The Mythical User

Taking the mystery out of User-Centered Design

Don’t make your users repeat themselves

There are basic principles and pillars that make an app usable. The very most basic is “Follow standard practices whenever possible.” Directly following that, I think, is:

Don’t ever make your users enter the same piece of information twice.

Image from

We live in the digital age of computers. One of the primary functions of computers is REMEMBERING things. If a user has ever entered something, you should remember it and carry it with them, across pages, across functions, even across applications if need be. If you want to be successful, you should be customer focused. If you are customer focused, the value proposition of saving information for a user and prepopulating forms or shortening workflows for them is fairly evident.

We have all had the experience of having to fill paper forms out in triplicate for government organizations. We’ve also had the experience from the meme above and been impatient.  Why would you put that on your users, when a few lines of code can avoid it?

I once worked for a fortune 500 company where 12 different applications (And a team of 5 people working in tandem) were required to order a single product. The tools were all built by different teams in the company, with different backends and different data output. They didn’t share data between themselves, so at each step, the user would be required to export data and then (in the best case) import it to the next tool.  Sometimes the user had to enter the data manually, and they would literally print it out and set it next to their monitor so they could type it in exactly the same.  I am not kidding you.

Probably, you’re not doing anything quite that bad.  If you are, hang your head in shame, and then fix it.

But you might be asking them for their name in one place, and then for first name and last name somewhere else.  Parse that bad boy out.  Maybe your phone system uses the phone number to look up the customer to see what tier of support they qualify for.  Then make sure your CRM is tied into the phone system so that when the customer is routed to the appropriate representative, that rep has the customer’s info and doesn’t have to ask for it again.

Google’s autocomplete is a prime example of this principle in action.  When you’re typing a search into the form, Google will autocomplete first with things you’ve searched for before, then with the most likely guess.  This is also one reason that Facebook sign ins are so popular – you use Facebook to sign up for a site and your info is there without you having to enter it in the new site. It’s like magic!

In the end, remembering things is probably one of the easiest tasks you can ask your software to do.  As users become more comfortable with computer interfaces, they will begin to expect this basic courtesy, as surely as we expect someone to respond “hello” when we greet them.

Today’s Interesting Link:

SVGeneration – This site is so much fun your head might explode. It’s got a ton of wonderful, tiled backgrounds that you can customize and then generate as SVG files. Go ahead, try not to get addicted. You’re welcome.

 Today’s Usability Quote:

It takes considerable knowledge just to realize the extent of your own ignorance.  – Thomas Sowell

Today’s Music To Design To:

Silent Shout, by the Knife. With a mysterious, slightly dark groove, this is synthesizer-based music that manages to be futuristic and uplifting and tickles your creative bones in all the right places.  The vocals even have that sort of airy but ominous feeling that lives somewhere between german synth pop and Lords of Acid.  Seriously, it’s worth a listen.

Leave a comment »

Design your back door

I keep company with architects and the children of architects, and I love buildings. I often think of my UI design process much the way architects think of the building design process.  You design your pages around the content.  You design a building around the primary space.  You ensure there are enough bathrooms for the number of people likely to be there at any given time.  You build in the features your users are likely to want just before, and just after your golden path task.  You make it accessible to anyone, regardless of physical ability type.

And I notice, that in much architecture, just as in much design, the back door becomes an afterthought.

Have you ever been amazed at how beautiful a building is, and wandered around it in awe, only to find yourself in a plain white hallway that leads to a mass-produced back door with chips in its paint and dirt pushed up against the doorjam?  Did you question it, and did it make you sad, or did you just accept it because it’s not the face of the building?

I got to tour the Pixar campus a while back.  It’s gorgeous.  And you know what?  Their architect even thought about their back doors.  Sure, they’re functional and not as beautiful as the rest of the building.  But the one I passed out through was Pixar-crisp.  A detail that the designers of the interior spaces looked at, thought about, and assigned.  It fit.

Why am I going on about architecture?  Because I want you to think about your back doors.  I want you to think about the alt tags and titles on your images and links.  I want you to think through the way your emails look with images off.  I want you to think about how your website looks with ads turned off ( does the neatest thing – they have a placeholder behind their ads that suggests you donate to have an ad-free experience).  I want you to DESIGN the voice browser experience.  I want you to think through your error pages and 404s and make them build your brand and help your customers in some meaningful way.  And I especially want you to think about your unsubscribe process.

I have a checklist for QAing and specing and even designing.  When I think I’m done, I go down the list and there’s always one thing I missed.  It includes things like

  • What does it look like in mobile?
  • How does it interact for touch?
  • What’s the SEO impact?
  • Does this affect other areas of the site/app?  Should it?
  • Are there emails associated and what do they look like?
  • What needs to be tracked?
  • What are the alt tags?
  • How does it share socially?
  • What’s it like to print it?
  • What does it look like when logged in? Out?
  • How do they get help?
  • What’s the back-out plan for the user if they’re not happy?

