Does your product feel like this?

What bad UX can tell us about building a better future. 


Look at that woman, sitting in the window. She looks dejected, demoralized. “In despair,” as a friend of mine put it. Her emotional state was so apparent I had to stop to snap a photo of it.

She looks the way the Verizon store makes you feel.


Last week I was chatting on the phone with the CEO of a pretty successful early stage startup. We started discussing his product and at some point he lamented that one of the core features of their product makes users feel “fatigued.” I told him that I felt when you start conceptualizing your users’ experiences in the language of fatigue, that is the death knell of your product, and eventually your business.

That may be true for many early-stage startups. But it is not so clearly the case for larger companies.

Why do companies like Verizon, Comcast (“xfinity!”), United, the USPS, with products, services, and brands that feel so terrible continue to exist?

The answer to that question is no doubt complex—“answers,” in plural, is surely more appropriate—but I want to suggest a framework for understanding this problem, and the beginnings of potential solutions to it.


I call it the taxi cab problem: riding in a cab is often an obnoxious experience. And yet, cabs never seem to get any better. Wouldn’t you think someone would eventually make a cab that didn’t suck?

You probably guessed that I chose that moniker because someone did—Uber—and I think it gives us big hints at the conditions that create the problems, and many of their possible solutions.

Here are the conditions that make things ripe for a taxi cab problem to arise (some of these are obvious but I think they all bear mentioning):

  • Reduce competition by raising barriers to entry: think medallions. It’s expensive to start driving a cab in most places—too expensive, in fact, for a driver to absorb the risk of starting to drive on their own. Regulatory and legal complexity also contribute. Think starting an ISP.
  • Incentivize mediocrity: cab passengers have no say in who their cab driver is, and cab drivers have no penalty for not picking up passengers; both of these allow or even encourage cab drivers to exhibit behavior that results in a mediocre experience for the passenger, such as having a driver who won’t pick you up because of where you’re going, how you look, or where you happen to be in town.
  • Minimize/eliminate the customer/user’s power in the sales cycle and/or distribution chain. My cofounder Evan and I dealt with this in the point-of-sale space: the people who interact with POS products the most are restaurant employees who typically have no say in the decision about what POS product to purchase.
  • Wield a leverage point over your users/customers: cab passengers often need to get somewhere quickly and have no other choice for transportation of the same speed. Verizon subscribers can’t complete daily tasks without their phones. Most companies that offer products/services with terrible user experience offer a product or service whose full loss would be life-altering to their users.

How does Uber address these conditions in their product? Here’s what I can come up with:

  • Use feedback to create measurable, two-sided accountability: drivers rate passengers; passengers rate drivers.
  • Make feedback an expected part of using the product so that it doesn’t feel like some unwanted burden or a chore you have to do for the company’s benefit. We all know how “lucky” it feels to be randomly chosen to give feedback on at a restaurant when the receipt prints out (for a chance to win!).
  • Keep cycles of accountability as short as possible: what is the minimum accountable interaction? As opportunities for feedback become more scarce, or more difficult, user frustration increases. I would assert that even when the service provided is exactly the same, and even when the experience is positive—for example, in a pleasant cab ride versus a pleasant Uber ride—providing feedback reinforces perceived value of the service to the user, and creates positive brand perception. Being heard feels good.
  • Remove/lower barriers to entry so that robust competition can work its magic to encourage improvement.

This is certainly an incomplete list; I encourage you to add thoughts of your own but for now want to take things in a bit of a new direction.


You could be forgiven for mistaking this piece as an ode to deregulation and a meditation on the perils of government intervention in markets. Indeed, Uber’s brand and behavior as a company has a distinct libertarian flavor—oh, the wonders of the “free market” and 8x “surge pricing.”

As that aside might begin to suggest, I am far from a libertarian; in fact, I am an unabashed big government liberal. But we’ll leave that debate for another essay.

For now, I want to put forth the following theory: what drives conservatism is that same visceral feeling about the government. Conservatives feel just like that woman in the Verizon window.

And—bear with me here—they’re not wrong. Not only are they not wrong, but I believe that understanding that feeling is the key to both creating liberal consensus and improving the way the government functions.

When I signed up for insurance via CoveredCA.com, I needed to talk to someone due to an error. I called the number on the website and a friendly recorded message hung up on me after telling me to seek in person help. I was in Nevada, so that wasn’t possible. So I hopped onto the nice online chat, and waited through a queue of 1471 people for about an hour and a half. Guess what happened when I got to the front of the line? A message told me no operators were available and the chat window was closed. I clicked back to chat and there I was again 1400 and some people deep.

It doesn’t take many experiences like that to start to feel like that woman in the window. And the correct response to that is not to tell people that the way they feel about the government is wrong; it’s to fix in government what made them feel that way.

The same technology and principles of good product design that have enabled drastically better UX for catching a cab should—must—enable drastically better government.

We as liberals should own using bright private sector innovations to fix the terrible way that the government so often can make us feel.

Denying that kind of visceral frustration can only widen the gap between us, which would be especially tragic when the frustration is the one of the best tools we have to show us how we can govern and live better.