What do snow plowing, car safety, investment management and face-recognition technologies have in common? While it isn’t obvious to most, they all have impacts that are gendered.
Take snow plowing: Most municipalities focus on getting the roads clear after a major snowfall. But when you clear snow from roads before you clear the sidewalks, it turns out that you get many more slip-and-fall accidents — and most of these are women, because they are more likely to be out walking kids to school in the mornings.
What about car safety? Women are 47 per cent more likely to be injured and 17 per cent more likely to die when they get in a car accident because most vehicle crash tests are done with crash test dummies that are male-sized and have male features.
In the realm of investment management, research shows that women are highly likely to leave their investment advisors when their spouses pass away because those advisors had never worked with them effectively. And facial recognition technology is coming under fire for many reasons these days, but one important one is that these tests are much less accurate in recognizing women’s faces, especially women of colour.
The fact is, many policies, products, services and processes are gendered, and those impacts are often intensified when you also consider other intersecting identities such as race, ability, Indigeneity, sexual orientation or socio-economic class. And these dynamics shape the risks, opportunities and impacts of many of your organization’s activities and outcomes. As I will show in this article, insights around gender can uncover hidden possibilities for innovation and value creation.
Putting Gender Insights to Work
Through my research and teaching, I have come to understand that one of the biggest barriers to progress is that people don’t know how to think and analyze with gender. My hypothesis is that if we all learn how to generate intersectional gender-based insights, we will be able to create more inclusive innovation. These innovations will be good for equity in our society and they will be good for policy impact and business performance.
These dynamics shape the risks, opportunities and impacts
of many of your organization’s activities.
Take the example of Ellevest, an app and investment platform designed to support women in building their wealth. It was founded by Sallie Krawcheck, a former senior executive at several Wall Street firms. Her experience in this male-dominated industry led her to understand that investment advice had not been designed with women’s unique lived experiences in mind. The app’s tag line is: “Ellevest was built by women, for women. The financial industry wasn’t.” In fact, recent surveys show that 85 per cent of investment advisors are men.
Most investment platforms are designed with the assumption of a linear earnings trajectory — meaning wages generally grow year after year with experience and seniority — and fail to factor in the different life events that motivate and constrain women’s investment decisions. But when it comes to investing, men and women do have different considerations. Because of the gender pay gap and career breaks to accommodate childcare, women earn less than men. And they also live longer, which means that women retire with less money and live six to eight years longer.
Keep in mind that the gender pay gap and career breaks to accommodate childcare are not caused by biological differences between men and women; they are caused by long-standing ideas about what is appropriate or inappropriate for someone on the basis of their gender. By failing to consider how gender differently impacts women’s career choices, investment platforms do a disservice to female customers, because their product offerings do not factor in issues specific to women’s career trajectories.
Why is analytics important here? Ellevest’s product uses an algorithm that incorporates factors that are specific to women’s lives, such as the gender wage gap, earnings over the entire career, career breaks, and longer lifespans. On the app interface, women (and people of all genders) are invited to input information about their investment goals and financial status, and the algorithm designs a portfolio and investment strategy tailored to their goals. It celebrates successes and helps with adjustments if savings goals aren’t being met. This is a prime case of an entrepreneur taking advantage of the fact that traditional products do not serve women well — and using analytics to innovate.
New Models Are Emerging
While Ellevest was among the first to embrace gender analytics, companies across industries are following suit. In order to distinguish itself in the market, an online grocery delivery company in the U.S. called Boxed took advantage of data about the ‘pink tax’ — whereby women and girls are charged more for products marketed to them, whether it be razors, shampoos, deodorant or shaving cream. In the U.S., many states also charge sales tax for feminine hygiene products, thus increasing their cost even though they are a necessity, not a luxury. So, Boxed developed its Rethink Pink strategy, charging equal prices for equal products no matter whether they were being marketed to men or women. They also reduced the list price of feminine hygiene products in states that taxed them. On top of this product strategy, they have also been testifying in front of legislatures to get ‘tampon taxes’ repealed.
This wasn’t just a strategy to do good. According to Nitasha Mehta, director of vendor marketing, “We’ve had tens of thousands of customers sign up for Boxed in response to our Rethink Pink initiative, and we’ve seen greater loyalty from these customers as a result. After launch, our customer service team received hundreds of positive emails from current as well as new customers applauding our initiative.” This is important because in online businesses, customer retention is one of the most important metrics associated with long-term profitability.
Real estate development is another male-dominated field: Only about one-quarter of architects in Canada are women, and the numbers are even lower amongst developers, city planners and construction firms. In Toronto, an all-women team of developers has redesigned what housing should look like. The condominium development team spearheaded by Taya Cook at Urban Capital came up with ideas based on intensive engagement with the community, using surveys and community meetings to understand people’s needs. They focused particularly on women’s voices — which tend not to be heard in traditional planning processes — and ended up with a design that included stroller parking on each floor, mini-marts on the premises so people could pick up staples such as diapers and milk, ample bicycle storage, and hallways that had no blind corners so women would feel safe. The condo also included units that could accommodate multiple generations of family members. In short, it was an innovative new approach to housing.