There should be no aspect of your app that’s accidental, or arbitrary, or forgotten.  Sure, you can have a perfectly successful website  without thinking about these things.  I’d say 90% of the successful websites out there do.  I once worked at a company that had a net promoter score of 93 (that’s astronomical, in case you don’t know) and they sucked at so many of these things – but the users were perfectly happy.

However, there is no downside to thinking about these things.  No one ever said “I liked that building until I saw that it had a beautiful back door and now I hate it.”  But maybe, just maybe – one person was on the fence about the building until seeing that beautiful back door gave them just the perfect subconscious feeling of elegance, refinement and conscious design.

Remembering your alt tags is a small investment for gaining one more delighted customer.

Today’s Interesting Link:

Trask Industries – This is the best movie promotional site I’ve ever seen. Even leaving aside my bias to immersive multi-channel entertainment experiences, and just pretending this is only some company’s brochureware site, it’s freakin’ awesome.  CSS animations build as you scroll, the page reacts and feels alive.  This is good design.  And it has Peter Dinklage.  You can’t get much more win than that.

 Today’s Usability Quote:

“Not having confidence will lead to bad decisions”  – Dave Mott

Today’s Music To Design To:

Coco Part 2 by Parov Stelar is, like much of what I recommend, at once exotic, energetic, and ambient enough to not distract you.  I love how it makes me sway while I work, and I think you will too.

Leave a comment »

Death by a thousand bugs

I have never met a company that wasn’t in a hurry.

Whether you’re a startup rushing to push product so your board of directors doesn’t get antsy, or a huge multinational corporation fending off the startups who move ten times faster than you – every company is in a hurry to get product out the door.

In that hurry, certain tradeoffs get made.  Bugs are allowed out there into the wild. They have to be, because if we are waiting for perfect, nothing will ever ship.  Only best-selling authors like George R. R. Martin can wait until their art is absolutely polished and perfect before it goes out into the world, and I’m willing to bet even he sees bugs in his stuff.

I’m a fan of bugs, anyways.  They’re often cute and they come in shiny colors like beetle green and butterfly blue.

But what happens after you let that buggy product go live? When do you go fix those bugs? And what’s the threshold at which there are too many bugs?

Everyone knows that if there’s a huge bug that stops people from being able to use the most basic feature of a product, that’s a showstopper and it’s not going to go live.  Engineers fix those as if their hair was on fire, and sometimes it is. But what about the hex value that is just a few letters off?  Or the corners that are square instead of rounded?  Those, we can let slip through, because they’re not, in isolation, going to significantly harm the user’s experience.

Where you have to be careful is in two places.

A thouand papercuts will kill you.

Although none of those little bugs in isolation are the cause that your users aren’t engaging, or your learning curve is high, or your bounce rate is high – a ton of them add up.  Three tiny bugs in this project, three tiny bugs in that project, three in this other one…. now you’ve got nine tiny bugs in your UI.  Remember that your users don’t see your app in terms of projects.  They see it in its entirety, and those little inconsistencies start to add up.  And let’s face it, you are WAY less likely to go fix that hex color bug than you are to build the next new project.  So every one of those little guys is going to join together on a million bug march against your users eventually.

  • Establish a maximum number of little bugs, per project and for the whole app.  Be faithful to this max
  • Have a commitment that every engineer fixes three tiny bugs in every sprint or dev cycle
  • Foster a culture of pride in workmanship in every employee, from the janitor to the CEO – so that everyone is eager to fix even the small imperfections.

It’s more fun to look forward.

If you work at a software company, you have probably either lamented the inability to go back and iterate on that project you just did, or you’ve gotten sick and tired of staring at the same screens of the same code for months and months and you never want to look at it again. Most likely, both at different times.  It’s crucial that you put out your minimum product, and then do sweeps through it again.  You MUST revisit the product, if not to update features, at least to fix bugs.

  • Have a plan for when medium-sized bugs will be fixed before the product even launches.  Be faithful to the plan.
  • Either dedicate one engineer to cleanup, or have every engineer fix one medium to large bug in every sprint or dev cycle.
  • Consider having a sprint just for cleanup and bug fixing, and let your devs and designers choose the bugs to work on.  This feeds into a culture of ownership and pride.

It’s important to understand that quality is a usability measure. It’s obvious why the big bugs hurt usability – they stop the user from doing what they need and want.  However, the little bugs add up and cause your users to feel like you’re not as professional as your competitors, or like you might not be trustworthy.  Credible research has shown that how attractive, consistent and brand-loyal a web site or application is significantly affects user’s opinions of the site – the exact same steps can feel totally different depending on the presentation.

Instill a pride of craftsmanship in every employee.  Empower people to take the time to fix bugs. And don’t let the little bugs pile up to take over your world.

