• By Everett McKay on November 24th, 2013
  • UX design for entrepreneurs

Design Hacking: Great UX without time, money, or design skills

IgnitersOn November 14, I presented Design Hacking: Great UX without time, money, or design skills to the Igniters: Stanford Entrepreneurs and Silicon Valley Founders group at Hacker Dojo in Mountain View, CA. I would like to thank Raj Lal, Cynthia Lee, and Wendy Soon of Vorkspace for inviting me and making this event happen. I had a great time, made some excellent contacts, and received a lot of encouraging feedback.

Igniters2
Design Hacking at Hacker Dojo. Apparently you are supposed to photobomb from the back.

Download the deck for Design Hacking: A Great UX without time, money, or design talent (8 MB)
Plus links to the event, Wendy’s blog notes, and photos.

A tall order—but doable

Obviously this topic is a tall order—is this goal even remotely realistic? I think so, but you have to believe three fundamental ideas:

  • We waste a lot of time and money on UX design now. Non-designers waste an enormous amount of time creating hard-to-use UIs because they don’t know what they are doing. Experienced designers often waste an enormous amount of time because traditional user-centered design processes are overly dependent upon user research, user testing, and lots of iteration.
  • We don’t leverage what we know. Generally, we know a great deal about our target users—we’re just so focused on what we don’t know that we don’t even realize what we do know. There are basic attributes that all users share that we can take advantage of and use to avoid wasting time.
  • The best is the enemy of the good. You need to make solid design decisions quickly and confidently—even if they aren’t ideal. As an entrepreneur, you don’t have luxury of making perfect decisions with the ideal design process. So, let’s define a “great UX” is one that provides value, and is simple, easy to use, and intuitive for its target users. If we nail those, any shortcomings can be addressed later.

What we need is a perspective to help us make better design decisions more quickly and confidently. Not only does such a perspective exist, but as you will soon see it is one that we already know!

Failing fast…or just plain failing

Traditional user-centered methods require user research, requirements gathering, sketching, prototyping, user testing, and lots and lots of iteration. But there are several challenges, especially if you are in a hurry:

  • User research Frankly, the typical result of most user research is to discover that users haven’t a clue what they want. Or worse—users think they know what they want but really don’t. There’s an old saying: listen to your users, but ignore what they say. What this really means: you have to ask the right questions and interpret the answers. Users aren’t designers, so we can’t expect them to design their own UIs. Good user research takes a long time and is hard do to well.
  • Requirements Gathering requirements works well if they really are requirements. Often requirements are arbitrary, ill-conceived stealth UI specs. The resulting UI might not work well, but at least it meets its acceptance criteria.
  • Sketching Sketching is a great technique during ideation to suggest and explore different design directions. What I see most teams do, however, feels more of what I call “sketching a pile of features,” where the focus is on variations of physical placement of features on the page. This approach can work, but great UX requires going beyond features and layout.
  • User testing User testing is the heart of user-centered design, but I have seen a lot of testing of poor designs that just weren’t test-worthy. A common question determined by user testing: Will users figure out this non-standard, hard to find, poorly explained interaction with poor feedback that doesn’t meet their expectations. We should already know the answer. It’s “no.” No testing required. Often, if you have to ask, you already know the answer.
  • Iteration Iteration is great if it leads to rapid progress towards your goals, instead of just flailing around. When playing golf, I never shank the ball into the weeds and think “Yay! I’m iterating!” A lot of initial UI designs are like shank shots, and the subsequent iterations are slightly improved shank shots. Polishing a poor initial UX isn’t going to get us where we need to be. That first shot needs to be good for iteration to be effective.

These are all sound design techniques, but they take a lot of time and great design is hardly guaranteed.

From what I can tell, agile and lean UX techniques don’t help much here either. They still take a lot of time and money—often with mediocre results, especially for larger, more complex projects. The main difference is that you’ll know your UX sucks much, much faster. [Disagree? Please provide counter examples in the comments.]