Gender analysis is not just about women or family needs. When we say ‘gender’, we mean all genders. For example, there are lots of insights to be had about men and masculinity that could benefit businesses. You might remember the 2019 ad that Gillette launched as part of its The Best Men Can Be initiative. The company’s tagline has for 30 years been, ‘The best a man can get’. But in response to the #MeToo movement and attention to ‘toxic masculinity’, they developed a new campaign that went along with a social responsibility program to donate to community organizations working on helping men be their best selves.
The ad was controversial. Just check out Twitter or all of the media articles on it when it was launched. Some worried that it might alienate Gillette’s core customers, but when you think that Millennials and younger shoppers are being enticed away from old razor brands by Dollar Shave Club and other online upstarts, a marketing strategy like this — which gets them back in the game with younger consumers who have different notions of gender roles — makes a lot of sense.
Avoiding Downside Risks
So far I’ve focused on the innovative upsides that can come from Gender Analytics, but I also want to address how these kinds of insights can help avoid downside risks. Most of these risks come about because of what we call the ‘male default’. That is, we think somehow that a man is a stand in for all people.
Many readers probably saw the story from 2019 that two women astronauts from NASA couldn’t go on a spacewalk together because there weren’t two appropriately sized space suits for them to wear. It would have been too dangerous for a woman astronaut to go out into space with equipment that didn’t fit her. But this type of danger occurs all the time: From settings such as biology labs to construction sites to bus driving, uniforms and equipment are often designed for men’s bodies, thus creating real hazards for women.
I indicated earlier that women are much more likely to be injured or die in a car accident because safety tests have used male-sized crash test dummies. When automotive companies finally realized this, they reduced the size of the dummies but didn’t change the anatomy, so it didn’t help the problem. Government regulators didn’t have policies in place to force companies to fix this. And, the same is true for pharmaceuticals, many of which are not tested on women and the results of testing are not disaggregated by gender. So in many cases, women are taking drugs that won’t actually work for them or might cause different side effects.
Some might hope that machine learning and artificial intelligence will save us from these kinds of errors. But, take the example of Amazon, which created a bot that was meant to do a better job of hiring new recruits than human recruiters would do. The theory was that it would take away human bias. However, they had to shut this down shortly after launch because the bot taught itself to discriminate against women. The data used to train the bot was based on historical hiring, so it learned to filter out résumés that included the word ‘women’, as in ‘women’s chess club captain’ and graduates of some all-women’s colleges.
In the medical field, one major healthcare system used an algorithm called Impact Pro to help determine which patients should receive what level of treatment. A study of its impact showed that it favoured more complex treatments for white patients than sicker Black patients. The problem was that the algorithm predicted Black patients would cost less, which signaled to medical providers that their illnesses must not be that bad. But in fact, Black patients cost less because they don’t use healthcare services as much as white people on average due to lack of access and a general mistrust in a system which has not served them well in the past. The research showed that if this bias were eliminated, it would triple the number of Black patients receiving the additional help of more complex treatments.
When we say ‘gender’, we mean all genders.
Similarly, when Apple launched its new credit card, it turns out that it was giving women lower credit limits than men. This was true even for couples that shared finances. The defense that was offered by Apple and Goldman Sachs, the issuing bank, was that the algorithm didn’t even use gender as a factor, so it must be gender blind! But in fact, that was just the problem: Because so many other factors used in credit calculations — such as prior credit history, where you shop, and were you live — have been shaped by biased processes in our society, if you don’t account for gender or race or other intersecting identities, you might just be reinforcing and replicating that bias. Being ‘gender blind’ is not enough. You need to do analysis to be gender aware in order to counteract historical inequities.
These kinds of biases can also be perpetuated in internal organizational practices. Research on downsizing — something that a lot of companies are doing these days — shows that supposedly gender or race-neutral decision criteria can have very gendered and racially-biased outcomes.
For example, in a study reported in Harvard Business Review, researchers found that if you do layoffs based on cutting positions rather than evaluating individuals, there was an immediate decrease in representation in organizations for white women, Black men, Hispanic men and women, and Asian men. When layoffs were based solely on tenure, white women and Asian men were disadvantaged. When layoffs were based on performance reviews, white men were the only ones to benefit — and the researchers explained that this is because performance evaluations tend to be over-inflated for white men relative to other people in organizations. So, supposedly gender-neutral criteria can actually undermine any work towards greater diversity that organizations have done. An intersectional gender analysis done in advance of deciding on furloughs would help avoid these negative consequences.
