research? What do you know about your business? They may have analytics, they may have nothing. There are plenty of start-up clients who don’t even know how to gauge their clientele because it doesn’t exist yet or they are just starting out.
How do people use this thing? Client might not know who their users are. They might have an idea but… in the valley, there can be young startups that have a perception of who their users are, in the sense that they are 25-35, educated, income level… but that’s not a user. Do you really know who your user is?
Can you pull out a prototype before you build the actual thing? Parts of research – what kind of research do you do?
Phil >
Then the client’s gut comes into play, they may feel extremely strong about something. They are a start=up, other sites they’ve seen..
Naz >
When a startup or any new company comes to you and wants to hire you… you got to do the best job you can do, so it’s hard to ask the hard questions; they don’t want to hear it. We had clients we had to let go, irreconcilable differences. but you got to take a stand someplace. If it’s a gut thing, or they are just wrong. Well… I can’t bring this to market, I can’t help you do this. But if you have a good client, willing to learn it, open, hired you for the right reasons…
(Phil) What are the most surprising or counter-intuitive designs you’ve seen?
Phil >
Pushing the envelope… using your gut to spark innovation. Henry ford, ‘if i gave everyone what they wanted it would be a faster horse.’
Do you know of some examples that come to mind? Maybe it’s not just design from the gut – the whole idea of sparking innovation. What helps drive everything forward?
Naz >
One of the apps that came out recently was Path. Path has interesting UI bits… in the little corner, it springs out – not anything we’d seen. Once that launched, you ask – is that good? Is it useful? The labels are in a weird spot – is that good?
Another one – clear, the to-do app. Pulls down. Pull-down is notification center, that’s weird.
bill >
Path – everything is moments… one flowing interface… I think a lot of about the sort of design for long time archiving… Twitter is very much a product of short time I guess… flickr.. if you think about… instagram vs flickr, instagram for right now, Flickr is for long-term. Flickr is trying to really position itself for immediacy again. Scaling interfaces over time. These novel approaches that feel impressive when they first come out (Path, clear), has a lot of wow factor. But how do they age? Path is iterating – does it make sense, scannability reduced if timestamp moves if you’re looking at the person you are scrolling? Will those things work well over time? Maybe it does work well? But a lot of considerations to be made around these things over time.
Laurel >
Pull-down-to-refresh is my favorite example. I’m disappointed when an app doesn’t do that.
It was lauren richter (?) who designed it.
Bill >
I remember when that came out, it felt natural and intuitive, but also – is this a gimmick? will it age well?
it’s a hand repeating gesture, twitchy mode…
Laurel >
I think i accidentally discovered it. if people can happen upon it, then you are aligning with human instincts.
Bill >
have to design for human fallacy instead of machine
Laurel >
not everyone will pick up on vernacular / patterns.
(Phil) When does being right trump consistency?
Bill >
human interface guidelines for iphone… interfaces will continue to evolve. sometimes comes from fatigue from designers… this feels so tried and true, not exciting anymore. how do we make this exciting again? make it fresh?
exciting to see other people doing fun, crazy, creative things. not always best approach
Phil >
boring. i want to challenge myself. i don’t want to just go and copy patterns. It behooves you to do some research to riff off of. Challenge yourself. Ask yourself, ‘What’s another way I can do this?’
I’m currently working with a client who loves pinterest… it’s crazy. pinterest, everyone thinks they’ve created a classical rel-obj design… i like to challenge myself, it’s good to look at those things. when i start a project, i like to do that kind of self research, look at other sites doing similar things, similar audience – see how they solved those problems.
sometimes my approach won’t work, and pinterest is the best way to do it.
Question from the Audience on the design process
A good informed design process might be broken into stages. Research, designer work with research, then going back to users and analyzing. Can you give us insight into what that first stage looks like, what sources you go to, what does that research entail? I work for a digital/print publiciation. We have real time users, 100’s per second, we want to analyze that.
Jane >
Early stage of research is usually… talking about a new product. before you start thinking about a specific design too much or a specific technical approach, take time to learn about the users themselves, the users of this product. part comes from your gut – who is going to be the userbase for thse products.
go out and find those people. if you are interested in a certain area, look for people who are extreme in that area. interested in a messaging app? find people who engage in lots and lots of communication, find something from their extreme behavior and try to adapt from their behavior.
or try to find a specialist.
or look for a represenative group or two, embed yourself with them, following them in an ethnographic way, one-on-one interviews… the research will be very open-ended, here’s a product, how do you react to it? no. you want to understand their workflows, lifestyle, etc, and see how the product could fit.
(Laurel) From Megan on Twitter, what if your client wants to respect their gut more than your gut?
Laurel >
i think this goes back to making sure you’re on the same page all the way through. no way to reconcile that otherwise. ultimately, i’ll be totally honest, its their decision, they’re the client. their gut wins by default because they are paying money. If you can reason with them and convince them why their gut is wrong, maybe you can solve the problem.
(Laurel) What if you’re working internally?
Jane >
have to understand bosses gut
Bill >
engineers as one sort of client you’re working with, because they will ultimately be implementing. additionally have product managers who define the product and its vision, then you have the product team…
relationship isn’t always perfectly clean. trade-offs. need to be okay, settle things in various ways. over time, as things change…
difficult at a large scale, a lot of people who care about the product, invested in it… it can be a challenge but generally works out well
Naz >
Out of time, thanks everybody.
Great notes. Thanks for sharing. This panel would have been much better if placed in a historical context. Do you think the mid century modernists that we love so much did market research? No way. I’m not saying that’s a good or bad thing, but there is a history here, a shift in ideologies that’s important to a conversation like this. Without it, the discussion was floating in space.