Slack is the bane and the lifeblood of my workplace communication.
Here at Quartz, the workplace messaging app is ubiquitous. Our newsroom uses Slack, which recently raised a quarter-billion dollars, to connect to our bureaus around the world and to colleagues one desk over, with news discussions, procedural questions, banter, gossip, and cute dog photos.
As a new reporter at Quartz, it took me a few weeks to adjust. I got used to the always-on aspects of the system, which pings my phone any time I get a message. I eventually stopped making eye contact when teammates messaged me, and learned the myriad meanings of the dancing pineapple emoji. I liked that Slack allowed me access to senior staffers, and colleagues in other cities and countries.
But I also quickly found myself anxious about messaging, especially in public channels (including office-wide channels and a slew of topic- or team-specific ones). I’d type an opinion and then delete it, for fear of sounding stupid. I’d cushion my ideas and links with modifiers like “maybe,” or “could be wrong, but…”. I was, and am, frustrated by my instinct to end every message with an exclamation point or emoji, as many women do. As a naturally frank person whose first job was at Bridgewater Associates—a hedge fund notorious for explicit, often brutal feedback—these tendencies were frustrating and uncharacteristic.
I noticed that some people seemed untroubled by any such self-doubt—the ones who posted blunt statements, or dropped in links with no context. They responded to others’ statements with sharp critiques, “no,” or radio silence. This behavior—standoffish at best, boorish at worst—conveyed power.
Many of these people, I noticed, were men.
Age, experience, and hierarchical position undoubtedly influence digital behavior. Does gender influence our office’s electronic communications? When I began asking my colleagues, nearly every woman said yes. Overwhelmingly, men said no.
Of course, not every woman experiences self-doubt on Slack. And many men doubt themselves, too—on Slack and in real life. That being said, research has repeatedly proven broad patterns in the way women and men communicate—in person, in writing, and digitally. Gendered communication patterns online mimic and sometimes exacerbate gendered communication patterns in real life, which derive from decades of social inequality.
Slack is such a new medium that people haven’t yet rigorously studied how gender plays out in its universe. But these patterns have been studied in listservs, emails, text messages, Twitter, Reddit, and more. Sexism is not, of course, Slack’s fault (or HipChat or Gmail’s).
And it’s not my company’s fault either. Quartz and its parent company, Atlantic Media, are enlightened workplaces. I’ve never felt more respected and supported as a young woman than I do here, in a newsroom where I have always had female supervisors, and where women account for 59% of the staff.
So, why do these patterns persist, even here? I dug into the history and research to find out.
The promise of a gender-free online utopia
In the early 1990s, Susan Herring had established her career studying oral discourse in southern India. As a linguist, she wasn’t particularly interested in English or gender. Then she was added to a listserv—essentially a massive chat room, in which all messages are delivered via email. It was created by the Linguistic Society of America and she, along with more than 1,000 other linguists, was subscribed automatically.
“It was the early ‘90s, and many were claiming that online, gender, and other social differences that might involve hierarchies—like ability, race, education—all of that would be invisible; you wouldn’t be able to tell who was who, or judge anyone based on their identity,” says Herring. “The internet, and discussion groups, were very young.”
Despite the utopian hopes for digital discourse, rancor and gender divisions seemed to thrive in the online discussions she followed. One debate, which went on for three months, especially caught her eye. Linguists from George Lakoff’s tradition were arguing those from Noam Chomsky’s over who laid claim to cognitive linguistics. The conversation should have had broad appeal. Linguistics is a long-polarized discipline, and there were women in both camps.
“However, it was almost entirely men engaging,” says Herring. “The only women participating were the former president of the association, and a high-status textbook author.”
Herring wanted to know why women held back. She had suspicions grounded in her own reasons: “This discussion is a waste of time;” “I’m too busy;” “I don’t want to expose myself to attack.” Trained as a discourse analyst, she sent out a survey.
Both the men and women she surveyed agreed that the debate was contentious, but they reacted to that contentiousness differently. Men would say things like, “Well, it was kind of aggressive, but as long as the slings and arrows weren’t aimed at me, it was fine,” or “This is just the way online conversation goes.” Some men said it was “kind of fun to go at each other’s throats,” or they brushed it off: “This is nothing; you should see the philosophy list.”
Nearly all the women, however, showed an aversion to the tenor of the debate. Common responses included things like: “The contentiousness made me not want to participate in discussion,” or “It made me want to drop off list all together.” Some went so far as, “People who speak like this are not good people,” and “This debate made me want to not be linguist.”
In general, Herring found, the women cared about politeness far more than men did. “What men really value, according to the study, is [not being] censored,” says Herring. “They perceive even politeness norms as a kind of censorship.”