The Concept of Gender Identity
Up until now, we have been talking about gender as if it is a binary — either/or — system. But this is just a product of our culture. In Western societies, we are finally learning something long known in other societies: that a lot of people don’t neatly into that binary or identify with the gender they were assigned at birth. Some readers will be quite comfortable with terms such as gender identity, transgender and cisgender, for example. But some might not be, so it’s worth taking a moment to discuss some definitions and bust some myths.
First, ‘gender binary’ describes a social system in which there are only two genders, a woman or a man. These genders are expected to correspond to birth sex: female or male. But we are learning that the gender binary is not representative of many people’s lived experiences.
The term ‘gender identity’ refers instead to each person’s internal and individual experience of their gender. That is, it is a person’s sense of being a woman, a man, both, neither or anywhere along the spectrum. It is important to note that a person’s gender identity may be the same as the sex they were assigned at birth, or it may be different. ‘Gender expression’ is how people outwardly and publicly express their gender. This could include clothing, hair, makeup, body language and voice. A person’s chosen name and pronoun are also common ways people express their gender. In English, my pronouns are she, her and hers. Someone who identifies as a man might use he, him and his as pronouns. Other people might use they, them and theirs.
A transgender person is someone who does not follow gender stereotypes based on the sex they were assigned at birth. A person whose sex assigned at birth is female and identifies as a man may also identify as a trans man. Similarly, a person whose sex assigned at birth is male and identifies as a woman may also identify as a trans woman. One thing that is very important to know is that ‘trans’ and ‘transgender’ are adjectives, not nouns. So, we would never call someone ‘a trans or ‘a transgender’; it should always be trans man, trans woman or trans person.
People might also be ‘gender non-conforming’, or ‘non-binary’ or ‘gender queer’. These are terms that are becoming more prevalent as we continue to recognize that the gender binary doesn’t represent how many people feel about themselves. Gender non-conforming people might identify and express themselves as ‘feminine men’ or ‘masculine women’ or something outside of the traditional categories of ‘boy and man’ and ‘girl and woman’. A person who is gender non-conforming may or may not identify as a trans person.
In many cultures, gender has historically been viewed as fluid. Samoan culture includes the category of ‘third gender’. People identifying as third gender are born biologically male but embody both masculine and feminine traits. Third gender individuals are considered an important part of Samoan culture. Among Indigenous communities in North America, ‘Two-spirit’ refers to a person who identifies as having both a masculine and a feminine spirit, and is used by some Indigenous people to describe their sexual, gender or spiritual identity.
Cisgender people, on the other hand, are those whose gender identity is in line with or ‘matches’ the sex they were assigned at birth. Some people say ‘cis’ for short. For example, when I was born, they yelled, ‘It’s a girl!’ and I still identify as a woman today. If we are going to label other gender identities, it is important that we also label the more traditional one, and that is why we need to use the term ‘cisgender’ as well. The problem is, the Western context has been built on assumptions that people are cisgender and that there is a gender binary. And as a result, when it comes to analytics, we often don’t have the data we need to understand the people we want to serve.
Think about it. How is data collected? Usually when you apply for a driver’s licence or a job, check in for a flight, or sign up for university classes, you are asked for your gender, and the only options are male and female. So, people who don’t identify as either a man or a woman or who change their gender designation to align with their identity will feel excluded and will not be well-captured by the data. These kinds of data are used for all sorts of analyses conducted by the government and researchers that inform policies and regulations, as well as corporate strategies.
In some places, governments and companies are coming up with better ways of capturing gender. For example, in the State of Washington and several other U.S. states, as well as some provinces in Canada and jurisdictions around the world, people have the option of marking ‘X’ or ‘other’ rather than making a binary choice.
Organizations are beginning to see ways that they can introduce new products and services to include non-binary or trans people. For example, Mastercard launched ‘True Name’, which allows people to get a credit card in their chosen name that matches their gender identity even if they have not gone through the often difficult and costly processes to change their names officially.
Gender Analytics as a methodology recognizes that data collection and analysis on gender — if it only includes the binary definitions — is inadequate to capture the realities of our societies, markets and communities. While, for some, the idea that there is a wide variety of gender identities is new, surveys show that one-third to two-fifths of people understand that there is a spectrum of gender identities; and among young people, it is the majority. Policymakers and corporate leaders will be behind the times if they don’t embrace this reality.
Gender Analytics can be useful both in creating innovation insights that improve impact and in avoiding downside risks, and it should be embedded into everything an organization does. It’s not just an evaluation tool to analyze retrospectively what impact your policy, product, service or process has had. It’s not just a nice-to-have feature you might discuss from time to time. This is something that can shape how you strategize, plan, innovate and serve customers.
Sarah Kaplan is Director of the Institute for Gender and the Economy (GATE); Distinguished Professor of Gender and the Economy; and Professor of Strategic Management at the Rotman School of Management. She is the author of The 360° Corporation: From Stakeholder Trade-Offs to Transformation (Stanford Business Books, 2019). In January, GATE launched a five-course Gender Analytics Specialization on Coursera. For details, visit www.genderanalytics.org.
Share this article: