What Numbers Can—And Can't—Tell Us About Gender In Tech

When the entrepreneur Gené Teare started regularly attending tech conferences in Silicon Valley in 1997, the female business leaders she met mostly filled public relations and marketing roles. In 2008, when she began working at CrunchBase, which collects data on startups and investments, Teare noticed a new trend at conferences: more women filling more executive and technical roles. But among all the data it had, CrunchBase had nothing to say about gender.

Diversity advocates frequently cite data to draw a dismal picture of exclusivity in the tech startup world. But in an analysis of gender on Crunchbase that began last year, Teare, who is now the company's Director of Content, began to uncover a more positive portrait.

"I didn't know what the data was going to tell us when we first went to look at it," said Teare, who co-founded Cyberia, the world's first cyber cafe, in London in 1994. "But what was very interesting was that anecdotal experience was mirrored in the data."

Last May, as the discussion over gender diversity in tech swirled—prompted in part by a discrimination lawsuit filed by investor Ellen Pao—Crunchbase released its first report on gender in tech. It contained a promising finding: between 2009 and 2014 the percentage of funded startups with at least one female founder increased from 9 percent to 18 percent.

Female-founded Investment Firms Invest More In Female Entrepreneurs

The gradual improvement in funding for female-founded companies "opened a number of related questions," Teare and Ned Desmond, TechCrunch's chief operating officer, write in a report published this week. "For example, are some venture firms more likely to fund female entrepreneurs than others?"

In the new report, "Women in Venture," Crunchbase counted the number of female partners at the top 100 venture capital firms and also found a slight increase in the number of venture firms founded by women: in the last three years, 20 VC firms, or 16 percent of newly launched firms, were founded by women.

More female founders might help nudge the culture of the tech industry into a fairer one, comprised of a workforce that more accurately reflects the industry's users. There is strong evidence that this is good for the bottom line too: the presence of female executives is correlated with increased profitability, according to a new study of nearly 22,000 publicly traded companies in 91 countries by the Peterson Institute for International Economics and EY, the audit firm formerly known as Ernst & Young.

Still, they found, almost 60 percent of the companies they surveyed had no female board members, more than 50 percent had no female executives, and just under 5 percent had a female chief executive.

The study's authors could not determine if firms with a female partner were more likely to invest in female-founded companies than those without female investors. But female founded firms stand out in the study: they were both more likely to hire other women as partners as well as to fund companies that were also founded by women.

"There is clear evidence," the authors write, "that the small number of venture firms with female founders and/or an unusually high percentage of female partners, invest at elevated levels in female entrepreneurs."

Investment Firms Founded By Women On The Rise

CrunchBase calls its analysis the most comprehensive study on women in venture capital ever conducted. At the outset, the company ranked global venture capital investing firms by fund size and the number of funding rounds led, and isolated the top 100. It scanned those firms' people profiles for employees listed as "partner" and independently verified that they were active investors.

It found that 7 percent of investing partners at the top venture capital firms are women. Two firms had four female investors each: Qiming Venture Partners in Shanghai and NEA in Silicon Valley. (One of NEA's female investors is based in India.)

Two of the top venture capital firms had four female investing partners, more than any other firm in CrunchBase.

CrunchBase then surveyed newer venture capital firms and accelerators that have only been active since 2014 and found that 12 percent of the partner roles were women. "Data in CrunchBase suggests that more women are entering the venture industry and potentially moving into position to win partner roles," Teare and Desmond wrote, noting that venture firms add new partners very infrequently.

It also found that the ratio of female-founded venture capital firms jumped to 16 percent in the past three years, compared to 12 percent in the period between 2010 and 2015. Teare and Desmond believe this shows creating and participating in female-founded firms could be a "potentially more rapid path for women to become partners."