Notably, one African-American man said he appreciated the frankly racist comments on the listserv because they allowed him to see where certain people stood, and then decide whether to argue with them or not. Women didn’t tend to value such extreme candor, and some saw any conflict as undesirable and to be avoided, Herring found.
Gender socialization helps explain these findings. “From age two or three, kids show patterns where little boys are more assertive and girls more indirect,” says Herring.
In one famous study, groups of male and female toddlers played with a toy pickle. Boys fought and shoved until one person won—at which point the fighting ended, and play resumed. Girls, while they too wanted the toy for themselves, employed indirect rhetoric and negotiation to get it, saying things like, “Maybe we should share the pickle? I could use it for a while, then you could use it later.” (I’ll spare you the inevitable pickle metaphor, but this contrast will sound familiar to women in all walks of life.)
“Already as toddlers, the idea that girls should take others’ feelings and desires into consideration before speaking or acting has formed,” says Herring. “And for boys, conflict isn’t just okay, it’s encouraged.” That’s why men are more likely to make declarative statements (using “boosters” like “always,” “definitely,” and “obviously”) to strengthen their assertiveness, while women often use “hedges” (words like “perhaps,” “might,” and “I think”) to soften statements into suggestions.
In subsequent email and listserv studies, Herring also identified gender differences in message structure: Men would commonly link to a previous message, declare that it’s wrong or disagree, tell the group “how it really is,” and then say (or imply) that the conversation should end. Women, inversely, would thank the previous contributor for her insights, suggest something like, “perhaps we could also think of it this way,” then inclusively appeal to others, asking, “What do the rest of you think?”
As countless studies have shown, women use exclamation points, emoticons, and emoji more—and more supportively (think: smileys, laughing smileys, and hearts)—than men. These images are the digital nods and smiles—the emotional labor women have long-internalized.
What’s more, Herring found, men posted messages that were sometimes 20 screens long, never apologizing for consuming others’ time—while women always apologized for long messages. After tracking response times, she also found that when a question was posed that many people knew the answer to, men were more likely to respond right away, whereas women were slower to jump in.
This delayed response can be explained in part by Dale Spender’s theory of man-made language. The Australian feminist argues that language and discourse conventions are created and enforced by men, for men’s advantage; so when women participate in public discourse, it’s almost as if they’re learning or adapting to a foreign language.
For the discourse style to shift toward female, the group has to be over 60% women, and include an active female moderator, Spender found in her studies. “Otherwise,” she says, “digital man-spreading prevails.”
In time, Herring was flooded with requests—predominantly women asking her to look at their listservs, where the same gender patterns were unfolding. She analyzed countless threads. “Time and again, I found that men were positioning themselves rhetorically as lone warriors—like it was one man against everyone else,” she recalls. “Whereas women were aligning themselves with one another, even in cases where they didn’t necessarily agree.”
And, she found, women often just didn’t show up for these online conversations. By the early aughts, fewer than 15% of participants in all public chatrooms were women.
The rise of social media, internet forums, and smartphones inevitably shifted these dynamics. When blogs boomed, women flocked to them. The ability to regulate discourse on blogs—whether by blocking trolls, deleting comments, or making content private—allowed for a “walled garden model” that suits many women better than the harsh, wide-open spaces of the unfettered internet.
The walled-garden effect is also the default on platforms like Facebook, where accounts can be set so that posts are only seen by people you’ve designated as friends. Today, women are as active as men on Facebook (though linguistic styles still diverge), and they dominate highly relational platforms such as Instagram and Pinterest. Still, open-discourse forums remain masculine environments—at Reddit for example, the users are 69% men.
Slack: Where work happens
If you know any story about Stewart Butterfield, Slack’s CEO and co-founder, it’s probably about the time he sent four black female engineers to accept TechCrunch’s 2016 award for fastest-rising startup.
The employees gave a powerful speech, exalting Slack’s diversity and inclusion efforts, and raising visibility for women and people of color in tech. By forgoing the stage himself, Butterfield—who is white—became an icon for inclusive leadership. Since 2016, Slack has raised representation of women in management from 43% to 48%. (As with a lot of tech companies, people of color remain vastly under-represented; at Slack, only 5% of employees in tech roles are black.)
Driven to make people’s working lives simpler and more productive, Slack does extensive research to understand its users’ workplace challenges—including surveys, focus groups, site visits, and user interviews. Complaints about the platform’s facilitation of gender bias haven’t come up, says company spokeswoman Julia Blystone.
“If we had seen a trend where women said they didn’t have a voice on Slack, we would work on how we might address it,” Blystone says, “but we haven’t heard that in our research.”
