Diagnosing Discrimination in Your Company’s Customer Service
Tami Kim:
We've really tried to expose discrimination occurring at the front lines. There are a lot of biases at play that individuals aren't aware of. So, what can marketing managers do? That is something that we're also hoping to develop a program around where we can actually identify a set of interventions for marketing managers and also frontline workers
Sean Carr:
As the global movement for racial justice and equity marches forward, businesses are looking within to find discrimination in everything from hiring practices to how they serve customers. But eliminating discrimination from customer service has been historically difficult, even for organizations with stellar reputations. Is there a way for companies to better identify their breakdown and eliminate discrimination from the fast food drive through to the hotel lobby?
Tami Kim is a marketing professor at the University of Virginia Darden School of Business and has gone undercover through her research to prove how discrimination is pervasive across the customer service landscape. She's also learned how companies can take a better approach to solving the problem. I'm Sean Carr and welcome to Darden Ideas to Action.
Tami, your research is incredibly relevant right now as our society reckons with racial inequity. So, thank you very much for making the time to speak to us virtually. We are socially distanced, compliant, and safe. I hope you're well and I look forward to our conversation for the Ideas to Action podcast.
Tami Kim:
Yeah, absolutely. Thanks for having me.
Sean Carr:
Today we're going to discuss your relatively recent research on discrimination in customer service and what companies can do to identify these issues. But let's take a moment to talk about the research and your approach, which I found incredibly interesting. Can you tell us a bit about the field experiments that you ran?
Tami Kim:
Customer service is an area where a lot of companies aren't really realizing that discrimination is happening. So, in my research collaboration with a good friend, Alexandra Feldberg, who is the Organizational Leadership Professor at Harvard Business School, we've really tried to expose discrimination occurring at the front lines. We've run a series of studies with hotels in the United States and also libraries, which are public institutions in the United States.
Basically I created these fictitious email accounts where we varied individual's names. So, we varied things like people's gender, people's race, and sometimes people's socioeconomic status with a label of PhD or non PhD. So, for instance, if you're a customer service agent, you receive an email from somebody named Lamar Washington versus somebody named Brad Anderson, you will very likely assume that the former is a black male customer, the latter is a white male customer. And by doing so, we're able to see variations in service quality by checking service worker's response rates, but also we're able to analyze the content of those email responses.
Sean Carr:
And so what did you learn? What did you find?
Tami Kim:
One result that kept popping up across many of our studies is that both black and Asian customers get discriminated against more than white customers. Hotels we asked if they had any restaurant recommendations in their local city, in local area. To libraries we asked if they have any book recommendations for us. And in another set of studies, we asked if we could get a tour of the library, like a brief tour when we were visiting the town.
And essentially across all of these inquiries what we were finding is that people are less likely to respond to both black and Asian inquiries. And more importantly, when we analyze the content of these emails, we saw that quality of those email responses varied as well.
Sean Carr:
How did you strive to measure discrimination? That sounds really tricky.
Tami Kim:
When it comes to email responses, for instance, there are a lot of email etiquette rules that people tend to follow. You'll get things like if I were to email you back, Sean, do I address you by your name? Like saying, "Hi Sean." Or do I close the email with a complimentary close, for instance, by using things like, "Best, Tami, or sincerely, Tami."
You can also go more in depth depending on what your email inquiry is. So, for instance, if you were to take our hotel studies, because we asked for restaurant recommendations, we could also assess the number of restaurants that our hotel representatives gave us. And we actually do see a big variation in the number of restaurants that they recommended depending on who the customer was.
Sean Carr:
So, were you and your co-author surprised by what you found?
Tami Kim:
I think we were definitely surprised by the fact that our Asian customers did so poorly. This is something that we saw consistently across all of our studies. I should also mention that it may be because it's not just about racial inequity, but there may also be something going on with anti-foreign bias. Asian names that we used, they not only saying that they were Asian, but they were also not Americanized names.
