User Experience

You walk into your local burger joint and the menu says: “The Ultimate Customizable Burger – $5″, and that’s it. Since it’s the only thing on the menu, you ask for “The Burger” and the person behind the counter responds:

“What kind of meat would you like? We can make it with beef, bison, lamb, chicken, tofu, pork, tempeh, seitan, jackfruit, or plain.”
“Well, what do you think is best?”, you ask in a daze.
“They’re all good. What kind of meat do you prefer?”

This is only the beginning of the ultimately customizable burger requiring you to choose your selection of 50 toppings (which are all given to verbally from the clerk who amazingly has them all memorized), what type of bread you prefer (8 types), what country you want your lettuce from (out of 4). Imagine the answer at each stage to the question, “What do you recommend”?, is always answered with, “It’s your choice, anything you want!”

Feeling overwhelmed yet?

Is having the ability to customize a bad thing? Not necessarily, but forcing your customer to make decisions about how to customize something that is not important to them is a bad thing, but many software packages neglect this principle. People want choice when it’s important, but otherwise, they would rather someone make reasonable default choices for them.

A debate rages between the book Paradox of Choice by Barry Schwartz (Excellent summary talk he gave at TED) and those that favor the book The Long Tail by Chris Anderson.  Even though recently it’s been difficult for behavioral economists to reproduce Schwartz’s studies, I think his main point is very well taken: The sum of all the choices we are required to make actually makes us less satisfied with them.

The Long Tail is all about the value of choice. When something matters, like “I really love Mexican lettuce”, then you’re more likely to be delighted at the ultimate burger joints ability to provide this to you. We absolutely love to customize and develop niche tastes that are all our own. But if I’m a lettuce connoisseur (is there such a thing?), not a mustard connoisseur, I really just want someone to pick a reasonable choice for the mustard.

So I reconcile these two views as follows:

• To the consumer: Satisfice, set high standards, buy the things that meet them and enjoy them rather than worrying about whether there is something better
• To the business: Make your product mass customizable, index comparitively so customers can quickly find what they need, and provide good defaults so that you enable satisficing

[My earlier blog summarizing the concepts in the Paradox of Choice]

All of us have encountered an insane application of technology at some point in our lives causing us to have the “what were they thinking” moment. The amazing thing is that they were thinking, and their thoughts were that this application of technology was perfectly normal. In fact, they usually were thinking it was the best thing ever.

Here are a couple recent example’s I’ve come across:

• 18 Button Mouse
  Seriously. Apple thinks more than one button is too many, but the designers of OpenOffice believe that 18 buttons is really the right number.
• Palm 650 Hard Reset
  Seriously, follow the link and imagine someone (and their friend) reseting this phone with a stylus in their mouth.  There is “making something difficult”, and then there is turning it into a sport.

These are obviously extreme, but exemplify that once you start down a line of thinking, without someone to questioning you, your further assumptions all make complete sense, but lead to complete nonsense.  This is a problem with buying too much into your own vision, you quickly loose focus on the real purpose — creating something that helps your customer achieve THEIR goals.

The only way to know their goals is to talk to them, not to decide for them.  As Steven Blank says, “Inside the building, there are only opinions.  All the facts lie outside of the company”.  This sentiment is something that we should all apply, not only to our product development, but also to the teams that we lead.

[P.S. If you come across other examples of odd technology applications, please send me the link]

Paul Graham, an essayist and also entrepreneurial investor, wrote an article called “Organic Startup Ideas” that indicates two different types of startups: Organic (which build something for themselves and then find others that are interested), and Inorganic (which build something they think someone else would want).

His essay is excellent and essentially states: Building something small for yourself because you desire it, is likely to be more successful than building something for someone else.  He doesn’t name the reason specifically, but it’s clear to me that the problem is empathy.  When you build something for yourself, you know what itch you are trying to scratch.  When you build something for someone else, you are only guessing at their itch.

