It’s kind of the same question as the video from Elisabeth Gilbert… it’s all about creative process isn’t it…just a little more math…and for you…simplicity. But the “art” of it still is the goal right…
http://www.ted.com/talks/view/id/453

Her take is just one way to look at it. Good design has to do with “showing up” everyday…right?

Joshua, you write:

most of the time designers simply don’t know if what they’re building is great, and they end up relying on process to get as far as they can.

I believe designers follow a process to quickly get them through the 80% that’s not so special, allowing them more time for the 20% that’s really special, and specific to the project (not necessarily in that order!).

Whether they know if (20% of) their design is great is, and whether a process helps them realize that, is another question

It would seem that UIE has been systematically attempting to determine what processes distinguish successful design teams from unsuccessful one. See http://www.uie.com/articles/the3qs/, with more detail at http://www.uie.com/brainsparks/2008/04/23/ia-summit-keynote-journey-to-the-center-of-design/). Apparently Apple is among those they’ve been able to get data from. UIE’s findings suggest there is no Process for great designs. Indeed, dogmatically following a Process (even one labeled “user-centered design”) seems to be associated with failure. Rather, successful teams instead mix and match design tools and techniques to solve each design problem.

That said, UIE reports there are three characteristics (if not processes) of successful design teams: 1. All team members understand the basic vision of the end product, 2. Team members regularly observe users using the product or surrogates of the product, 3. The team celebrates failures, using them as learning experiences. So this suggests there is a process for successful designs, kinda, sorta, involving establishing a vision, employing user research, and conducting iterative design. However, there’s also a lot of latitude for exactly how you do what and when. That’s the art part.

Getting insight –getting the actual design idea –is also the art part. I think that although we experience the moment of insight as something internal to ourselves, it, like much of human behavior is actually driven by our environment. It would seem that insight comes from studying the world around us, both designed and natural, for inspiration. Observing users and celebrating design failures are among the ways to study the world, but I think inspiration can come from studying in other ways too. More tools to the toolchest.

The other feature of insight is that most insights are wrong, so you need a way (process? tools?) to weed out bad insights before the design gets too far. That’s one place where the iterative design part comes in.

agreed on all points

@dl – thanks for the link!

@peter – that’s an interesting take…I would buy that approach much more than I buy the “you must do it this way” approach. Part of this post is a frustration of hearing that so often…

Josh, you just knew I’d have to quibble with your choices of questionable products…

However, Apple has also produced suspect things like the Apple TV, the Cube, iMovie ‘08, and iCal.

The question is how to define “failure of design” and what parts of the product are and are not “design.”

The Cube is one of the finest computers I’ve ever used. It was silent and beautiful. It’s part of the permanent collection at MoMA. Failure of design? Because they priced it about 15% higher that it’s performance warranted? By that measure, how does the 20th Anniversary Mac stack up?

I continue to hear people call the Apple TV a failure. How is that? It sports a beautiful and usable interface that works from across the room with a tiny remote with three buttons. Apple has sold millions of them. Is that a failure because they’ve sold many more of other products? By that measure, Audi’s sportscars are failures when compared to the A4, because Audi sells many more of the A4. However, no car enthusiast would call them failures. It just appeals to a smaller audience.

iMovie 08, I’ll agree with that… they replaced something capable and relatively easy to use with something less capable and less easy to use. iCal certainly has its problems… but nothing that couldn’t be solved in one revision if they assigned one talented guy who really cared about the app…

The point: when do you stop pointing the finger at designers? Companies are comprised of many more people.

Great points, there are pros and cons to process, the cons mainly being those who chain you to all steps in the process when unnecessary.

I’m with Brian, I’m not sure Apple TV was a design failure. Perhaps a product line design failure, but the aTV in and of itself, not so much. A lot of the aTV critique I read is why it’s so close to being a Mac Mini and why someone wouldn’t buy a Mini instead. I think the reasons are obvious, but that’s another conversation.

On Apple process being a mystery. Michael Lopp gave us a small glimpse into one design process used during his an Gruber’s “Blood, Sweat, and Fear” preso at SXSW ‘08. E-mail Wars has some decent notes on the presentation.