So, for instance, if you were to get an email from someone named Mai Chen, that individual feels more foreign than somebody named Juliet Chen, while still holding their Asian constant across both names. What we've done in a followup study is try to vary foreignness across all the different racial categories. So, for our white male individuals, it will be something like Bjorn Anderson versus Brad Anderson.
So, essentially there you see that we've varied the foreignness while holding white and male constant. And we've done that for whites, blacks, and Asians, and essentially seeing that regardless of racial categories, we do see this consistent bias against foreign individuals, which may also speak to the anti-foreign sentiment that is so prevalent in the United States right now.
Sean Carr:
I understand you also did some work looking at gender. Can you say a bit about that?
Tami Kim:
We see the exact same thing happening in the marketplace when you take a look at all of the different gender labeled products. So, we have whiskey for women, we have pink pens for women, we have things like beer for women. All of these things are playing up a stereotype that woman don't like whiskey, don't like beer, they like the color pink. And oftentimes targeting based on customer's identity can be really effective, but in our set of studies what we're finding is that actually that can backfire, especially if you're making customers feel very conscious about their minority identity.
One of my favorite studies that we did was essentially give female participants two different calculators. One was purple and one was blue. One group of female participants we told them, "Okay, you'll run into a bunch of math problems, choose which color calculator you want to use."
So, we gave them and then they chose between blue and purple calculator. To another group of female participants we said, "You're going to do a bunch of math problems. Choose a calculator that you want. Here's a blue calculator. Here's a purple calculator for women." So, all we did was just tack on that label for women to the purple calculator to one set of participants.
And what we were finding is that generally when neither is labeled, female participants tend to prefer purple over blue calculator, but we see that percentage greatly reduced once we tack on that label for a woman. We see this reluctance happening where people all of a sudden are saying, "I'm going to choose blue." So, just by simply tacking on that label, we're seeing the customer behavior changing. This can also backfire in that consumers might choose an option that is objectively inferior.
Sean Carr:
So, you did some research looking at the 2016 election, right? Tell me more about that.
Tami Kim:
So, one of the things that we found was that customers might actually go for an objectively inferior product if they encounter this gender label that evokes a stereotype about women. So, what we did during the election cycle in 2016 was we had a bunch of us go out with different types of campaign materials for Hillary and approach female voters in Massachusetts.
What we did was to basically give one set of voters two options to choose from, a button or a sticker. Button being a higher quality, more expensive product for free. To the other group of participants, we did the same thing except on the button was a gender labeled slogan. So, it said, "Hillary Clinton, a candidate for women." And what we were finding is that when individuals encounter campaign materials, neither which is labeled, they tend to go for not surprisingly the objectively better, higher quality button over a sticker that had a Hillary campaign very neutral slogan on it.
But once we tack on this label for women for the button, we see that actually a lot of people end up going for the sticker. That is, again, objectively worse quality, but you're seeing this avoidance behavior happening such that people are choosing non-gender labeled product.
Sean Carr:
What do you think is going on there?
Tami Kim:
We're thinking about two factors that are really important when it comes to identity marketing, effectively marketing based on customer's identity. One is that you should not evoke a stereotype when you are utilizing someone's identity in marketing message. But also importantly that stereotype shouldn't be about what we call a stigmatized identity.
So, in many of our studies, we actually see that men are totally fine with gender labeled products. Actually, in the one instance, when we labeled purple calculator for men, we see that a greater number of men decide to go for the gender labeled calculator when otherwise they would've never gone for a purple calculator. We're seeing something very different with our female customers where when they encounter a product that evokes a stereotype about them and it's about them being woman, then that becomes a really bad thing. They react very negatively against it. And so we think that it has to do with the fact that being a woman is more stigmatized than being a man. And the stereotype that is in both is associated with that identity.
Sean Carr:
There are some who clamor for the removal of ethnically oriented labels for products within stores and then there are others who are asking for more of a visible presence for products that speak to their ethnic or racial group, right?