Today’s Interesting Link:

Photoshop Secrets – This site is like my candy store.  Sometimes there’s stuff I already know, but most of the time the tips are super-awesome power tricks that I would never have known without it.

Today’s Usability Quote:

“My goal is to omit everything superfluous so that the essential is shown to the best possible advantage.”  – Dieter Rams

Today’s Music To Design To:

Ayurveda Buddha Lounge #2 – it’s largely ambient, energetic music that soothes you without distracting you. I don’t know how to describe it beyond that. It’s not particularly unusual music, it’s just really great for creative endeavors.  I also listen to this when writing, so it’s probably great when you’re doing a lot of copy-intense work.

1 Comment »

What’s Your Ultimate Goal?

IMG_20130107_150314People who work with me hear the same questions over and over again, almost to the point of shellshock.  One of my proudest moments was when a VP sheepishly told me that in a meeting she had channeled me and asked those questions.  She felt guilty, but I thought it was the best thing ever.  It means she’s internalizing them!

What are these questions, and why am I so obsessive about them?  These are the questions that get the conversation moving in the direction I need it to, so I can understand the intent of the product or feature.  Without intent, you’ve got nothing and you’ll waste everyone’s time with design and spec cycles that are never right.

1.  What’s the goal for this project?

You have to ask this to get the conversation started.  You’ll get a complicated, detailed answer, probably.  It might involve metrics.  It will probably involve the name of an executive.  All that’s fine – it’s your starting point to dig deeper.

2.  Who is going to be using it?

Even if the goal answer told you this, you need to ask this question to get them thinking deeper about the end-users and the end-user result.  The answer is almost always slightly more complex than their initial response, so you should ask about other types of users and dig deeper to make sure you don’t miss those incidental, corner-case “other users”.

3.  So what, ultimately, do you want that user to do?

Follow this question out several steps.  So often the answer is “click the button.” but that’s not the ultimate goal.  The reason they want them to click the button is to proceed to the form, because they want the user to submit the form, because they want the user to sign up and take classes at a university.  So ultimately, what they want the user to do is to become a student, not click a button.

4.  What user problem or pain point are we addressing here?

Be cautious of a company that consistently does projects that don’t address a user issue.  They won’t be around forever.  However, understand that sometimes, the perfectly valid answer will be “This isn’t something that bothers users, it’s something we have to do to remain profitable.”  In those cases, use your past and current research to find user pain points you can address in the course of this project.  Never do a project without some tangible benefit to the user, if you can avoid it.

5.  So what’s the ultimate philosophic goal here?

No metrics allowed in this answer.  Make them talk about how the user should feel, what they should do, and how that affects the user’s relationship with the company.  Those are the key things that will influence your product design, whether you’re making the pixels, writing the code or penning the spec.

Sometimes, as you uncover intent, you discover an entirely new way of approaching the problem.  Sometimes you realize it’s not a project worth doing.  Sometimes you realize it’s way bigger than you previously thought.  But in every case, you’ll be able to approach it with a stronger understanding of the requirements and conviction that you’re doing the right thing.

Today’s Interesting Link: – There are so many ways to view your site across multiple devices out there.  This one is nice because it’s very quick, and it’s not bad to send clients to.

Today’s Usability Quote:

“For, is it not written the Holy Manual of Style :  “Lo, and the goddess of Eeuu Ayee did come forth down from the hill.   She did look upon the masses and Spake Thus to all.  ‘THOU SHALT NOT COMMIT THE CRIME OF BLINKING BUTTONS’ and lo it was so.”  – Martin
Bogomolni, referring to me.

Today’s Music To Design To:

I’m frankly surprised that I haven’t mentioned these guys yet, so if this is a duplicate recommendation. Soundtracks are so good for working, and Lagaan: Once Upon a Time In India is brilliant like that.  It’s exotic but familiar, sweeping and dramatic.  It’s hard not to be both soothed and energized when listening.  Check it out.

Leave a comment »

What Do You Do?

In social settings, this is one of the most common questions.  Especially in America, we tend to define people by what they do for a living.  With UX people, that’s actually quite apt.  Like teachers, we don’t leave our work at home.

However, the answer to the question, “What do you do for a living?” is a faceted truth for a UX person, and we should apply the principles of user experience to it.  Not everyone will understand what we mean when we say “I am a user experience designer/researcher/evangelist” and it’s unkind to make people look at you with that head-cocked-to-the-side-puzzled-dog expression.

So, to each type of person, you can give a different answer without in any way being dishonest.  After all, they don’t really NEED a treatise on the entire umbrella of UX.