Startups with at least one female founder received $35.1 billion in venture funding between 2010 and 2015. These female founders received 10 percent of all venture capital globally, and won funding from 12 percent of all funding rounds that took place during that period. The authors did not go further to reveal a trend over how money much female founders received compared to previous years. Rather, they present them as benchmark figures for future studies.

CrunchBase also determined that female-led venture capital firms or firms that had an exceptionally high number of female investors were more likely to invest in female founders. To do this, it isolated the firms that had the highest number of funding rounds for female-founded startups and looked at their investors' gender ratios.

Women won 12 percent of funding rounds between 2010 and 2015, globally. Graphic: Crunchbase

The authors found a curious data point, however: Four of the five highest in terms of percentage of total rounds in startups with a female founder—SOSV, VTF Capital and Metamorphic Ventures—have no female partners, though the top firm, MassChallenge, at 42 percent of its portfolio, has a female managing director in Mexico. This suggests, says the report, "that female founders may not have to worry much about the arrival of more female investors in order to make more progress."

On the other hand, when counting by round, the five most active seed investors in female entrepreneurs—500 Startups, Wayra, Startup Chile, Y Combinator and TechStars—each have at least one female investing partner.

More female and non-white investors, and more female-founded companies, could be effective at encouraging a broader change in culture in the industry in ways that typical recruitment efforts cannot, advocates argue.

"Getting more women to work in tech isn't going to happen by hard-sell recruitment," Minnie Ingersoll, co-founder and chief operating officer of Shift, wrote in a piece this month on Quartz that examined Silicon Valley's "culture problem." "Rather, tech firms need to create and foster environments that women actually want to be a part of—ones that value different viewpoints and work styles."

In an op-ed in October, "Bros Funding Bros: What's Wrong With Venture Capital," Chamath Palihapitiya, the outspoken founder of investment firm Social + Capital, called for a "wake up call on Sand Hill Road."

"We need to recapture our potential and open the doors," he wrote on The Information, a website focused on the tech industry. "Invite more people into the decision making: young people, Blacks, Latinos, females, LGBT and others who aren't necessarily part of the obvious majority. Surround ourselves with a more diverse set of experiences and maybe we will prioritize a more diverse set of things. Maybe we will find more courage to do the hard things."

While the data indicates positive trends, there may be reason for concern. Amid a slowdown in investing—venture capital and otherwise—investors have lowered their valuations on companies that had previously been valued at billions of dollars. No tech company went public in the first three months of this year.

Still, said Teare, "The numbers that I'm seeing don't seem to reflect that female founders are impacted by the slowdown. The slowdown hasn't been as dramatic in the first quarter in Series A and seed rounds, but there was a little bit more of a slowdown in B rounds." But, she added, "there is the question for me of how much of a slowdown there is."

The Promise of CrunchBase

Advocates for gender equality insist more data about Silicon Valley is crucial. Tech publications repeatedly reuse the same 2014 Babson College study to play up how much men dominate the tech industry. The study surveyed 6,793 companies, of which 985 had female executives. It also found that the ratio of female venture capital investors had decreased, from 10 percent in 1999 to six percent in 2013. Google finally divulged its own data in May 2014 after it and several other major technology companies including Apple, Oracle and Yahoo resisted publicly reporting it under federal law for many years, on the grounds that doing so would amount to divulging trade secrets.

Crunchbase may be specially positioned to understand how the gender imbalance in tech impacts innovation. Because anyone can view and contribute to this data on CrunchBase's website—names of executives, funding amounts, and investor information—it has become a powerful tool for transparency in an investing community known for secrecy. Last year it added a gender option to its personal profiles, and to conduct an analysis of the industry's gender balance, began tagging user data with genders where possible.

Anyone can contribute data by creating a CrunchBase profile. In 2015, over 100,000 people contributed data this way, Teare said. Users can also modify most company or profile pages without a moderator's approval. Through its Venture Program, CrunchBase partners with 2,300 venture capital firms, incubators, and accelerators around the world who volunteer up-to-date information about its investments. In September, CrunchBase spun out of AOL's TechCrunch and became a venture-backed startup itself. (Among the company's 27-member team, 10 are women, according to Teare; the 11-person engineering team includes two women.)

