Meet the Author

Ever wonder why many large companies have bad websites? The answer isn't: they have bad designers, or low budgets. The real reason is structural. There are cultural, procedural and bureaucratic issues that relegate them to mediocrity. In the parlance of political science, it's a structural issue that requires reform. But, reform requires a commitment most leaders are unable to make. It's much easier to maintain the status quo than to drive forward change.

It's the same reason failing companies languish with bad customer service. Corporate leadership refuses to confront the structural issues that underly their flaws. Instead they make excuses and institute incremental changes that do nothing more than paper over the underlying issue. It's short term fixes at the expense of long term reform.

So what are some of the most common structural issues and how can they be fixed?

1.) Interest Group Politics:

Every large company has numerous business units that compete for limited resources. The core concerns of these business units don't always align with the best interests of the company. Individual goals are often achieved at the detriment of the brand. If you're a manager for a particular business unit, your number one concern is hitting your own numbers. It's not the customer experience. It's keeping your job. This is why so many large companies end up with crowded, unfocused homepages. It's the reason there are 15 banners promoting 15 different business units crammed onto a single page. The corporate leadership conceded so the web team has to make it happen. The worst part is this undermines the customer's experience and sets up the web team for failure. It becomes impossible to design an elegant interface with clear calls to action and focused paths to resolution based on competing corporate priorities.

How to fix it:

Luckily fixing the problem isn't impossible, it merely requires the ability to say "no". The person responsible for the website should be accountable for it's performance as a whole. This person can act as a counterweight to the silo-ed interests of the individual business units. There is only so much real estate on a homepage. If a business unit isn't a top 5 priority then it shouldn't be on the homepage.

2.) Flawed Approval Cycles

Another common issue with enterprise websites is the approval process itself. I've previously addressed the golden rules of giving feedback, but unfortunately these rules aren't often followed by decision makers in the enterprise space. Instead stakeholder feedback tends to focus on exactly the wrong items. Managers look at a review as a chance to put their stamp on the creative. They give opinions, not analysis and the work goes through endless revisions. Often these changes are to it's detriment. Each review/approval is just as likely to make the work worse, as it is to improve it. More approvals don't lead to a better end result. It usually leads to watered down creative that is not only less impactful, but also less effective.

How to fix it:

Limit the number of people involved in the approval process, and limit the types of feedback that will be accepted. Communicate that not all feedback will be acted on. And institute a process for evaluating feedback that weighs the priorities of the brand, with the impact on customer experience. The person responsible for the feedback process needs to be able to differentiate personal preference from actionable feedback.

The Bottom Line

The reason good companies have bad websites is simple. There are structural flaws that make their homepage into a battle of competing interests at the expense of the overall customer experience, and the approval/feedback process is flawed in such a way that feedback becomes more detrimental than helpful. However, fixing these issues requires a commitment from management. It's not enough to just hire good designers, or UX talent. Companies have to reform their structural biases to make success possible. Talent by itself is not enough.

A quick note:

The inspiration for this post was an article in Fast Company on why the American Airlines website is such a disaster.

Your ability to give good feedback can help make or break a project. If you follow these simple rules your projects will go quicker, run smoother and turn out better.

Listen carefully, and ask questions.

The first step to providing good feedback is understanding the rationale behind the decisions that were made. "I don't like the red." isn't good feedback. It's a personal preference disguised as feedback. Focus on giving feedback that is non-opinionated and provides an opportunity for the person that created the work to provide context. A question like, "This color isn't in our brand standards. What's the thinking behind using this particular red?", will help you understand the reason the decisions were made so your input will be more informed and valuable.

Start with the nice.

Try to accentuate the positive in the first part of your communication. If you always go straight into critiques/revisions then the person you're working with will adopt a defensive stance. The result will be a combative conversation in which you'll be attacking and they'll be defending. You're on the same team and should act accordingly.

We end up working with a fair number of B2B clients. It amazes me how often they complain about their services/product being dry and boring. Most of the time they sit down and pound a few meaningless pages of copy, or they work with a mediocre B2B marketing consultant to develop their website content.

The result is almost always the same, boring copy that reads like a laundry list of ultra-generic product benefits. I can recite these by memory: increased ROI, lowered cost, lowered support burden, greater flexibility. Whether they're selling hosting, contract manufacturing, or consulting — the benefits always end up being the same, regardless of what the product actually does.

The reality is that too many B2B companies are scared of actually having a personality. They confuse being interesting with being controversial. They forget that their target audience is made up of ordinary people who desire nothing more than to learn a little bit more about what their company does in as clear and conversational a manner as possible. A website shouldn't try to dazzle customers with bullsh*t and buzzwords.

It's a fair question that deserves to be addressed. How should a company measure the value of design? What is the contribution that design makes to the bottom line?

This is a difficult problem. Sure there are numerous anecdotes that provide evidence that a focus on design leads to success (Apple, Google, Facebook etc). But, anytime you're attempting to isolate and measure a single variable in a complex environment it will be difficult.

A few years ago the British Council of Design commissioned a study that does provide some quantitative evidence. They created a stock index tracking the 61 firms that had won major awards in british design contests. This index fund outperformed its peers by over 200% during the study. This may not prove a causal link, but perhaps it proves at least a correlation between design and performance.

The most seamless user experiences are often the least noticeable. They just work. You google something and it returns the right results. It's a variation of Clarke's 3rd law, instead of "Any sufficiently advanced technology is indistinguishable from magic", it is: "Any sufficiently advanced user experience is invisible to the user."

This is the reason that projects include user personas and audience goals. The core objective of a website should be to allow users to accomplish the tasks they set out to accomplish in the most efficient way possible. Whether people are visiting to purchase, to request more information or to be entertained. A good user experience shouldn't get in the way of the user's tasks.

Now, there are times that you want the experience to be notable. That you want to communicate a little bit extra in your user interactions. In these cases an interface is designed to draw attention to itself. But, at these times the core experience still needs to be as seamless as possible. If you were to strip away the extras (the humorous responses etc) the core experience should be as close to invisible as possible.

So next time your deciding on whether to run unaffiliated network ads on your website, or whether to require a form before downloading a white paper—think to yourself, does this help or hinder my user's experience. Am I interrupting them on their tasks? Is there anyway to make this a more elegant process?