  • To a layman who knows nothing about computers, you might say “I work with computers. I try to make them easy to use.”
    I also say this if I don’t know anything about the person I’m talking to, since this is the safest of the answers.
  • To an artist, you might say “I’m a graphic designer.”
  • To a psychologist, you might say “I’m a software usability researcher.”
  • To someone who doesn’t know much about computers, or isn’t interested in them, you could say “I make software/websites.”
  • To Cory Doctorow, when getting a book signed with a long line behind you, you might say “I’m a UI Designer.”
Simplifying what we do down to a single bite-sized sentence might seem self-deprecating, but it’s really just a kindness.  If someone demonstrates knowledge of or interest in the field, you can expand on it.  If they don’t, you’ve done them a favor by not listing off a ton of things that make them feel stupid by comparison.
There  is one exception.
  • To a person who might possibly hire you someday, you say “I’m a [give your exact title].  I do [brief list of the things you do].”

Why?  Because this might be the only time you ever get a chance to talk to that person, and they should have all the information they need to know whether they want to continue talking to you.  You should keep your description of what you do to two sentences, and only expand if they ask you to. My own example:

“I’m a Director of User Experience.  I do usability research, interface design, information architecture, evangelism and ux team management.”

I hope this helps you have smoother conversations, and helps you see that the principles of user experience apply to everything you do, not just software design.

Today’s Interesting Link: is a useful PNG compressor for those who have trouble using other tools.  It’s easy to use and does a darn good job.

Today’s Usability Quote:

“Our role goes beyond just making visible products; we need to make objects that are relevant, meaningful, and empowering.”  — Stuart Karten

Today’s Music To Design To:

Flamenco Arabe by Hossam Ramzy is exactly what it sounds like – Flamenco music performed by an Arab musician.  If you enjoy either style of music, you are certain to love this album.  I find it evokes warm colors and intricate designs when I listen.

Leave a comment »

User Experience goes beyond your website

From time to time, I find myself inspired to remind a company:  User Experience is the ENTIRE experience your customer has.  It’s not just the clicks and pixels on your website.  It’s how your customer feels that you, as a company, feel about them.

With this in mind, there are a few rules you should follow on your website, even though they might seem counter to your business needs:

1.  Make sure it is easy to contact you by email and phone.  Make sure this contact information is displayed on your home page at LEAST.  Better is that it should be displayed on EVERY page of your website.  Most users will look to the footer of your site, so you can either put the contact info right there (best) or have a “Contact Us” link (not bad).
I know that there’s an objection to this because if you give customers a way to contact you, you’ll have to pay someone to answer emails and phones.  Expenses go up.  However, if you do NOT give customers a way to contact you, they will choose to do business with your competitor who has a prominently displayed 1-800 number.  Trust me on this.
By not giving them a way to speak to you, you are essentially saying “We don’t care what you, the customers, want or have to say.”

2. Make it easy to unsubscribe.  In emails, have the word “unsubscribe” be a link to an instant unsub. Don’t use some funky sentence that beats around the bush but makes marketing people feel comfortable.  Use the word “Unsubscribe”, because that is what users are looking for.
On your website, have an unsubscribe link in your megamenus or your footer.  My rule of thumb is that a completely new user should be able to find the unsub link within three seconds.
The objection to this is that if you make unsubbing easy, your unsub rate will go up.  I won’t lie; it might.  But honestly, the RIGHT way to get unsubs down is to improve your offering.  Make it better, and people won’t even look for the unsub link.
By making unsubscribe hard to find, you are trapping users into a subscription they don’t want.  They will resent your company, which does harm to your brand – and they will either delete your emails without opening, or set up a filter so they never have to see them.  Worst case scenario, they’ll mark you as spam.

3. Don’t make your users pay for basic support.  If they paid for your product, it’s in your best interest to give them basic support for free.  The more they use your product, the more they will want to buy the next version, and the more they will want to  tell their friends about it.  If they have a problem, and you don’t support them, then they will give up on you and tell everyone they know that your app doesn’t work.  This is most especially true in the first week of a user’s acquisition.

3a. A knowledgebase is not enough support.  Unless you are an open source application, then you need more than a knowledgebase.  You should have some version of live chat capability, a support phone number, and a good support ticketing system.  Don’t make your users rely on FAQs and poorly edited form answers to get help.

There are plenty more rules for making sure your customers have a good all-around user experience.  What are some of your favorites?

Today’s Glossary Term: 
Multivariate Testing – Everyone knows that A/B testing is when you show some of your audience one version of a page, and the rest of the audience another version of the page.  This is great for testing drastically different designs to see which one performs better.  Multivariate testing is what you use next – once you pick a design, you start to tweak things.  You will run five, ten, maybe twenty MV tests at once, changing a tiny thing here, a tiny thing there.  It’s a very rigorous, scientific test that allows you to learn that the combination of headline A with button color B and button placement C and image D works best for your audience.

Today’s Interesting Link: is a fascinating look at A/B tests and what did better.  The author of this blog has a great understanding of the fact that every audience reacts to things in a unique way, and there’s no best practice that works for everyone.

