In the early 1990s, newly minted Internet evangelists promised a gender-free utopia. Hierarchical identifiers like race and class would be obscured online, they argued, and biased judgements would therefore become obsolete. That didn’t quite work out.
Gender norms infiltrate digital communication today as powerfully (and as detrimentally to women) as they do in-person, show decades of linguistic analysis. Whether on listservs, text messages, Facebook, or Reddit, men tend to “digitally manspread,” as SusanHerring, a leading linguist studying digital communication dynamics, calls it. Meanwhile, women self-segregate in private, women’s-only spaces, like direct messages.
Slack, the increasingly popular workplace communication platform, is not exempt from this phenomenon. As I wrote in “Your company’s Slack is probably sexist,” women across industries say that their male colleagues dominate public-channel conversations with the same authoritative communication styles they deploy in meetings. Meanwhile, women are more likely to use supportive, friendly punctuation, and to modify their opinions with hedges like “I could be wrong, but…”
Now, Slack CEO Stewart Butterfield says Slack is pioneering products that will provide individual Slack users with data on whether their digital communication changes when they speak with people of different demographics. He says this data will help promote more equal, inclusive workplace cultures, and make employees more efficient and effective.
Bias on Slack
As Slack continues to replace email, becoming the primary means of internal communication at over 50,000 companies worldwide, women’s inhibitions on the platform pose a formidable threat to organizational culture, innovation, and business success.
Of course, gendered or otherwise hierarchical communication norms aren’t universal; some women are comfortable speaking bluntly, especially when other women do the same. And some men self-question to the point of Slack paralysis. Nor is Slack (as a company or product) to blame for the prevalence of gender norms that we start internalizing before we can type—or even speak in full sentences. But while Slack holds that no part of its product facilitates bias, the company now appears to acknowledging that women and people of underrepresented minorities could be silenced on Slack—and looking into product development that addresses these trends.
In November 2017, Slack told Quartz that complaints about the platform’s facilitation of gender bias hadn’t come up. “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,” said Julia Blystone, head of communications at Slack. “But we haven’t heard that in our research.”
Just a few months later, CEO Steward Butterfield indicated that Slack was beginning to address concerns, by developing tools to analyze communication trends on its platform, at the Wharton People Analytics Conference in Philadelphia on March 23. In response to a question from Wharton management professor Mae McDonnell on whether Butterfield ever worries that private Slack chat “channels” can reinforce exclusion, CEO also joked, “I worry about everything. I have a Jewish grandmother.”
“If there are deep and systemic problems at an organization Slack can exaggerate them,” he said.
Butterfield added that the platform can also enhance an organization’s positive characteristics. “If there are real positive attributes and successful [negotiating and conversational] skills within an organization, those can be supercharged,” he said. “So I don’t think there’s anything inherent to [Slack’s] structure… or any inherent visible characteristics that would inhibit diversity.”
Butterfield emphasized that for less loquacious employees of all identities, Slack can be a godsend. Nearly every week, Slack hears from customers who identify as introverted, or previously struggled to participate in meetings “where some of the participants are louder, or more aggressive,” or just prefer to think more slowly, says Butterfield. “They reach out to say thank you, because now with Slack, they can participate asynchronously, and they feel like they have much more input, and are much more active participants in their company’s conversations.”
“Personal analytics” could expose communication bias
While apparently gendered or racial slights in face-to-face communication can be distorted by perception and memory, Slack’s digital archives provide invaluable opportunity for linguistic analyses.
Butterfield says he’s “really interested in the idea of personal analytics.”
“These are analytics that no one else has access to you except for you,” he said. “And they don’t present you with any real moral value either way, but [they answer questions like], do you talk to men differently than you talk to women? Do you speak to support groups differently than you speak to superiors? Do you speak in public differently than you speak in private?
Butterfield’s New York staff are creating those analytics tools to identify those personal communication styles, he says. “There’s a handful of APIs Slack employees use to do their own queries,” he said. “Our plan for the next couple of years is to expand that as much as possible—so to provide customers with insights about their organizations and individuals.”
Blystone says the personal analytics initiatives are “in the early stages and will continue to develop over the next couple of years.”
As CEO, Butterfield says he’s interested in using Slack communication analytics at a more macro-level to identify dysfunctional teams or mismatched partnerships within his organization. Slack has publicly committed to diversity within its own ranks, and 2016, has raised representation of women in management from 43% to 48%. Nevertheless, people of color remain vastly under-represented, only 5% of employees in tech roles at Slack are black, a disproportion common in tech companies.
The fine line between analysis and surveillance
Early as these products may be, their potential to put data behind damaging (and positive) communication dynamics on a person-by-person basis is unprecedented.
Women and people of underrepresented minorities sometimes don’t speak up about coworkers whose Slack habits make them uncomfortable due to fear that they wouldn’t be believed, or wouldn’t have data to back up their accusations. Convincing as linguistic studies on gendered communication patterns may be, nationally representative samples pale in comparison to easily accessible, real-time data about the people literally sitting (or Slacking) alongside you.
Of course, privacy as it relates to people analytics remains pressing. It’s an issue Slack has yet to resolve.
“We’re a bit stuck in the middle on these conversations about access to information, because most of our large corporate customers have employee provisions which already grant them the right to access all employee communications,” said Butterfield.
Automatic analysis of how users communicate would be a further step. “It’s a fraught area, because you want people to be empowered by the feedback they’re getting and the tools they’re using, without them feeling like they’re being surveilled,” said Butterfield later.
“That would be useful feedback for any employee, but it’s probably something that people don’t feel very comfortable sharing with their managers or with their peers, so the consent question is really interesting.”
However, even if unfavorable data were to to be exposed about an individual, it can—and, in Butterfield’s books, should—inspire positive change. “So if the result of that [data] is not ‘Hey, it turns out you’re a jerk and we’re firing you,’ but ‘Hey, it turns out we’ve identified some set of problems around communication, or management structure or organizational design, which inhibits the kind of progress we want to make, and therefore we’re going to rectify them,’ that’s a good thing,” he said.
This story is part of How We’ll Win, a project exploring the fight for gender equality at work. Read more stories here.