Empathy is a topic that I have been thinking a lot about lately because it is core to building something that your customers desire. While Paul focuses on Organic companies in his essay with some excellent examples. My guess is that many Inorganic companies start from a group enamored with their technology. As a result, they become a solution looking for a problem.

In the end, we all empathize with ourselves perfectly, but it is not so with others.

Other good reading on this subject:

• What’s Your Problem? Build software for yourself – from 37signals ebook “Getting Real”
• Steve Blank – Why understanding the customer is so critical. Plans are made, but they don’t survive the first customer contact.

I’ve started to loathe renting cars. Not because they aren’t high quality nor because of the ridiculous hoops that agencies make you jump through. Rather because each car was designed to “innovate” on a bunch of things that were already good enough.  Does it really make a difference if the open trunk button is to the left of the steering wheel, near the radio, or on the floor near the seat? Either designers considered the problem and believe they invented the best place for this button or they didn’t give it a second thought.  Either way, the fact that every car puts many controls in different “innovative” places means that every time I get in a car, my head spins (literally) trying to find the one I need. This is when I consider innovation a curse, and this is but a trivial example of something that occurs all the time with startups.

Startups begin with a Big Idea that will make the world a better place by providing their customers with something that they need; however, too often, those who have the vision to start companies, also see just how inadequate or ill-designed many other things are as well.  Fortunately, they have a solution and so they set out innovating not only on the “Big Idea”, but on all those little things that should be better.

The result is that in addition to providing a breakthrough for their customers, they also provide innovations that improve the many things that were already  good enough.  This places a huge burden on their customers because now they have to invest time relearning things that they didn’t think needed any improvement to begin with.  While some customers (especially early customers) will push through this and be delighted at all of the innovations, the remainder will see your product as too complex and cling to their existing well-known solution.

A long time ago, a venture capitalist friend mentioned advice he received when starting his fund: The Three Change Rule. He was told by an astute advisor that his fund can be different in no more than three ways and in everything else needed to be exactly the same as others.  This does two things.  First, it provides a foundation for investors to compare the key areas that you are different (since many things are the same).  Second, it forces you to pick only the most important differences to focus on.

As innovators, we think that because we know a better way, all we need to do is educate our users about why they should change and how to change. In this, we try to solve the problem of unfamiliarity by providing documentation. Clearly, the user will see the incredible value of our solution once they read about it. We assume they are going to be as excited about our innovations as we are.

But they aren’t.  In fact, the necessity of reading documentation in order for them to use your product, will either scare them, or worse, cause them to look elsewhere. Sure, customers are  interested to learn the minimum they need to in order to take advantage of “The Big Idea” but everything else just gets in their way. I’ve begun to view the size of the documentation as a good way of measuring how much you’ve failed to design the product with your customers in mind. Poor design that is documented is better than poor design without documentation to be sure, but the more documentation that is needed, the more the solution will be rejected by the market.

Just because there is room for innovation in a particular area, doesn’t mean that it needs to be improved.  Pick three differentiators, and in everything else, be the same.

If an airport was designed like most software it would look exactly like the Phoenix Suncity Airport (PHX) and, like many software companies, PHX offers consulting and tech support. Fortunately, their consultants are free and take the form of an army of purple-clad senior volunteers. In spite of this, every time I visit this airport, I’m shocked at its poor design.

To illustrate this point let me take you through my first hand experience of this airport. I was flying with my mother-in-law from Guadalajara to Phoenix. She had to catch a connecting flight to Denver and I to San Jose.

After clearing immigration and customs, we proceeded on a long walk to the end of a hallway, took the poorly marked elevator, accepted the free hand sanitizer and were dumped into what we later learned was Terminal 4. Let the games begin.

Upon entering Terminal 4 your first goal is to find your connecting flight so you look at the first set of monitors you see, which we did. Unfortunately, this set of displays contains only the list of international arrivals. Not much help.