In practice, the data has remained free with some limitations. Teare promised that CrunchBase data would remain free, even if CrunchBase, the startup, now needs to worry about generating revenue.

Now anyone can view the database online or download pre-2014 data for free. But people who want to download and analyze it in Excel or need up-to-date data can choose between a $999 and a $4,999 monthly subscription. The latter subscription enables app developers to build new consumer apps on top of CrunchBase data.

In 2013, an app called People+ repackaged CrunchBase's entire database into a mobile app, which prompted CrunchBase to restrict its Creative Commons license to non-commercial applications in December 2013. Since then, it has screened potential commercial app developers and upgraded its pricing model.

We Need More Data

Even with millions of data points on hand, CrunchBase isn't able to tell the whole story around gender inequality in tech. For example, women who enter the tech industry tend to drop out after a few years. Teare wasn't immediately sure how CrunchBase data could add a voice to this issue, but she was interested in using the data to explore it.

CrunchBase's global data could also provide clarity on foreign influences on Silicon Valley and its effect on female investors and entrepreneurs. Between 2010 and 2015, the number of female international students studying computer science in the US increased by 116 percent. Meanwhile, many of the most important tech companies in the US were founded and are led by immigrants, and foreign-born students earn more than half of advanced STEM degrees.

Crunchbase's gender diversity report from August showed that many startup founders in the US are educated in the United Kingdom. But more data about foreign-born or foreign-educated founders could give clarity to issues in the academic pipeline that could affect women and minorities, while spurring a discussion over Silicon Valley's geographical dominance. Teare is pushing her team to analyze CrunchBase data on global and regional levels outside of US-only data, as well as other types of data.

"I think it would be very interesting to look at, outside of gender, racial diversity. The challenge for us, being a global database, is that race means different things for different parts of the world. For us, it's been difficult to figure out how we would introduce that, but it's also a very strong interest," Teare said.

In October 2015, Social + Capital and The Information examined data on 71 firms representing more than $160 billion in assets under management and broke out the racial and gender mix of the investment leadership. Their data on female investors appear to confirm CrunchBase's findings: 8 percent of "senior investment teams" are female.

Drilling down into lower-level investors or non-partners, the survey found that women occupy 20% of "junior investment teams." In terms of race, it found that 20 percent of venture firm's senior teams are Asian, while blacks and Hispanics make up only 1 percent each. In junior teams, black and Hispanic investors made up 2 percent each, and Asian investors made up 32 percent.

Minority women wanting to offer a business idea might be deterred from pitching in the first place, simply because of their physical appearance. A 2010 CB Insights study found that investors simply routinely choose to invest in Caucasian males over other ethnicities or genders.

More Than Data

Meanwhile, some wonder if the tech culture problem might need a less numbers-heavy solution. The effect of placing too much value on data risks trivializing women's personal experiences. If Google one day released tech diversity data that showed gender parity in its recruitment, a woman who voices discontent with its culture might be seen as wanting too much.

Meanwhile, setting quantified human resources targets might start to shift the gender ratio at tech companies, but that won't address the underlying problems that make tech culture and executive boardrooms so unwelcoming to women.

"I think the mistreatment that women and entrepreneurs of color experience at the hands of VCs will go away when the entrepreneurs have somebody who respects them, understands them, and looks like them to invest in their companies," Freada Kapor Klein, a partner at venture capital firm Kapor Capital, told Fast Company last year.

For Teare, fixing the problem begins with the principle of "know thyself." "Venture capital firms view and use CrunchBase on a daily basis, so correctly showing what they have been up to and giving people the tools to understand the deals they have been making is extremely important," she said.