Slack believes its product helps promote both “an equal playing field,” and accessibility to colleagues, including executives: “It can create more opportunity to participate in conversations—where in real life you have to be invited to the meeting, in Slack you can pop in and out and share your ideas, get information, and not rely on people to invite you,” Blystone says. “We’re building an inclusive product, a product that’s useful for everyone, so it’s important that we take the time to understand the needs of all kinds of knowledge workers—women, men, people of all ages, people from different cultures, those who work for themselves, those who work at a 100,000-person company.”
As a communications platform, Slack, like a lot of social-media networks, can empower those who otherwise struggle to find a voice, adds Blystone, citing this tweet:
Women’s reality on Slack
Alix Devendra knows the value of messaging platforms: She and her startup co-founders met via Twitter DMs. As a lawyer with many interests, she was quickly captivated by Slack. “I’m in nine Slacks at the moment,” she says. “I trimmed that down from 12 or 13.”
Many of the Slack communities Devendra belongs to are large, national, and interest-specific (much like Herring’s listserv). One 1,200-member group focuses on organizational design. Another, “Agile Attorneys,” has 200 members—lawyers who are “a bit more tech savvy,” she says. Both are predominantly used to discuss industry innovations, share articles, or pose questions. Both are about equal in gender breakdown.
“The philosophy of these groups is to share information that’s interesting to the community,” says Devendra, “but many men tend to use the platform for self-promotion.” While some men are explicitly salesy—to the point where some have been kicked out of the group—many more write posts telling others to check out their writing, achievements, or events, says Devendra. Men also tend to dominate public channels, she says, often responding to others’ posts with declarative statements and dropping in links with no context.
Chelsea (who asked that her last name be excluded to protect her job), a user-experience researcher for a company that builds consumer products, is frustrated by the same dynamic: “I always feel like I have to ‘justify’ sharing content by introducing a link or video with a quick summary or excerpted quote before posting.” She says her male colleagues don’t exert the same labor. “They just toss [a link] in because their interest in it was enough to warrant sharing it—they’re assuming you’ll receive their gift with graciousness, then they walk away.”
Devendra, whose startup helps lawyers design their ideal legal practice, was particularly taken aback by an exchange with a participant in one of her Slack groups over an article about the experimental approach of providing a Universal Basic Income (UBI). The fellow group member (a man, and a frequent poster) responded negatively to the idea. Curious to hear his thinking, Devendra posed a question. The man’s identity has been obscured, and replaced with “M” (for man):
Reading this exchange, you might not notice anything wrong. That’s how unconscious biases and microaggressions work: Unless you’re subject to them, they’re invisible.
But they’re there: Note how Devendra hedges her ideas with “I guess” and “perhaps,” and is upfront about all her failings (as women often are), saying, “I have not done much reading in this area.” Meanwhile, M is blunt, shutting down Devendra’s ideas with comments like “I live on the cusp of the future, not in fantasy,” and tangential facts like “Norway has .7% growth btw.”
Alone, Devendra’s hedging and M’s comments don’t seem like much. This is the struggle I faced while reporting this piece, looking at Slack conversations in my own workplace to find the smoking gun. With microaggressions, there rarely is a smoking gun. But over time, these aggregate power displays can wear down women and minorities, leading us to question whether it’s worth sharing our thoughts at all.
“Optimistically, we’ve hoped that digital communication would flatten the playing field between men and women,” says Janel Anderson, a workplace communications expert who studies modern discourse platforms. So far, this has only happened to a limited extent. “Research continues to show that when there is an audience online, like in Slack public channels, men will be more aggressive than women. So it stands to reason that on an incessant platform like Slack, men will be more aggressive than women in advocating for their ideas.”
Many of my female colleagues at Quartz have noted that on Slack, men (and sometimes women) capitalize on the “bold privilege.” Not all, but many, male editors frequently respond to ideas with little more than a “k” or “no,” while female editors are more likely to explain what’s wrong with ideas they reject, and thank reporters for their work.
I spoke with industry colleagues from other digital newsrooms to see if their experiences and observations aligned with the ones I found at Quartz. They did.
Elizabeth Plank, a senior correspondent at Vox Media, says she frequently behaves in stereotypically “feminine” ways on Slack, just as she does in person. “I feel like I need to be nice and warm and inviting, even if I’m having a conversation about an issue I want corrected or taken care of,” she says. “When I have to ask someone to do something I feel like I need to sugarcoat it with my lady language (exclamation points, emojis, gifs or I love yous) so that I don’t come off as shrill, bitchy, or any other stereotype about women. Men can be direct. Women often don’t have that luxury.” And Yahoo reporter Melody Hahm says she feels the need to frequently contribute to her team’s Slack channel—or else it “gets dominated by the ‘boys’ club’ of sorts.”