Today’s Usability Quote:“Things should be made as simple as possible, but not any simpler.” – Albert Einstein

Today’s Music To Design To:
Have you heard the TRON: Legacy soundtrack?  The incomparable Daft Punk does classical.  And it’s brilliant.  It will totally put you in the creative mood.
Download the MP3 Album or Buy the CD

Leave a comment »

Finding a UX Job – In a Recession

We’ve all heard the lines – if not delivered directly to us, delivered to someone we know:

“We’re tightening our belt right now, and User Experience is not core to our business model”

“We’re taking the interface in a new direction, and we need fresh eyes to do that.” (and then they hire someone at half your salary)

“The product managers have learned so much from you, they’re going to take over your role.”

Whether you’ve been laid off, or you’re just looking to move on, job hunting for UX people in a recession is HARD.

Eight months ago, my employer (at the time) and I mutually decided it was time for me to move on.  I spent two months looking, and in the end I found what I love to refer to as a “pony” job – the perfect job in the perfect company, working with the perfect team.  I’ve never been happier.

Unlike the dot-com crash of 2001, when there simply WEREN’T any jobs out there, there are still plenty of UX openings for the taking.  I still get 3-4 emails a week from recruiters who have opportunities that are great matches, in my area.  The challenge comes from the fact that there are a lot of really excellent candidates out there, hunting for good opportunities.  The competition is stiff, my friends.

But it’s not hopeless.  If you’re in the job market, expect to be looking for around two months.  If you’re not picky, possibly only one month.  I applied to 93 total positions.  I interviewed at about a quarter of that.  I got several offers, but it was important to me that I find the RIGHT place, not the first place.

Whether you’re looking for the perfect job or just a paycheck, here are a few helpful hints:

Job Boards

Post your resume on every single one you come across.  Update it weekly.  I don’t care if all you are updating is your skillset or one word in your description.  Every time you update, you get bumped up in search results.


I’m not going to enter the debate about one page or multiple pages.  Do what you think is best, there.  However:

  • Have a list or grid of skills.  Not sentences, phrases.  Example:  graphic design, user interviews, heuristics
  • Forget the “goal” – that’s an outdated relic.  Instead, have an introduction that describes what you can do, and how long in total you’ve been doing it.  It’s like a two sentence cover letter.
  • Treat your resume like an Information Architecture project.  What’s the most important info?  Make that very prominent.  What’s next?  Put that nearby.  Etc.
  • Include all your contact info on your resume.  Recruiters will tell you to take it off, but the job board resumes should have it.
  • Include a link to your portfolio.   If you do ANYTHING, people are going to want to see samples of your work.


A lot of people feel recruiters are like used car salesmen.  Honestly, some are.  However, I’ve had fantastic success with particular recruiters and I swear by them.  I like them so much that I keep touch with them for years afterwards, and refer friends.  Build relationships with some good recruiters and they’ll take great care of you.

  • Reformat your resume however the recruiter asks you to, for them.  You don’t have to use that format anywhere else, but the recruiter knows his clients the best, and he knows what they are looking for.
  • Insist that your recruiter tells you every company/position they submit you to.  Bad recruiters will send your resume to positions you don’t fit, and it reflects badly on you.  Plus, if you then submit your resume directly, the hiring manager will only remember that your name is associated with something negative and you’ll hit the round file.
  • Communicate with your recruiter.  Think of them as your advocate.  Or your best friend, or your bodyguard.  Whatever it takes for you to treat them like a partner.

Social Web

Use your friends and your social networks.  Announce that you’re looking, unless you’re doing it in secret.  Talk to everyone.  Go to networking events.  Do side projects.  Offer free advice, consultations or etc.  Word of mouth is invaluable, and nothing gets you an interview faster than some impressed acquaintance saying “I know this amazing person…”

Also, momentum is a great thing.  If you know people with similar skillsets, ask them to pimp your resume out when recruiters reach out to them.  I regularly pass on resumes when recruiters touch base with me – it helps the job seeker and the recruiter.

Keep Upbeat

I know this sounds weird, but the more positive, cheerful and upbeat you are, the better your chances are.

As a job hunt drags on, you can start to wonder, “What’s wrong with me?  Why don’t I have 10 offers and a bidding war yet?  Am I unemployable?  Am I outdated?  Are my skills too weak?”

Don’t fall prey to this way of thinking.  It shows through in your body language, your written communication, and it even causes you to make bad choices about where to apply.  Keep your eye on finding a “perfect” fit.  If you haven’t been hired yet, it’s because you haven’t been exactly the right fit, not because you haven’t been good enough.


Try to find out as much about the people you are interviewing with, how long the interview will take, and what you should bring before you ever go in.