Tami Kim:
Yeah, exactly. And I think that's what makes today's marketing manager's job so challenging. There's just so much that they need to consider. And oftentimes we get asked by practitioners, "What can we do?" Oftentimes these scandals, especially in this age of social media where everything is so amplified, a lot of people don't ever intend to offend any certain group, but oftentimes stereotypes work as a heuristic.
There are a lot of biases at play that individuals aren't aware of. So, what can marketing managers do? And I think that is something that we're also hoping to develop a program around where we can actually identify set of interventions for marketing managers and also frontline workers.
Sean Carr:
Are there examples of companies where they've inadvertently stumbled into discriminatory practices? They didn't realize what was happening.
Tami Kim:
Starbucks comes to mind. Two African American patrons were sitting in the store without having ordered anything and turns out they were just waiting for another colleague to come and meet them. And one of the workers basically called the police on them. I should just note that oftentimes frontline workers don't really realize that they're discriminating against certain types of individuals, mainly because a lot of these discriminatory behaviors are implicit. They are oftentimes unintentional.
And that is why it is so important for organizations to actually implement processes in place where they're regularly auditing internally and making sure that they have protocols for their employees to follow to minimize these types of behaviors. One of the things that Starbucks did after that incident happened was that they closed many of their stores to do racial anti bias training.
I think it was a great move, but at the same time we see a lot of companies, a scandal arises and a lot of companies try to do a lot of things, like Band-Aid, where they hired a diversity officer, or they form a committee internally to investigate issues like this without actually really diagnosing the problem. And that is something that we're really hoping to communicate to organizations that they have to diagnose the problem first.
Sean Carr:
So, that brings us next to what should companies do? Are there any practical steps in other words?
Tami Kim:
First and foremost, what we've been telling practitioners is that you have to diagnose a problem first before you start doing this patchwork solution implementations. So, I think there are many different ways that organizations can actually diagnose the problem. One of the things that we've been telling organizations is to just simply go out and talk to people who are involved in the day to day operations.
So, if it's in customer service context, go and talk to customers, or go and talk to frontline workers, seeing what their experiences are like and seeing what their customers are seeing. And also importantly, a wide range of customers so that you're not only getting a snapshot from certain types of customers is really important. Another thing that we've been telling organizations is doing regular audits can be really powerful, because that is one way that you can really systematically assess whether there's something amiss in your operations.
Oftentimes I think when we talk to practitioners, they get really afraid of doing things like all these studies, or randomized controlled trials, because that seems very resource intensive. What we would really love to communicate to them is that it really isn't and it can be really powerful when done right.
Sean Carr:
That's an interesting insight. It's a scientific approach, but one that need not require big resources to achieve.
Tami Kim:
Right, absolutely. And given that we're in the digital age, they already have so much data that they're collecting and they may not even really know that some of the data that they haven't been looking at can actually provide them with a ton of insight. And that is also why just going out to the front lines and talking to people can be helpful, because it can help you develop what the right questions are to ask.
Sean Carr:
Thank you, Tami. This has really been a pleasure and this is really an important topic. I'm so glad we had a chance to explore your research today.
Tami Kim:
Thank you. Thank you for inviting me.
Sean Carr:
I'm Sean Carr and that's it for today's episode of Darden Ideas to Action. Tami Kim is a professor at the University of Virginia Darden School of Business. Her research focuses on the topics of firm transparency, consumer empowerment, and implicit contract.
Join us next time for more research analysis and commentary from the University of Virginia Darden School of Business. You can subscribe to Ideas to Action on Apple Podcasts, Spotify, or Podbean. To read more expert insights on this topic and more, visit ideas.darden.virginia.edu.