While incessant communication in digital spaces theoretically permits everyone to avoid interruptions and take as many turns as they like—Slack’s argument for the platform as a democratizing force—the reality isn’t so equitable. Publicly, on average, women don’t take as many turns as men, and their turns are more likely to be offering social support, says Anderson, the workplace communications consultant. “That’s as true online as it is offline. And when women see men taking many turns, being disrespectful, or shutting down a conversation, they’re likely to switch to another channel, where the communication is more collaborative and supportive,” she says.
One of my male colleagues, an editor, is acutely aware of this dynamic. “Men who tend to be explainers find Slack a lovely place to over-engage in that practice,” he says. He also notes that on Slack, it’s easier to instantly say “no,” rather than explore how to get to “yes.” “You lose the nuance of real back and forth that is available when we speak with one another,” he says, “and this quippy debate does feel like more of a male thing, at times almost used as a weapon. It doesn’t usually invite further discussion.”
It doesn’t have to be this way
Slack’s dream of bias-free discourse in the workplace is, for some, a reality.
Steph Korey, CEO and co-founder of the luggage startup Away, says she doesn’t notice any difference in men and women’s public channel or direct-message (DM) communication on Slack. Notably, Korey’s co-founder is a woman, and Away’s present team—which Korey says has ranged between 10 and 60 people in the past year—is two-thirds female.
It’s not just the female majority that ensures Slack civility—or standards most women would consider civil, anyway. Unlike many companies, Away also has an evolving set of Slack communication guidelines. One rule is that very few situations warrant private group conversations (besides schedule coordination).
“If three or more people are having a business discussion, it should happen in a channel where others have the ability to join and where new team members in the future have the ability to see the history of how that decision was made,” says Korey.
It’s not just a women’s problem
While most of the men I queried said they don’t believe gender influences Slack behavior, some actively work to prevent mansplaining, and promote women’s ideas. One man in Devendra’s legal industry Slack makes a point to publicly call out female lawyers’ writing and expertise whenever relevant—a strategy he learned from women. Similarly, a male reporter at Quartz says he always makes an effort to acknowledge his female teammates’ posts in Slack,with an affirmative emoji or written comment.
Some men also shared insecurities about their own public-channel behavior. One Quartz reporter says he feels “that the silence after [I post] a link is people thinking ‘that’s dumb, why did he post it here?'” This anxiety led him to stop posting in some public groups entirely. And one editor, turned off by constant nay-saying in public channels, tells me he almost exclusively DMs people when using Slack: “It’s just more effective time-wise, and it seems people are more engaged one-on-one, where they won’t feel as judged on the quality of their quips,” he says.
The inherently competitive nature of male discourse, combined with emotionally stifling masculinity norms, make it likely that many men struggle silently on Slack. Which is to say: Ignoring gendered communication patterns on Slack, or any digital platform, is a game where women lose the most, but no one truly wins.
How do we fix this?
Ultimately, improving Slack communication comes down to the trendy but important concept of “psychological safety,” according to Caroline Simard, senior director of research at Stanford’s Clayman Institute for Gender Research.
Coined by Harvard Business School professor Amy Edmondson, psychological safety is the ability to take risks and share your thoughts and ideas openly, honestly, and without fear of judgment. Extensive research has proven psychological safety key to employee performance and satisfaction—which correlates with a strong bottom line—in countless professions, from Google engineers to air-traffic controllers.
Improving psychological safety and collaboration on Slack isn’t complicated: First, we must accept that, regardless of our intentions, some colleagues may be offended by our digital communication. Then we must empathize with this reality. At the least, this empathy makes us contemplate others before posting; at best, it drive us to elicit feedback about our communication, and, if necessary, modify it. Without empathy, digital communication becomes hostile, ideas are not shared, and innovation (and its financial pay off) is stifled.
Groups should also align on digital communication norms, which everyone is accountable to uphold, says Simard: “If there’s a norm around sharing airtime, it becomes pretty easy for someone to course-correct by invoking the norm—as opposed to ‘calling out’ the person.”
She also suggests educating employees on implicit gender and racial biases, and the specific dynamics that come from being underrepresented within a group—such as stereotype threat, low sense of belonging, and isolation. (There are plenty of findings that people of color, regardless of their gender, are silenced on digital platforms far more frequently than white people, and especially white men.)
Since women and minority employees are far less likely to be seen as experts and receive credit for their ideas, vouching for their authority and technical competence can help mitigate the negative professional effects of these biases, says Simard.