  • Take a printed copy of your resume AND the job description when you go.
  • Take the time to read the website, try the product, and get familiar with the company you’re interviewing with.
  • While you’re doing that, come up with three questions about the company or product that dig deeper in than their website, FAQ or press materials go.
  • Dress nicely, but not TOO nicely.  There’s nothing wrong with asking how you should dress for the interview.  Especially with dot com companies, wearing a suit might lose you the job.  They might decide you’re too stuffy.
  • Be prepared to answer some canned questions.  Think through the answers ahead of time so you aren’t caught off guard.  Do an internet search on interview questions and you’ll find a ton of lists.
  • Be prepared to do a test project.  This isn’t free work for the company, so don’t get irate.  They aren’t going to use what you’re doing, unless they’re really unethical.  They just want to get a sense of what it’s like to work with you.  Be pleasant, flexible and fast and your skills will speak for themselves.

Be Ubiquitous

Unless you’re looking for an entry level position, your potential employers are going to Google you.  Be found.  Blog posts, forum comments.  LinkedIn profile, facebook profile, livejournal.  Make sure you’re everywhere and that everywhere you are reflects your professional persona.  You want potential employers to know you’re not going to embarrass them, and that you’re respected and passionate about your profession.

There’s no magic wand that will fix the economy and make our 10% unemployment rate go away.  But at least for UX people, the situation isn’t as dire as it was eight years ago.  Chin up, carry on, and go find that “pony”.

Today’s Glossary Term:
A/B Testing –  this is a process by which you create two versions of something, either slightly different or very different.  You then serve up version A to some of your users, and version B to the rest of your users.  This is, of course, simplifying the concept, but you get the idea.
A/B testing can answer little questions like “does a red button or a green button encourage conversion?”  It’s invaluable, but it takes some infrastructure to set up unless you have…

Today’s Interesting Link:
Google Web Optimizer – GWO is a free (for now) tool which lets you do A/B testing without building an enormous infrastructure on your own.  It’s not going to tell you everything you might want to know, it’s not going to be as convenient as having a home-grown system, but it’s nearly immediate, and it’s darn easy to use.

Today’s Usability Quote:
“Perhaps the most difficult thing an artist has to do is evaluate the quality of his own work.” -Peggy Hadden

Today’s Music To Design To:
Boards of Canada was introduced to me by an artistic genius and all-around neat guy a few years ago.  You can’t go wrong, with Boards in your headphones.  It’s downbeat, but happy, musical but not lyrical, and energetic without being thumpy or making you anxious.  It’s good driving music, and great for those long hours coding.
Download some MP3s or
Buy a CD


Evolution vs. Revolution: How to effect change

There’s a term that people use for a job that is everything you could ever want, right off the bat: a pony.  For a designer, the pony job is one in which you come in to a well thought out, well researched product that has no pre-existing interface and will be built using the best technology for the job, after you’ve had ample time to do  your user research and design.

But that NEVER happens.

More often than not, the case is that you hire on to a company that either has an existing product with established design conventions (good or bad), or an emerging product whose original design was done by engineers and the CEO while they tried to find and hire you.  Don’t get me wrong – great products can and have come out of these situations.  But frequently a designer joins a team or a company and in addition to designing the ever-flowing panoply of new features required to stay competitive and build brand, they also have a mess to clean up.

So how do you do this, without losing ground a competitive marketplace?  Design takes resources – Product, Marketing, Engineering, QA, sometimes even executive.  While those people are working on requirements, testing, coding and verifying all these fancy fixes you want, they’re not doing the things necessary to whoop your competitors into forlorn lumps of also-ran.

Yes, we all know that having a usable website or application is an integral part of remaining competitive.  I take great pleasure in reciting a story about a director of E-commerce at a Fortune 500 company that hired me as a consultant.  When I was giving him the standard ROI speech on usability, he actually said to me “We got this far without doing any usability.  Why should we start now?”
That director no longer works for that company.

Joking aside, though, it is a tricky balancing act – how do we update our past code while still driving forward with new?

The answer is evolution.
Evolution is a multi-general process in which an organism changes to adapt to its environment.  Your website or application SHOULD be a living organism of sorts, growing and shedding and hopefully healthy.  There’s no reason your website can’t evolve.

If you think about it, stopping everything to have a revolution – major relaunches, complete identity changes – is an enormous and painful act.  It costs unimaginable amounts in money and time, not to mention your users’ pain as they relearn everything they were comfortable with.  It sets you back in brand identity, because you have to rebuild everything you’ve built.  And the potential for things to go wrong, the risk of failure, is enormous.
It’s not that there’s never a time for revolution.  We can all think of one or two incredibly successful relaunches, or prominent redesigns.  But honestly, it’s a drastic measure that should be left to the giant behemoth corporations, or used as a last resort.

Because you can always evolve.
Start with a plan, a vision for your end goal.  I like to draw pictures and make spreadsheets, but other people might work differently.  This becomes your strategy, and is much more powerful than any buzzword-filled strategy statement will ever be.
Then take the most heinous wrongs and right them first.  What are the usability holes your users are complaining about?  Will converting static pages to AJAX enable ten other improvements later?  Can you revamp the golden path all in one fell swoop?  Start one at a time and break out all the tiny tasks that make up the grand vision.  Let your site mutate.