Eliminating discrimination from customer service has been historically difficult, even for organizations with stellar service reputations. Is there a way for companies to better identify their breakdowns and eliminate discrimination from the fast food drive through to the hotel lobby? University of Virginia Darden School of Business Professor Tami Kim joins the Batten Institute’s Sean Carr to discuss how her research proved discrimination pervades the customer service landscape. She shares what she’s learned about how companies can identify the issues unique to their service and take a better approach to solving the problems.
Read full transcript below:
Tami Kim:
We've really tried to expose discrimination occurring at the front lines. There are a lot of biases at play that individuals aren't aware of. So, what can marketing managers do? That is something that we're also hoping to develop a program around where we can actually identify a set of interventions for marketing managers and also frontline workers
Sean Carr:
As the global movement for racial justice and equity marches forward, businesses are looking within to find discrimination in everything from hiring practices to how they serve customers. But eliminating discrimination from customer service has been historically difficult, even for organizations with stellar reputations. Is there a way for companies to better identify their breakdown and eliminate discrimination from the fast food drive through to the hotel lobby?
Tami Kim is a marketing professor at the University of Virginia Darden School of Business and has gone undercover through her research to prove how discrimination is pervasive across the customer service landscape. She's also learned how companies can take a better approach to solving the problem. I'm Sean Carr and welcome to Darden Ideas to Action.
Tami, your research is incredibly relevant right now as our society reckons with racial inequity. So, thank you very much for making the time to speak to us virtually. We are socially distanced, compliant, and safe. I hope you're well and I look forward to our conversation for the Ideas to Action podcast.
Tami Kim:
Yeah, absolutely. Thanks for having me.
Sean Carr:
Today we're going to discuss your relatively recent research on discrimination in customer service and what companies can do to identify these issues. But let's take a moment to talk about the research and your approach, which I found incredibly interesting. Can you tell us a bit about the field experiments that you ran?
Tami Kim:
Customer service is an area where a lot of companies aren't really realizing that discrimination is happening. So, in my research collaboration with a good friend, Alexandra Feldberg, who is the Organizational Leadership Professor at Harvard Business School, we've really tried to expose discrimination occurring at the front lines. We've run a series of studies with hotels in the United States and also libraries, which are public institutions in the United States.
Basically I created these fictitious email accounts where we varied individual's names. So, we varied things like people's gender, people's race, and sometimes people's socioeconomic status with a label of PhD or non PhD. So, for instance, if you're a customer service agent, you receive an email from somebody named Lamar Washington versus somebody named Brad Anderson, you will very likely assume that the former is a black male customer, the latter is a white male customer. And by doing so, we're able to see variations in service quality by checking service worker's response rates, but also we're able to analyze the content of those email responses.
Sean Carr:
And so what did you learn? What did you find?
Tami Kim:
One result that kept popping up across many of our studies is that both black and Asian customers get discriminated against more than white customers. Hotels we asked if they had any restaurant recommendations in their local city, in local area. To libraries we asked if they have any book recommendations for us. And in another set of studies, we asked if we could get a tour of the library, like a brief tour when we were visiting the town.
And essentially across all of these inquiries what we were finding is that people are less likely to respond to both black and Asian inquiries. And more importantly, when we analyze the content of these emails, we saw that quality of those email responses varied as well.
Sean Carr:
How did you strive to measure discrimination? That sounds really tricky.
Tami Kim:
When it comes to email responses, for instance, there are a lot of email etiquette rules that people tend to follow. You'll get things like if I were to email you back, Sean, do I address you by your name? Like saying, "Hi Sean." Or do I close the email with a complimentary close, for instance, by using things like, "Best, Tami, or sincerely, Tami."
You can also go more in depth depending on what your email inquiry is. So, for instance, if you were to take our hotel studies, because we asked for restaurant recommendations, we could also assess the number of restaurants that our hotel representatives gave us. And we actually do see a big variation in the number of restaurants that they recommended depending on who the customer was.
Sean Carr:
So, were you and your co-author surprised by what you found?