Simard also suggests inviting input from colleagues, being open about mistakes, and inviting feedback about what is and is not working in their Slack, and offline communication. “Rather than thinking about psychological safety as a ‘woman’s problem,’ think about the behaviors that will create the conditions for contribution and innovation for everyone,” she says.
Ultimately, no rules or regulation will fix gender, or any other bias hindering psychological safety on Slack, or in-person. “Education about bias can be a step to raise awareness, but education by itself will not work without those norms being created, agreed upon, and actively modeled by leaders,” says Simard.
Don’t confuse women’s Slack quietness for weakness
The goal of creating a more inclusive Slack environment is ridden with complex questions. Among them: Are women who avoid the scrum of public chat really at a disadvantage? There are certainly benefits to the more strategic, less public communication women frequently favor.
Women tend to use conversation to maintain and build relationships, while men use it to exchange and display knowledge, says Deborah Tannen, a linguist, and the author of Talking from 9 to 5: Women and Men at Work. When Tannen implanted herself at various organizations, she noticed that men and women both use small talk to build friendships, but the topics differ: Women often discuss their personal lives, thoughts, and relationships, while men rely on playful insults and joking. The two tactics don’t always mix well.
“Women’s friendly, personal communication struck many men as a waste of time,” says Tannen. But what men miss is that such discourse isn’t just friendly—it’s strategic. “It keeps lines of communication open for women so they can call on colleagues for work when they need it,” says Tannen, “as it’s easier to receive help from someone you’re already friendly with.”
Meanwhile, women often took men’s insults and banter literally. “A man’s retorts were seen as evidence that he thought an idea was seriously flawed, or that he doesn’t like the person he’s arguing with,” she says. “But for many men, challenging ideas is sign of respect and playful insults are a sign of friendship.”
Many of the women I spoke with for this article say they frequently DM other women to vent, digitally roll their eyes, and make fun of mansplaining. This phenomenon is beautifully unoriginal: DMs are the digital evolution of the water-cooler knowledge-sharing that women have relied upon since entering the workforce.
And it’s not a terrible idea for men to know that it’s happening. As Quartz At Work’s Lila MacLellan has noted, the mere presence of private backchannels has been demonstrated in team environments to actively reshape the power dynamic between majority opinion holders and those with other ideas.
DMs and private groups—the “walled garden” features of Slack—aren’t just about gossip. They’re about safety and honest expression. When one male colleague told me he doesn’t understand why women often DM him with comments “that could easily be posted in public channels,” I laughed. The pressure that exists in public channels to preface or cushion comments and links disappears in DMs and private groups, many women told me. I, too, become far more direct, and simply myself, over DMs. And without this private release valve, I can confidently say that I, and many of the women I spoke to for this article, would struggle to produce our best work.
All this begs the question: Since the early 1990s, when Herring’s digital discourse work began, has anything really changed?
A small victory
I experienced a small victory recently, in a public channel here at Quartz. It happened when US president Donald Trump embarrassed himself and an Irish journalist, beckoning her over to his desk to “compliment” her smile. A female reporter posted the news in our editorial chat room, and tagged a handful of specific reporters and editors to see if they thought Quartz ought to pursue the story.
“I’d hope we’re beyond having to explain to our readers why they shouldn’t do that,” a male reporter offered.
Clearly his intentions were good. His reply conveyed that he considered Trump’s behavior beyond the pale, and believed our readers would feel similarly. But I didn’t want the conversation to be shut down so quickly.
At first, too nervous to step into the line of public-channel fire, I DM’d my then-editor. She felt as I did.
Three minutes later, another woman stepped in suggesting a story was warranted.
I waded back in. And then, for 38 minutes straight, more women than I’ve ever seen participate at once swarmed the channel with arguments that built upon one another’s points:
Not once does a woman shoot down her colleague’s idea, force her opinion on the group, or ignore the comment that came before. Instead, just as Herring observed on those early listservs, the women, rather than dictating solutions from on high, used hedges to introduce ideas (“maybe it is a story for the women being creeped on, not the men doing the creeping”). We asked questions (“is there a way to have made it clear that it was inappropriate without getting burned”) and invited in colleagues with relevant knowledge (“that sounds like something @aimee would have thoughts about”).
And, perhaps most importantly, we spoke as a collective (“we all know how that would have gone down”) instead of, in Herring’s words, “positioning ourselves rhetorically as lone warriors.” The conversation epitomized the power of women’s communication style.
At the end of the conversation, I got the assignment: an article framing Trump’s boorishness as an opportunity for female solidarity. It was a story I felt plenty qualified to take on.