@Brian, @Dustin – agreed wrt the Apple TV improving…the first version was rank…chained content to the desktop. It does appear to be improving, but even Steve Jobs says that it’s an “experiment”.

The best designers, IMHO, are those designers who submit themselves completely to their process, or who have experience with many processes and are able to design the the process they use for an individual project based on the unique properties of that project.

2 factors seem to work for me:
1) Trust: Design takes trust. it is risky, but taking big risks is part of great design. Fail Big!

2) Flexibility: doesn’t really need an explanation

Now, I did say that there is success in strong adherence to a process/method, but you have to go all the way with it. It can’t be short cut along the way b/c if you do the foundation of the work falls apart.

Too many people say they are doing Contextual Inquiry or Goal Directed Design, but what they are usually doing is a bastardization of it, or a “discount” version of it.

The one thing that I find a constant in my “process” is the desire to have good insight into user needs, wherever it comes from. This is the one thing that gives me a lot more confidence in if I’m designing the right thing.

In the Design Observer article in which Bierut delivers the above quote, commenter Gunnar Swanson observes the following:

The phases in our proposals don’t represent the way we work. They represent that way we get paid.

This is so true for consultancies, but I imagine it is also applicable to people working internally. A good deal of what a design process is for is not to support creativity and innovation necessarily, but to ensure people in charge that the people working for them are doing the best job they can, not cutting corners, not skipping potentially valuable research, not leaving creative/conceptual stones unturned, not investing themselves in pet ideas, not avoiding hard work… making sure the ream is documenting their work well enough that their teammates will understand it correctly… basically smoothing over and mitigating all the mistakes that designers, left to their own devices, are at risk of making and, most importantly, allowing budgets and timelines to provide the primary structure for the design process.

I am absolutely sure that individual designers, working all by themselves on their own work almost never work according to a predictable process. I certainly don’t.

Basically design processes and methodologies are largely business and management devices, designed to bring order to what is often a chaotic and unpredictable process.

With respect to Apple, we often lament that Managementâ„¢ just doesn’t understand the value of design and daydream about how much better everything would be if we had that extra week or few thousand dollars to get things right. I suspect that isn’t a fantasy for Apple designers.

Many companies also wear their prototypes on their sleeves, whereas Jobs himself has said that Apple sweeps theirs under the rug, so you never see the half-baked ideas.

As for a process that yields consistently good results, I think there are at least parts of processes that, if used correctly, can consistently make it easier to get good results. One I’ve resurrected and have been almost obsessed with is the fitness graph concept central to Christopher Alexander’s Notes on the Synthesis of Form. Because:

1) Design (at least to me) is a problem of fitting a shape to a set of requirements (called misfits by Alexander) together, many of which are interlinked by a relationship of mutual support or conflict.
2) While it’s easy to get consensus on whether a particular requirement is valid, it’s almost impossible to get people to agree on whether or not it’s important.
2b) It is also hard to argue how a given requirement might be connected to others.
3) An excellent way to handle a complex problem is by splitting it up into smaller problems that you can recognize and act upon.
4) Humans are notoriously bad at understanding heavily-interlinked structures, and even worse at taking them apart.
5) Therefore, throw the whole mess into a computer and let it figure out how to cut the proverbial cake — because unlike people, computers are good at that kind of problem.

What the system described in the book will then produce is a program of smaller problems that you can confidently tackle independently of one another — like with sketches, storyboards, skits, paper prototypes or wireframes — and feed those results back into the system.

What the graph doesn’t do, however, is govern what requirements get put into it. However, the conventional process of stakeholder interviews, demographic and ethnographic research, personas (personae?) and scenarios can fuel those. The requirements entered into the fitness graph also need a significant amount of massaging to get them all non-overlapping and about the same conceptual “size”.

Although Alexander didn’t cover it, I don’t see why the structure couldn’t handle engineering feasibility, as well as iterations and/or stepwise refinement.

As a closing note, in writing this comment I resolved that no matter what process you choose, it is absolutely essential to treat the problem as novel and gather as much information as possible, for design is necessarily bespoke. After all, if you already had all the answers to a given problem, why not exploit that as a product?

You’ve probably read about how checklists in hospitals dramatically reduce death and illness. Not only is there value in using the list as an external memory device, but as Chris says, methodologies are tools that designers use to ensure that teams are in synch and help to de-politicize and de-personalize the creative process.

I think there is something more here, though. The nature of design problems – that they are often indeterminate – means that the solution isn’t apparent until it has been arrived at. While we may not know what the solution will look like, we do know how to get there. A good designer may have a sense of how to adjust his or her approach to get to a better solution.

“The implicit assumption is that if you perform some particular UX method then you’ll produce consistently better design: the right process = the right product”

Anyone who assumes that must also have assumed that the ‘process’ is controlled and predictable. That may be so for some engineering processes, but it isn’t so (and can’t be so) for a designer exploring a new set of challenges.
An abstracted process model might be the same for several projects, but it can only be so by ignoring the detail which makes the situation suitable for a design approach.

@christopher – excellent point…yet I’m still surprised at how many process pariahs there are out there…

@Dorian – awesome mention of Alexander’s framework. And I really like the idea that you “treat the problem as novel”…I must go away and think on that one some more.

@Neil says:

“methodologies are tools that designers use to ensure that teams are in synch and help to de-politicize and de-personalize the creative process”

I agree, and I like the idea even more because it’s for designers…not yet another deliverable for a non-designer. Keeps people honest, in effect. What I like about Bierut’s explanation is that he’s willing to keep some order but also allow himself that magic moment, when he sees where he must go. I think that designers do so many visual representations in part to elicit that magic.

This is a little tangental to the topic of processes, but I think a lot of this type of discourse depends on the assumption that designs are finite, stand-alone products. If a designer works in a continuous and holistic manor, constantly refining and improving the product, it becomes much harder to count successes and failures.

I think that Apple sees not just each product as an evolving design with iterations, but rather their whole strategy for a particular problem as a product. For instance, you can look at iTunes, the iTunes Music Store, and each type of iPod as separate products, or you can look at Apple’s overall strategy (to profit from improving the user experience of finding and listening to music) as a product.

I believe Apple’s success is founded less on process and more on the fact that they take a holistic and iterative approach to “product” design that integrates everything and everyone involved from design to engineering to manufacturing to business strategy.

On the post title, I think that there are practices that can be followed that will help a designer or design team limit the amount of uncontrollable variables encountered as their design is developed. That’s not to say that these practices themselves are controllable; it’s just that within each project there are practices that can be followed, as needed, that can help to reduce unwanted unpredictability in the design process – (e.g., unwanted = not helpful to the outcome of the design).

I understand the general argument for “good” design vs. “bad” design, but I wonder if what is being asked here is how the design process helps to guarantee a successful or unsuccessful product? I’m disagreeing on the term “good” vs. “bad” because it’s subjective, as opposed to “successful” vs. “unsuccessful”, which is measurable.

I also disagree that Apple is the only company to produce a string of successful products. I know it’s not a computer/digital company, but wouldn’t Porsche be an example? I do not own a car or a Porsche to test out my theory (donations accepted!), but I’m sure that they would agree with me.

If you use a process for design you get a “processed” design.

“Is there evidence that someone following a certain process produces great design every time?”

Of course not, that notion is silly. IMHO I’m not sure its even a good question.

Process can increase reliability, diminish risk, and provide a pattern of technique that can be methodically enhanced and improved upon. These are all good things, and when used optimally, process can have a positive impact during and on design.

You see, design is a process, not just an ouput – it’s both a verb and a noun.

Apple – who clearly have a defined process to diminish failures and risk, but are not always 100% successful – of course design success/failure is not a binary either. Without a process, the failure rate would be much higher.

Ross wrote:
Process can increase reliability, diminish risk, and provide a pattern of technique that can be methodically enhanced and improved upon.

Is there even any evidence of that? Is it possible that no process, gut feeling, or even random chance doesn’t produce success as often as, if not more than, thanprocess does?

At risk of this becoming a semantic debate – beyond randomness, is there such as thing as no process?

As I said before, design by its very nature is a process – from a large documented system, to the little iterations you do in your head whilst you play in photoshop – and that process is more reliable than randomness.