Tami Kim:
I think we were definitely surprised by the fact that our Asian customers did so poorly. This is something that we saw consistently across all of our studies. I should also mention that it may be because it's not just about racial inequity, but there may also be something going on with anti-foreign bias. Asian names that we used, they not only saying that they were Asian, but they were also not Americanized names.
So, for instance, if you were to get an email from someone named Mai Chen, that individual feels more foreign than somebody named Juliet Chen, while still holding their Asian constant across both names. What we've done in a followup study is try to vary foreignness across all the different racial categories. So, for our white male individuals, it will be something like Bjorn Anderson versus Brad Anderson.
So, essentially there you see that we've varied the foreignness while holding white and male constant. And we've done that for whites, blacks, and Asians, and essentially seeing that regardless of racial categories, we do see this consistent bias against foreign individuals, which may also speak to the anti-foreign sentiment that is so prevalent in the United States right now.
Sean Carr:
I understand you also did some work looking at gender. Can you say a bit about that?
Tami Kim:
We see the exact same thing happening in the marketplace when you take a look at all of the different gender labeled products. So, we have whiskey for women, we have pink pens for women, we have things like beer for women. All of these things are playing up a stereotype that woman don't like whiskey, don't like beer, they like the color pink. And oftentimes targeting based on customer's identity can be really effective, but in our set of studies what we're finding is that actually that can backfire, especially if you're making customers feel very conscious about their minority identity.
One of my favorite studies that we did was essentially give female participants two different calculators. One was purple and one was blue. One group of female participants we told them, "Okay, you'll run into a bunch of math problems, choose which color calculator you want to use."
So, we gave them and then they chose between blue and purple calculator. To another group of female participants we said, "You're going to do a bunch of math problems. Choose a calculator that you want. Here's a blue calculator. Here's a purple calculator for women." So, all we did was just tack on that label for women to the purple calculator to one set of participants.
And what we were finding is that generally when neither is labeled, female participants tend to prefer purple over blue calculator, but we see that percentage greatly reduced once we tack on that label for a woman. We see this reluctance happening where people all of a sudden are saying, "I'm going to choose blue." So, just by simply tacking on that label, we're seeing the customer behavior changing. This can also backfire in that consumers might choose an option that is objectively inferior.
Sean Carr:
So, you did some research looking at the 2016 election, right? Tell me more about that.
Tami Kim:
So, one of the things that we found was that customers might actually go for an objectively inferior product if they encounter this gender label that evokes a stereotype about women. So, what we did during the election cycle in 2016 was we had a bunch of us go out with different types of campaign materials for Hillary and approach female voters in Massachusetts.
What we did was to basically give one set of voters two options to choose from, a button or a sticker. Button being a higher quality, more expensive product for free. To the other group of participants, we did the same thing except on the button was a gender labeled slogan. So, it said, "Hillary Clinton, a candidate for women." And what we were finding is that when individuals encounter campaign materials, neither which is labeled, they tend to go for not surprisingly the objectively better, higher quality button over a sticker that had a Hillary campaign very neutral slogan on it.
But once we tack on this label for women for the button, we see that actually a lot of people end up going for the sticker. That is, again, objectively worse quality, but you're seeing this avoidance behavior happening such that people are choosing non-gender labeled product.
Sean Carr:
What do you think is going on there?
Tami Kim:
We're thinking about two factors that are really important when it comes to identity marketing, effectively marketing based on customer's identity. One is that you should not evoke a stereotype when you are utilizing someone's identity in marketing message. But also importantly that stereotype shouldn't be about what we call a stigmatized identity.
So, in many of our studies, we actually see that men are totally fine with gender labeled products. Actually, in the one instance, when we labeled purple calculator for men, we see that a greater number of men decide to go for the gender labeled calculator when otherwise they would've never gone for a purple calculator. We're seeing something very different with our female customers where when they encounter a product that evokes a stereotype about them and it's about them being woman, then that becomes a really bad thing. They react very negatively against it. And so we think that it has to do with the fact that being a woman is more stigmatized than being a man. And the stereotype that is in both is associated with that identity.