Your users will complain.  Unless you have the most complacent user community on the interwebs, SOMEONE is going to complain about every single thing you change.  I have actually had users complain when I reduced clicks.  Listen, but don’t be discouraged.  Most of the time, if you did your research well enough, those same users will fall in love with the new stuff after a few uses.  They’re habit driven.  And if you were wrong, and what you did actually DID make things worse, well, you only have a small mutation to roll back, rather than a huge redesign.

It’s not ideal.  Who wants to change one page at a time and have potential inconsistencies with their names attached?  Who wants to wait a year or two for pages to be tiled?  But in the big scheme of things, it’s worth the wait, and it’s a great way for a design team to empower the corporate strategy, rather than slowing it down.

Today’s Glossary Term:
Golden Path -The primary task flow or navigation flow of a website or application.  If you can’t instantly identify the golden path for your app, you should immediately sit down and figure it out.  What is the one thing that everyone should or wants to do when they come to you?  And how do they do it?

Today’s Interesting Link:– Both beautiful and brilliant.  Imagine a web interface in which you don’t click.  Then follow that link and go experience it.

Today’s Usability Quote:
“The greatest mistake you can make in life is to be continually fearing you will make one.” Elbert Hubbard

Today’s Music To Design To:
Dead Can Dance have been around for too many years and have too many albums for me to recommend just one.  Honestly, almost anything you find from them is great, and it’s all got a mellow, exotic groove that lends itself amazingly well to creative work.
Check them out at

Leave a comment »

Rapid Prototyping: Slideshows

First of all, let me express to you how important rapid prototyping is.

You absolutely ought to do requirements gathering first, then come up with a theory of what your users want, then build a quick-and-dirty prototype and take it back to users – both the ones you gathered requirements with, and others. This can take you a few days and cost a few hundred dollars, and it can save you hundreds of thousands in development, materials, or man hours.

So now that we’ve established that you all agree and will employ rapid prototyping before you build anything big, let’s get into the meat. There are a lot of ways to do make a prototype. Depending on your medium, you can build a cork-board model, draw pictures on a piece of paper, or whip out wireframes and screenshots. Regardless, you shouldn’t spend too much time on them – depending on the size of the project, we’re talking an hour to 2 days, no more. They don’t have to be perfect, fancy, interactive, or anything along those lines. They are nothing more than a tangible representation of an idea or concept. Nothing more.

One way, particularly useful for an experience (like a tour), or an application (web-based or desktop) is to use a slideshow.

Irfanview is a free, easy to use downloadable application. It has lots of features, but the only one I ever use is the slideshow builder. You can drag screenshots in, and then set controls on each for mouse click, arrow, etc. This enables you to get on the phone with a remote user, and bring up the screenshots in any order you want, without opening gifs from a folder. It can completely fake user interaction, as far as clicking goes – click with the mouse and it can take you to a next screenshot that shows a list expanded or a field filled in.

It saves out to an .exe file, so you can then send the “prototype” to anyone you want. You can make it go from slide to slide automatically, or control each slide individually. File sizes are reasonable. I also use it for training, and believe it or not – I have a version of my portfolio in an executable slideshow. Many possibilities for a little free gem!

Irfanview’s interface isn’t the prettiest or simplest, but the basic functionality only took me a few minutes to figure out. It’s incredibly powerful, when you’ve got a test you need to keep running smoothly.

Enjoy this tool, and if you find others like it, please comment and let everyone know!

Today’s Glossary Term:
AEIOU – A research method focused on Actions, Environments, Interactions, Objects, and Users. It’s primarily used for ethnography (studying people and making a mental model of how they see the world) but can also be a good mnemonic to spread throughout your team. It helps to get everyone thinking of the world from the user’s point of view.

Today’s Interesting Link: – Is one of the places I go when I’m feeling boxed in by a design. It’s a great resource for inspiration, or if you’re looking to improve your CSS skills. If you don’t already have Zen Garden bookmarked, you really ought to.

Today’s Usability Quote:
“The greater danger for most of us is not that
our aim is too high and we miss it, but
that it is too low and we reach it.”
– Michelangelo

Today’s Music To Design To:
Getting away from the edgy electronica, it’s time for a Jazz recommendation. However, Jenna Mammina is not “just” jazz. She uses her voice like an instrument, switching between belted lyrics and then lilting back into a soft, sultry croon.
Download the MP3s or Buy the CD

Leave a comment »

24/7 Usability Methods

Mark Hurst has a great post on “Listening Labs” in his excellent blog. In it he talks about a method I can’t recommend highly enough – a freeform session in which you watch your user actually use your product, their way. No test script, no tasks, no standardized objectives. Just “show me what you do”. I’ve been using this method for about 10 years now, to identify problem areas, build personas, and flesh out requirements.