The danger of this data, some investors say, is that it could encourage women to limit their funding options, perpetuating a kind of funding segregation. Women might seek out venture capital firms that have historically been friendly to female founders or minority founders, while attempting less to get a meeting with another firm that would have invested anyway.

"Does it become self-fulfilling? What if you have a VC yet to fund a woman, but because someone sees that stat, women won't pitch him, even though he's completely open-minded and willing?" wrote Charlie O'Donnell, investor at Brooklyn Bridge Ventures, in an email.

Data can also be deceiving. In February, Annalee Newitz at Ars Technica reported on a study that found that, even when controlling for programming language and length of code, Github users were more likely to accept contributions from female users than from male users.At first glance, this surprising finding appeared to be a positive sign that men are becoming more accepting of female coders.

"78.6 percent of women's pull requests were actually accepted and merged into the code, while only 74.4 percent of men's pull requests were," wrote Newitz. "Not only that, but 25 percent of women had almost 100 percent of their pull requests accepted, while only about 13.5 percent of men reached that exalted 100 percent acceptance rate."

When men know the women who are contributing to an open source project, they appear to favor those women's contributions over those of men they know. This finding, she writes, suggests that "women's participation in projects is helping them overcome existing bias."

But the numbers also tell of that bias. When men did not know the contributors to a project—where they were "outsiders"—the women whose contributions were accepted most often were those whose profiles were gender-neutral. Generally speaking, men were less likely to accept contributions from women who did identify their gender. In other words, all things being equal, contributions from unknown women programmers were accepted less often than contributions from unknown male programmers. If anything, the study validated the assumption that men respect female coders' work—at least those they don't know personally—less than that of men.

At first glance, the numbers show that women in open source projects are considered to be more competent than men. But in explaining why that may be, the researchers point to underlying challenges for women.

One explanation is survivorship bias: as women continue their formal and informal education in computer science, the less competent ones may change fields or otherwise drop out. Then, only more competent women remain by the time they begin to contribute to open source. In contrast, less competent men may continue … Another explanation is self-selection bias: the average woman in open source may be better prepared than the average man, which is supported by the finding that women in open source are more likely to hold Master's and PhD degrees. Yet another explanation is that women are held to higher performance standards than men.

Meanwhile, amid an ongoing discussion over unequal pay in the tech industry, the career site Dice released a study in March, titled "Gender Plays No Role in Tech Pay." But the headline may mislead: The report states, "When it comes to bonuses, compensation satisfaction, primary motivators and career concerns, all of which greatly impact overall job satisfaction and career growth, there are clear differences by gender."

Data is useful for making any case. And as more organizations find new ways to quantify diversity trends, there will be more data, and more ways—right and wrong—to support tactics for improving diversity in tech. But data alone won't be enough.

Open commitments to change the tech world's culture, using quantifiable targets, could be the path toward more diversity. Some organizations have created online pledges that companies and conference organizers can choose to sign to show their support for changing the tech gender ratio. Last May, Barnard College's Athena Center for Leadership Studies launched the Athena Pledge campaign to encourage accelerators and incubators to support at least 33 percent women-led businesses by 2017.

In July, Quibb founder Sandi MacPherson started the 50/50 Pledge to encourage tech conference organizers to invite female technologists to speak at their events. Tech conferences are notorious for hosting all-male panels, but no human resource department holds them accountable for creating a diverse lineup. The same holds true for accelerators.

Meanwhile, tech companies and venture capital firms do have the resources to implement and monitor diversity programs to consciously upturn the gender divide. While Teare and Desmond point out in their report that more female founders are receiving venture capital, they do not examine a decade-long decline of women on the investing side. CrunchBase's open data uncovers a few lessons from the greater story of tech's gender divide. To uncover more, tech companies and organizations will need to be more open not just with data, but with their ideas on how to actually change it.

Silicon Divide is a series about gender inequality in tech and science. Follow along here.