Sean Carr:
There are some who clamor for the removal of ethnically oriented labels for products within stores and then there are others who are asking for more of a visible presence for products that speak to their ethnic or racial group, right?
Tami Kim:
Yeah, exactly. And I think that's what makes today's marketing manager's job so challenging. There's just so much that they need to consider. And oftentimes we get asked by practitioners, "What can we do?" Oftentimes these scandals, especially in this age of social media where everything is so amplified, a lot of people don't ever intend to offend any certain group, but oftentimes stereotypes work as a heuristic.
There are a lot of biases at play that individuals aren't aware of. So, what can marketing managers do? And I think that is something that we're also hoping to develop a program around where we can actually identify set of interventions for marketing managers and also frontline workers.
Sean Carr:
Are there examples of companies where they've inadvertently stumbled into discriminatory practices? They didn't realize what was happening.
Tami Kim:
Starbucks comes to mind. Two African American patrons were sitting in the store without having ordered anything and turns out they were just waiting for another colleague to come and meet them. And one of the workers basically called the police on them. I should just note that oftentimes frontline workers don't really realize that they're discriminating against certain types of individuals, mainly because a lot of these discriminatory behaviors are implicit. They are oftentimes unintentional.
And that is why it is so important for organizations to actually implement processes in place where they're regularly auditing internally and making sure that they have protocols for their employees to follow to minimize these types of behaviors. One of the things that Starbucks did after that incident happened was that they closed many of their stores to do racial anti bias training.
I think it was a great move, but at the same time we see a lot of companies, a scandal arises and a lot of companies try to do a lot of things, like Band-Aid, where they hired a diversity officer, or they form a committee internally to investigate issues like this without actually really diagnosing the problem. And that is something that we're really hoping to communicate to organizations that they have to diagnose the problem first.
Sean Carr:
So, that brings us next to what should companies do? Are there any practical steps in other words?
Tami Kim:
First and foremost, what we've been telling practitioners is that you have to diagnose a problem first before you start doing this patchwork solution implementations. So, I think there are many different ways that organizations can actually diagnose the problem. One of the things that we've been telling organizations is to just simply go out and talk to people who are involved in the day to day operations.
So, if it's in customer service context, go and talk to customers, or go and talk to frontline workers, seeing what their experiences are like and seeing what their customers are seeing. And also importantly, a wide range of customers so that you're not only getting a snapshot from certain types of customers is really important. Another thing that we've been telling organizations is doing regular audits can be really powerful, because that is one way that you can really systematically assess whether there's something amiss in your operations.
Oftentimes I think when we talk to practitioners, they get really afraid of doing things like all these studies, or randomized controlled trials, because that seems very resource intensive. What we would really love to communicate to them is that it really isn't and it can be really powerful when done right.
Sean Carr:
That's an interesting insight. It's a scientific approach, but one that need not require big resources to achieve.
Tami Kim:
Right, absolutely. And given that we're in the digital age, they already have so much data that they're collecting and they may not even really know that some of the data that they haven't been looking at can actually provide them with a ton of insight. And that is also why just going out to the front lines and talking to people can be helpful, because it can help you develop what the right questions are to ask.
Sean Carr:
Thank you, Tami. This has really been a pleasure and this is really an important topic. I'm so glad we had a chance to explore your research today.
Tami Kim:
Thank you. Thank you for inviting me.
Sean Carr:
I'm Sean Carr and that's it for today's episode of Darden Ideas to Action. Tami Kim is a professor at the University of Virginia Darden School of Business. Her research focuses on the topics of firm transparency, consumer empowerment, and implicit contract.
Join us next time for more research analysis and commentary from the University of Virginia Darden School of Business. You can subscribe to Ideas to Action on Apple Podcasts, Spotify, or Podbean. To read more expert insights on this topic and more, visit ideas.darden.virginia.edu.