Process is evolved (based on an analysis of the past) in order to become better. You figured out approaches last time, and used some of them again this time – sometimes you try something random too.

Bierut may be right able formalised, locked down design processes, but just because it’s not predictable doesn’t mean it’s not a process… He also points out that beyond ‘creativity’, process does diminish things like risk and mistakes – which typically does make for a “better” design output – provided you measure a solution beyond it’s ‘creativity’ – which of course we all do right?

“A designer’s “process” is generally more subjective and sporadic than most are willing to admit.”

And definitely more subjective and sporadic than a potential manager or client is willing to hear. I tried telling it like it is once, believing I was in the company of people who understood that, beneath the surface, the process is in reality far less than stable. It didn’t go over well.

Honesty, it turns out, isn’t necessarily the best policy.

I think that with anything the way people create and design is a personal thing. Many work from re draft and designing what they may have previously found worked, adapting this principal to create something new, is that not a process in itself?

Maybe few people always design from scratch everytime they produce something – obviously their are exceptions-but you learn from your previous work to create new designs and you find your own niche through this trial and error ‘process’.

There definitely needs to be a process for design, most designers do not have the luxury of creating without a process (and a deadline, and a budget. Sometimes it’s a case of “too many cooks spoil the broth”. If our team designs to client specs, it takes away some elements of creativity if those specs are too detailed.

Great design can be replicated, but should never be copied. It should serve as an inspiration for other designers, who can then adapt certain design elements. In this way it can be replicated, but in an updated fashion.

Most designers I know are really proud of a small number of projects, and quite ambivalent about the rest…

This certainly resonated with me, thinking back to the various projects I’ve been involved in over the years (from being an ID at Yahoo, to working freelance, and everything in between).

And when I think about what led to the failures, more often than not it had to do with “soft skills” stuff more than anything — politics, egos, crummy managers, team dynamics…b.s. basically.

It’s sad and shocking how often talented teams and good ideas are handicapped by the aforementioned b.s. Cue the sad face

I should say there is that good design that is replicable. But Original designs are genuine that designers should be proud of.

Detox Patch

The process in which one makes a design should not matter. what should matter is the outcome and beauty is in the eye of the beholder.

voice over talent

I believe designers follow a process to quickly get them through the 80% that’s not so special, allowing them more time for the 20% that’s really special

Goggles

Looking back on my past experiences, I can say that some good designs fall through because of other reasons: the politics, the drama, or negotiations falling through completely.

indoor antennaplasma mountsoutdoor antenna

There is no process that guarantees good designs, only exceptional talent.

If I didn’t need my walking canes, I be out surfing

I struggle with this all the time… as a graphic artist/photographer/web designer I am constantly looking for the perfect balance, and great design, but at the same time a “process” to get the damn project done in a reasonable amount of time… like, on time would be good!

One of my new sites, LifeForceMentor.com, was a trip to work on. It showcases Life Force International. The difficuly with the project was the balance between “usability” and making sure that the site visitors get the “information” they need, but at the same time have a pleasant “visual” experience. I tend to lean on the visual side in my designs, but now that I am venturing into designing for a more “commercial” application and focusing on conversions, it seems the copy is a higher priority than how the site looks; for the most part.

My workflow is still pretty standard, cruise the web, books, magazines & music for inspiration, scetch it out, hammer out a mockup, refine & tweak, test, rinse, repeat. Not sure if it’s “genius” design but it works for me.

I would have to agree with some of the other comments too… some people just “have it” and others don’t. Process to me is simply keeping the “mechanics” in alignment and tuned for overall workflow. But you still need “creativity” and “talent” to pull off “good design.”

Apple may have the most kickass, high tech gadgetry available to mankind to create just about anything they want… but if their “designers” had zero “talent” then the iFamily very well could have bombed.

That’s my $0.02… keep the change.

PS:
Sweet blog by the way, glad I found you.
Will definitely be back.
Cheers!
~ Richard Bravo

I try and vary the design process for each project hoping it will have a different outcome. The problem with being too strict when it comes to having a process is you notice that parts seem to be nigh on identical across the board.

Consistency is good and a great way to get yourself involved but making things a bit dynamic can help too (I’ve found myself reversing my usual order sometimes and it worked quite well)