The traditional approach favored by developers, which often boils down to putting raw the data structures required by the back end on the screen, doesn’t work at all. Find out why in Don’t design like a programmer.

Reality check

Don’t believe me? Let’s take a look at a well-known UI that is currently in the news.

Igniters-aca
How many HealthCare.gov’s Account Setup usability problems should have been obvious? All of them!

Nielsen Norman Group did a guidelines review of HealthCare.gov’s Account Setup step. Here is a sample of the guideline violations they found:

  1. Allow users to see the product/information before registering.
  2. Keep calls to action (CTAs) and other pertinent information above the fold.
  3. Use email address as username.
  4. Simplify password requirements.
  5. Provide specific and actionable error messages.
  6. Display passwords as users type them.
  7. Remove unnecessary steps.

There are no surprises here—the designers should have known that these would be issues. Lots of time and money wasted here. My bet: this UI wasn’t designed as much as it was specified based on arbitrary, ill-conceived requirements. Usability wasn’t much of a concern but acceptance testing was. [Am I wrong? Please let me know in the comments.]

What we (should) already know

To design great UX without money, time, or design skills, we need to leverage what we already know. If you were to ask a group of people why UX design is so hard (which I routinely do), invariably they will agree that one of the biggest challenges is that we don’t understand our users well enough.

This is certainly true. But if this were the root cause of poor design, I would expect to see many UIs that are well designed, just for the wrong target audience. In practice, I never see that. Instead, what I see is UIs that are poorly designed for everyone. It’s doesn’t matter who you are or what you are doing—they’re poorly designed.

Of course, it’s true that different users have different knowledge, motivation, preferences, workflows, etc. But there are many attributes that we can safely assume that all users have that we usually don’t even consider. I have compiled dozens of these and put the top 50 in UI is Communication.

But to keep things simple, let’s start with one user attribute that every UX designer should know:

Unless they have been trained or have prior experience, users only know what your program tells them.

I believe the real root cause of most poor UI is a failure to communicate to its users. Well-designed UIs are self-explanatory, and shouldn’t require trial-and-error, a user’s manual, help, or training to use.
Igniters-rtfm
Most user’s manuals explain how the UI should have been designed in the first place.

UI is Communication

A user interface is essentially a conversation between users and technology to do tasks that achieve users’ goals. A UI is a form of human communication, and if it communicates well it will be naturally intuitive. My UI is Communication book explores UI design from this point of view, and I will explain this concept in detail in a future blog post.

But to get started, a great way to design UI is to think about how you would explain a task to a target user in person. Think about the steps, their order, the language you would use, and the details that what you would bother to explain. Also think about what you wouldn’t say—those unnecessary steps or details you know the target users just don’t care about.

That conversation is a high-level guide to what your design should look like—great UIs feel like a natural, friendly conversation. But to be clear, I’m not saying the UI should be a literal conversation—that might be tedious—rather it should feel like a human conversation in terms of the steps, order, and communication. And it should never be chatty—applying this process usually results in less text, but much better text.

If there is a difference between what we say in person and what we say in UI, usually it is the human conversation that is right and the UI that is wrong. We naturally explain things in person in a way that the target users will understand. Great UIs should mirror human conversations because that is the most effective way to explain things.

At this point, I’m sure that you would like to see some examples. If so, please check out the Look Inside feature on Amazon. There is a very generous preview with many examples to get you going.

Putting it all together

Now let’s cut to the chase! Here is a communication-focused design process that requires a minimum of time, money, and design skills.

Most of this process is fairly traditional. The main difference is that we are using our understanding of the target users and top scenarios to determine how to communicate the tasks effectively on a human level. Scenarios and communication drive the process. At best, traditional design processes discover that communication accidentally. Instead of focusing on a mechanical solution and hoping to stumble across a usable human solution, let’s flip that and start with an intuitive, human solution and make it work mechanically. And save a whole lot of time and money by doing so.