What surprised me about Mark’s article is that he called “Listening Labs” an unorthodox method. I’d never thought of this as particularly clever or unusual, so it got me thinking.

There is a mindset amongst us product people that user feedback is best solicited through guided methods. Interviews, surveys, lab time, observations… all of these things structure your time with the user and are great for targeting specific stuff. But I feel that it’s absolutely crucial to do an additional sort of data gathering – we’ll call it 24/7 Usability.

24/7 Usability means that at any given moment of any given day, you are soliciting feedback on your existing product. There are a number of ways to do this, and you’ll be amazed at how simple they are.

The Feedback Link
First and foremost, the very minimum you should do. You should have a way for a user to give feedback from any page or screen in your application. The feedback should have a form, so you can get enough info to respond to them – and it should be called “Feedback”. Seriously “Tell us what you think” or “suggestions” or “comments” or any number of other cute, non-standard things may give your site/app personality, but it’s not necessarily going to be instantly recognizable to a confused or frustrated user.
The result of this form should go to customer support, of course, but the user experience team or professional should be CC’ed. Just work out who is the responder, and make sure you’re not double-teaming your poor users with responses.

Message Boards/Forums
I can’t tell you how valuable the Usability Forum for my product is. I post survey questions there, I answer interface questions (each one identifies a problem with the design!) and I get requests. I can even share screenshots of proposed designs, since only registered users can get to the forum. I keep a spreadsheet that logs every feature request and confusion point, along with how many users have requested it, and what type of user they are. When I’m building my strategy, I refer to this spreadsheet like a bible.
When a new feature gets released, or an old one gets improved, I head back to the forum and comment on every post that asked for it, letting users know we listened to them. As a result, our users have confidence that their feedback is being heard and acted on.
A bonus is that because the forum is public, if a user requests a feature that doesn’t make sense to other users, those other users will dispute their request, and they often work out a compromise between themselves – without my interference!

Your Customer Support Team
If you’re not already good friends with every one of your customer support folks, you’d better bake them cookies and get on it. These people are the front line of usability. They hear all the complaints – and the compliments!
I’ve got an arrangement with my support managers – whenever anyone in support gets a call where the problem is interface, or they can’t find a button, or don’t understand a word – support sends me an email. These all get tracked in that Holy Spreadsheet too, with special annotation that it came from a support critter. Those get higher weight because they were bad enough that someone picked up the phone and made a call.

Your Sales Team
Believe it or not, your sales people are just as important as the customer support people. Here’s where you have to be careful, because there are often qualifying factors. However, a salesperson knows that when she is demo’ing a product, and three potential customers ask the same question, there’s a real problem there. I’ve even taught my salesmonsters how to ask followup questions in a non-leading way, empowering them as mini-moderators. These guys are a fantastic source of feature requests, and they can give you the pulse of a customer segment you REALLY want to make happy – the ones you don’t have yet.

All too often, apps – whether web-based or dowloadable – are built in a hurry, with an eye to timelines and no thought of the future. However, it’s important to build into your architecture the ability to track everything a user does.
There are privacy issues, of course. However, it’s incredibly simple to assign a different token to each of three links to the same page, so you can tell which one is getting used the most. Qualitative data is precious, but nothing beats sheer quantitative data to show you what you’re doing right and wrong.
Take the extra day, or week, or month, to make sure that you’re tracking user behavior without violating privacy or user trust. It’s well worth it.

The point is, usability shouldn’t be an on- and off- thing. If you only test in spurts, you’re likely to miss some very important issues and opportunities. Consider every contact with a user, whether that contact is via website, person, or product as an opportunity to do 24/7 usability.

Today’s Glossary Term:
Wireframe – A rough diagram of a page/screen, either sketched by hand or made with black & white lines, that generally indicates basic layout and what info/functionality is shown. These are good for first walkthroughs, high-level validation of a concept, and finalizing requirements. They’re also great when you’re playing with multiple layouts or methods, because they take very little effort/time and can be very illustrative.

Today’s Interesting Link: – It’s got your usual web 2.0 look and feel, but the layout is useful, primary functionality is well highlighted, and the concept is great. I frequently have the problem of having too much of X and not wanting to let it go bad – with this site you enter an ingredient and it pulls up recipes.
I particularly like the suggestions on the search box, the mouseovers on everything, and the reactive nature of the site. Well done!

Today’s Usability Quote:
“If it was magic, how would it work?” – Alan Cooper

Today’s Music To Design To:
Ape of Naples is a fantastic album by Coil. If you’re familiar with Coil, you’ll enjoy the dark, funereal remixes, and if you aren’t familiar with them, it’s a good introduction. Excellent for designing dark, edgy stuff.
Buy the CD or Download an MP3

1 Comment »