We looked across and saw another bank of monitors. We went to look up the flight and saw that the only airline listed was US Airways, our airline. But her flight was not listed. The only clue we had was that the person who checked her bag after customs had suprisingly stated she was on a United flight, which now led us to believe that it was a code share flight. But how were we supposed to find her gate?

At this point, like with so many software packages, the option you’re looking for simply isn’t where you think it should be. So we walked over to the very friendly consultants in purple. Our consultant seemed a little bored, but gave us the helpful hint to take the escalators marked “baggage claim,” down to a bus to get to Terminal 2 (where United flies). This is the software equivalent of,”Oh, its easy, just hold down the shift key while double-clicking with the mouse”.

Having read the slightly more helpful training manual (brochure of the airport), we came to understand that there are four different terminals. Each terminal has airlines assigned to it and while I can’t speak for all of the terminals, Terminal 2 has simply numbered gates. Terminal 4 has 4 wings (A-D) which each have about 30 gates. This doesn’t include the international arrivals which dump you into Terminal 4. Terminal 4 is only US Airways flights.

After talking to our purple-clad consultant the first time, we looked back at the escalator and seeing markings that indicated it was to baggage claim, we asked for confirmation that we really needed to shift-double-click. She reconfirmed, “yup, take the escalator down.” Then she added some new information. “Find door 22 and take the bus from there.” Note, the instructions were not, “follow the signs that say: Connecting to other terminals.” There are none. Having received our quest from the oracle, we proceeded into the lair.

Now that we knew the secret door number we set off in search of the magic bus. Upon exiting the door, we saw a sign labeled “Airport Bus.” Again, no indication what the “Airport Bus” is, or where it goes, we ask the driver if this is the bus that takes us to other terminals. Upon confirmation, we board and take a brief rest for our journey.

After successfully finding the United Airlines counter (very thankful for inadvertant tip from the baggage recheck guy), we arrived at the correct gate, said our farewells and I then embarked on the solo journey back to my own gate.

I eventually made it back to Terminal 4, searched in vain for some indication on how to find my flight but to no avail. By now I had given up on any illusion of self-sufficiency and asked the nearest purple friend who replied, “Oh, you’re on the wrong side of the airport, but you can take the elevators up over to the left.” I looked over, and there were two down escalators. But now armed with the pointer about the elevator, I ended up finding my destination.

Terminal 4, Wing A, Gate 13.

Why do they call this airport “The Friendliest Airport in the US” inspite of being so user unfriendly? Because even the most skilled way-finders have to ask for help from the friendly purple brigade. While it’s hard to know how they could design this airport any worse, but to their credit they do have people wearing bright colors offering some semblance of assistance. What they should probably do is replace their signs with, “Stop trying to figure it out on your own, just ask a purple person.”

Navigating this airport successfully exemplifies most software interactions. If you’re a regular user, like someone who flys through Phoenix often, you learn your way around and it becomes second nature. But if you are an occasional user you’re in for an adventure that will later require some form of nerve tonic. Oddly, you are grateful for the purple people sitting next to you telling you, “Yes, now hold Alt-A, while pressing the enter button. See, it’s exactly what you wanted. Pretty cool, eh?”

It’s more friendly to simply help people achieve their goals in a self-sufficient way by designing things so they are easy to find. For instance, why have Terminals, Wings and Gates? Why not simply have a terminal letter and a gate number? People can deal with numbers over 30. Moreover, why not have all of the flight panels contain all of the flights so that anyone can figure out where they need to go. And why not add signs that indicate exactly where you should go to change terminals?

Too often, when we build our companies or products, we look at them the way I do the Denver airport (or probably how my friends see the Phoneix airport), as something that seems obvious. As such, we don’t consider the pain that new users go through in accomplishing their goals. Empathizing with our customers experience to help them meet their goals leads to having a business that doesn’t have to be staffed by hundreds of friendly consultants.