Planning

  • Define your target users (hours)
    • Determine all the facts and assumptions about your target users, and make sure you have the right market.
    • For more information, check Personas: Dead yet?
  • Define your product’s value proposition (hours)
    • Determine what will motivate your target users to buy and use your product.
    • Merely solving a problem is a weak value proposition. Instead, you need a solution that provides real value—one that’s worth the trouble.
  • Determine your top scenarios (hours)
    • A good scenario describes a specific target user with a specific task or goal, in a specific context or environment, but without specifying a solution.
    • More simply: Who, what, when, where, and why…but not how.
    • For more information, check Design scenarios—and how thrilled users ruin them.
  • Reality check with real users (days)
    • Make sure you have these decision making tools right before proceeding.

Design

  • Hire a good professional visual designer—one that you can trust (days)
    • The process outlined here will get us a long way in the interaction design, but we’ll need help with the visual design.
    • Let’s not pretend that non-designers can do the visual design. We can’t! (And trust me, it’s obvious!)
  • For each top scenario, design the human “conversation” (hours)
    • Determine how would you explain the task to the target user in person.
    • Determine a simple, familiar, efficient task flow that mirrors that conversation.
    • Analyze, optimize, simplify to get it right.
  • Use the conversation to design the pages (days)
    • Does the page communicate well and mirror what you would say in person? Does it have integrity—does everything need to be there?
    • For more information, check Effective Prototyping.

Refinement
Some evaluation techniques to consider:

  • Scenario-based reviews (hours)
    • Use the top scenarios to help you evaluate the design from the users’ point of view.
    • Are those top scenarios easy to perform? Are the underlying user goals achieved?
    • The traditional “look at a screen and throw spears at it” approach doesn’t work well.
  • Communication reviews (hours)
    • Have someone explain the design and carefully listen to the explanation. Can you find any differences?
    • If so, explore why. If there are differences, most likely the in-person explanation is right and the UI is wrong.
  • Value proposition reviews (hours)
    • What is going to motivate users to use the product?
    • Are the benefits maximized, the costs minimized? Is its value immediately obvious?
  • Intuitive reviews  (hours)
    • Review the attributes required for an interaction to be intuitive.
    • Any missing attributes will be a problem unless the interaction is advanced, infrequent, or optional.
    • For more information, check Intuitive UI: What the heck is it?
  • Reality check with user reviews, preferably with paper prototypes (days)
    • Perform informal usability studies by giving target users important tasks and seeing if they can perform them successfully.
    • You can use paper prototypes for this. Functional prototypes are usually too expensive to create and result in too much commitment in the current design.
    • For more information, check How do you know when a design is done?

I put a rough time estimate for each of these steps (in parentheses) to back up the “without time” claim. While designing all the pages and doing some user-based reality checking will take days, most of the remaining steps can be done in a matter of hours for simple projects.

UX Design Essentials for Entrepreneurs

For the sake of brevity, I glossed over several important design techniques that I mentioned here, but they are all covered in detail in my UX Design Essentials class. My next UX Design Essentials in San Francisco (San Mateo, actually) will be on April 23 – 25. You can register now without payment.

While I believe that UX Design Essentials is at least 90% relevant to entrepreneurs with extremely limited time, money, and design talent, I’m thinking of creating a new UX Design Essentials for Entrepreneurs class that is 100% relevant, then delivering that version exclusively in the Bay Area, New York, and Boston.

A good idea? Please contact me to let me know.


Leave a Comment / What do you think?

One Response

Links to this article

  • Notes on “Design Hacking” by Everett McKay | walkerux - December 1, 2013 at 8:24 pm
  • […] on the deck from a presentation by Everett McKay to the Stanford Igniters, Nov. 14, 2013. See uxdesignedge.com for his summary and to download the deck. I’m focusing mainly on his discussion of UI design […]


For more information, please contact info@uxdesignedge.com

All Content Copyright © UX Design Edge