We often generate and evaluate product ideas based on what we know, and, in consequence, based on our biases. But every design decision we make can raise or lower barriers to participation in society.
If we use our own biases as a starting point, we end up with products designed for people of a certain age, gender, language and physical ability.
Inclusive design is a method of designing products, services and experiences that draws on the full spectrum of human diversity.
Here are our top 5 suggestions for creating more inclusive products and services:
- Reimagine how inclusion looks like
Assuming that all senses and abilities of your potential users are fully enabled all the time may lead you to ignore a large portion of humanity.
In the past years, disability has evolved to mean not solely a human’s physical impairment.
If up until recently disability was seen as a personal attribute (a restriction to perform an activity, resulting from an impairment), today disability is seen as context dependent.
What does that mean? That disability is a mismatched interaction between a person and society.
Sometimes, disability is temporary — can be a short term injury, or can happen while ordering dinner in a foreign country. Other times, disability is situational- becoming new parents (sometimes you can only use one hand in daily activities, because the other one is busy with holding the child), or being in a loud crowd.
So disability is not only a personal health condition, it can also be very context dependent.
When was the last time you designed a product meant for folks 60 years old or older?
Age diversity is seldom addressed, but presents such huge business potential.
Baby Boomers account for half of consumer spending worldwide and hold majority control of disposable income in western countries.
2. Solve to one, extend to many
According to Microsoft Inclusive Design, we could use design solutions for people with permanent disabilities as designs that benefit people universally.
A product that works well for someone who’s blind might also work for someone who is driving a car. High contrast screen settings may work well for persons with vision impairments, but also when a person uses a device in bright sunlight.
A solution for a person with an arm impairment could benefit a person with an arm injury, as well as a new parent, who cannot use both hands, since one is holding the baby.
We have to start seeing human diversity as a resource for better designs, and constraints as drivers of design innovation.
3. Don’t create your user personas around demographics
Create your personas without reference to demographics. Demographics can cause assumptions and subconscious stereotyping by team members.
Try using phases of life or life events, instead of age. A person who is 25 with motor difficulty in their fingers has same problem hitting buttons with a 90 year old.
Something else that you can do is to purposely add demographics that are not common, in order to challenge these stereotypical assumptions- e.g. for low digital skills, use someone younger, not the 60+ stereotype. This way you’ll keep the team focused on the thinking, not the demographic.
Empathy requires not a face or demographics, but the inner thinking, the underlying reasoning of the subject.
4. Words matter
Words are incredibly powerful when it comes to creating the best possible user experience.
Whenever we are designing forms, we have to pay attention to the language we use in their description and microcopy.
Offering inclusive gender options (trans, bi-gender, non-binary) and asking what pronoun do your users prefer (her, him, them) is a way to start with inclusivity. Facebook offers 58 options when it comes to gender identity.
But don’t worry, a simple set of gender options makes it easier to pick something. Letting users pick as many labels as they want is also desirable. You can also offer a open-ended field, to allow for self expression, and a “Prefer not to say” option.
Be transparent, explain why exactly you are asking, and how it will benefit your users. Reassure them that your company strives to be inclusive of everyone so they can feel welcome and protected while disclosing their information.
If you need gender information for market segmentation, ask yourself first whether your audience’s attitudes are really different based on gender.
Same with choosing ethnicity. Most commonly used options are: White, Hispanic, Asian, Black/African American. When requesting ethnicity, make sure you ask the question in a mindful way (“To which racial or ethnic group(s) do you most identify?”) and also make sure you give the option for multiple choice, or the option of not picking either of the labels provided.
5. Inclusive AI
Bias in AI will happen unless it’s built from the start with inclusion in mind. The most critical step in creating inclusive AI is to recognise where and how bias permeates the system.
According to Microsoft Inclusive Design, there are multiple kinds of biases, but the most common are probably the dataset bias, the associations bias and the confirmation bias.
- Dataset bias
When the data used to train machine learning models doesn’t represent the diversity of the customer base. Examples are web cameras used to track user movements that only work well for white users, because the initial training data excluded other skin tones.
b. Associations bias
When the data used to train a model reinforces a cultural bias. Examples are language translation tools that make gender assumptions (e.g. pilots are male and flight attendants are female).
c. Confirmation bias
When oversimplified personalisation makes biased assumptions for a group or an individual. Examples are shopping sites which show recommendations for things the customer has already bought.