At which point are ethical guidelines and practices applied in a UX design process?

Question

One area that definitely does not attract enough attention in the topics being discussed in UX design is where the line needs to be drawn when it comes to design practices that influence user behaviour. We have seen many examples where UX design practices and techniques have not been used to serve the interest of users (e.g. in-app purchases disguised as action triggers or call-to-actions designed to entice young users that are easily influenced into making purchase decisions).

In a normal product/service design process, where does ethical guidelines in design come into consideration (if at all)? And should the UX designers have a say in creating and upholding this guideline as part of ensuring that their deliverable meets best UX design practices? I would be interested if there are any examples of this incorporated in a design guideline or framework document.

Answer 1

I am not a UX designer, however, I am a technologist working in ethical technology and I regularly present at conferences regarding these factors. I also am an NIH-funded clinical researcher, and research ethics are central to my daily work. I believe that UX ethics do differ from the ethics of graphical design as well as design research.

First, UX issues differ from graphical design ethics simply because UX is not merely graphical design. This is easy proven when we can and do design devices that have non-visual feedback mechanisms: tactors that vibrate, speakers that beep, etc. Furthermore, graphic design regularly elides the concerns of blind or vision-impaired persons, but ethical technological integration requires that accessibility be treated as a fundamental principle.

UX ethics also differ from design research ethics because research ethics are based upon core principles such as beneficence and equipoise. In product design, many of these factors don't come into play; for instance, we do not need to have a legitimate research question to ethically justify the inclusion of human subjects.

I strongly disagree with @Ades's answer. It is trivially simple to construct user experience that fail ethical tests under the guise of serving the user. I'll provide an example from some of my talks.

Samsung is selling a smart refrigerator. They don't really call it that, though---they call it a "Family Hub." Among the many features of this device is the ability to take a photograph of the interior of the refrigerator every time the doors shut.

These photographic data are made available to the owners via the cloud. This is a great user experience! If you're at the store and you forget whether you need eggs, you can simply use your smartphone to find out instantly without any need for prior preparation (e.g. no list-making). This is a powerful convenience feature designed with users' best interests in mind.

Unless you or a family member have, or have recovered from, an eating disorder. Now, you have a photostream of every time you closed the door; an endless stream of food pictures, a quantifiable reminder of how frequently you hunger or how much you eat.

Or maybe you have Multiple Sclerosis and you take Capaxone, which requires refrigeration. Now, you are entrusting sensitive medical information---the kind of information that if leaked can and has led to people being fired or not-hired---to a provider not governed by medical records laws.

These issues represent ethical concerns with well-established practices in their fields. There are clinical guidelines for treating eating disorders. There are medical records laws. People who work in these fields are bound to adhere to standard ethical praxis, so by default, there are already ethics in place governing these experiences.

The UX designer has a challenging role to be aware of and consider all these concerns. It is no more permissible to tell the user "just don't use that feature" than it is to tell a patient "if you don't want your medical data getting leaked then don't go to the doctor."

Not every feature has to accommodate every possible user. But to the greatest extent possible, the designers must consciously decide which users to leave behind. Those ethics matter, because we live in a world where normal, intended use can cause significant user harm. I don't believe these issues are suitably covered in either the fields of graphical design or design research.

Answer 2

Ethics and UX are on the same side.

It's kind of a non-issue. Because the objective of UX, always serves the interests of the users. The point where UI starts to confuse users and tricks them into pressing something they wouldn't have pressed otherwise, then the UX has failed on that application.

And yes, you should step in. As a UX practitioner you might not have the authority to enforce right UX approaches. But the least you can do is to point out the 'wrongdoings' and suggest the 'best practices'.

Answer 3

I'd say

1. Ethics and UX are orthogonal. You can choose to be ethical or unethical in applying your UX skills.
2. Ethics are domain dependent. Medical systems have different ethical priorities than commerce systems. If you want to be ethical (or unethical) you need to understand the ethics of the domain. I think the domain experts should have the final say on ethics, not the UX experts.
3. Some domains have more clear ethics (e.g medical systems) and some domains have less clear ethics (e.g. social networking systems).
4. The ethics of software systems is constantly in flux. What was ethical 2 years ago might not be ethical today.

Answer 4

I am sure a lot of people around here have read most of the Dark Patterns all around the web. It is hard to say where to draw the line especially when the UX designers have to deal with product managers and sales/marketing people. It is hard because each of position within a product team will have 'equal' weight. So yes, some dark patters will make more money in the short term, and that might be what the company wants/needs for the time being.

In terms of ethics, it will be down to each and every individual to conclude what works and what doesn't. I believe this is part of the skill of being able to do UX work.