The 10 reasons bad design happens

This list explains where things go wrong and every designer should know them

Two quick notes:

1. I’m doing a live AMA event with the Design Better podcast next Tuesday at 9am PST. You have to be a subscriber to join in, but it’s easy to sign up for a 30 day trial.

2. If you live in Vancouver, BC I’m speaking in person at the VanUE meetup next month. Hope to see you there. Please help spread the word.

---

Always remember that no one plans to make something that’s bad. Most people try to do well. If we’re disappointed by something it might have just been the best they could do at the time.

I tell you this because finding empathy for other creators is important if you want to be a better designer. It helps you stay curious about what really went wrong so you can learn from it. The lazy, ego-centric alternative is to assume incompetence so grand that the entire organization was populated by idiots. We learn nothing by assuming if they had they only hired us, we could have magically fixed everything.

In 2002 I wrote a list of the top reasons bad design happens on projects. It’s a question I studied for years. When I made this list my role at Microsoft was to help teams across the company to overcome these challenges. It’s now 20 years later and I went back to reread the essay.

Here’s my revised and simplified list:

1. General incompetence. Some organizations fail at good design simply because they fail at most things. It’s not just design. To diagnose what’s going on it helps to take design out of the equation: how effective is the organization at doing anything? There are likely bigger dysfunctions at work. Good design is a kind of quality and you need an organization that has the capacity for it. Many do not.

2. Better design isn’t the primary strategy. A more cynical way to explain this is bad design makes money. Consumers often choose products based on affordability and convenience, not quality (e.g. fast food), and corporations usually want to maximize their market share and profits. When you look at the world this way, most organizations are not in the high quality as strategy business. Raising quality can be an effective strategy but it takes more skill to deliver on and not every business model benefits from it. Yet designers often believe everything should be high quality, a belief most people do not share. People can be convinced, but someone has to persuade them.

3. Executives see limited strategy choices. Most tech executives have engineering or business backgrounds and the strategies they rely on come from those perspectives. They may not be willing to take the risks of betting on design as a strategy for the first time in their career. Or they may be stubborn, incompetent or rendered impotent by the company founders or board of directors. This means the ability for a design leader to influence executives may be zero for no fault of the design leader.

4. Project leaders are unfamiliar with creative process. Developing better products requires a different mindset than the typical engineering led build trap (e.g. measuring outputs instead of outcomes). Mature organizations, like the best automobile companies or architecture firms, know how to enable creativity and exploration, which raises quality. It requires a longer term commitment to quality as a strategy and making investments for it. However most tech companies are immature at protecting the creative process (or have no creative process at all) or leaders exclusively reward short-term returns.
   

5. UX leaders fail to influence decisions. Bad design suggests the UX leaders in an organization failed to influence the strategy, which is a primary part of their job. It’s the most senior UX staff in an organization that have the most responsibility for how designers are perceived. If designers are ignored, discounted and underestimated, design leaders are accountable for it. However, many organizations do not even have UX experts on staff to even attempt to gain influence and raise quality.

6. Too many cooks or the wrong ones. When authority over the design of a project is distributed across too many individuals, the likelihood of a quality result decreases. The worst approach is design by committee, where people without shared goals torture each other with compromises, resulting in mediocrity that everyone blames each other for. An organization can have many talented, well-intentioned people, but they are often set up to fail because their goals diverge and authority is too distributed. Even with the right number of cooks, if the primary cook fails to prioritize good design, bad design happens.

7. Confusion over who is the customer. It’s common for people to care about their customers, but this is problematic when they have diverging ideas for who the customers are, what they know and what they need. The fracturing of this perspective translates into UX incoherence, where the product feels like it was designed by dozens of people who didn’t talk to each other. Instead the goal should be for everyone to try to make the product feel like it was designed by one very thoughtful person who knew all the right details to care about.

8. Feature culture instead of scenarios. Product marketing often treats features as binary: either a product has it or it doesn’t. What makes a good design is that features are designed for the real world scenarios customers have in their lives.
   During the development process, it can be easy to focus on features and checklists, since it’s the lowest bar required to claim a feature is in a product. A scenario culture aims to solve real customer problems, not just ship features. Thoughtless, self-serving use-cases (e.g. “The goal of this feature is for it to exist”) are a clear sign of feature culture.

9. Good design is not an actionable project goal. If good design and ease of use is never a stated project goal it will lose resources to other goals. Good design requires tradeoffs and without giving everyone on the team the tools to defend good design, it will always lose to the more familiar pressures of schedule, budget and marketing constraints. Even simple concepts like the iron triangle make it easier to see the design costs of certain decisions.

10. Customer data is misunderstood. Each kind of customer data, such as focus groups, surveys, or usability studies, are only good at answering certain questions. Just like you wouldn’t use a hammer to pound in a screw, you shouldn’t use focus groups to tell you how easy to use your product is (that question is best answered by a usability study). It’s common for immature teams to depend solely on the first kind of customer data they have, and overstate the questions it can answer. Customer data is vulnerable to confirmation bias, we’re prone to seeing what makes us feel good and ignoring what doesn’t. Unless someone with the proper skill and experience is involved, customer data can do more harm than good.

You can read the original post, and all 14 reasons I listed back then, for some advice on how to overcome these challenges.