Looking For A Career Mentor You Love? Let Cold Data Be Your Guide

The startup Everwise conducts a more personal search using analytics, and counts Walmart and Oracle among its clients.

Looking For A Career Mentor You Love? Let Cold Data Be Your Guide
[Image: Flickr user Matt Biddulph]

When Amy Dobler sought to boost her career at Jive Software, the HR manager didn’t learn from a case study or a certification class. Instead, she went online, where predictive and recommendation software matched her with a complete stranger outside her company to be her mentor for the next six months.


That’s right, the age-old process of mentorship just got Netflixed.

Some of the country’s biggest corporations, hoping to boost employee career development, are tapping the startup Everwise, which employs the same kind of prediction and recommendation tech as Spotify, Amazon, and Netflix and the same kind of matching software used by dating sites like eHarmony. And they do it with an added human touch–staff “concierges” who guide the relationships. HR executives are starting to see the benefits of automating what’s normally a labor-intensive job: running a mentorship program and introducing employees to mentors with a perspective that’s outside company walls.

“People didn’t necessarily have access to someone who could be their ideal mentor,” says Everwise cofounder and CEO Mike Bergelson. “We wanted to reinvent mentoring.”

The company is a two-year-old San Francisco startup created by Yahoo chairman Maynard Webb and several former Cisco Systems and Sun Microsoft executives. Backed by $2.5 million from investors, the 23-employee startup has a list of 50 enterprise clients–including Walmart, Genentech, and Oracle–some of which pay $125 per employee per month to manage the mentorship relationships. “It’s the future of HR to draw on this kind of data analytics,” says John Boudreau, a management professor at the University of Southern California who is familiar with Everwise.

Dobler admits she was initially cautious of the idea of computerized matching for mentorship. Yet she gave Everwise a test-drive by completing an online questionnaire, which personalized its queries based on her LinkedIn profile, education, how she moved through her career, her goals and personality. After a phone interview with relationship manager Angela Miller, the software then matched her with Edel Keville, a human resources VP at denim giant Levi Strauss & Co.

The women had fairly similar personalities–quiet, reserved, and thoughtful–and both had followed similar paths in human resources and tech. A few weeks later, the Everwise manager organized a conference call, making an introduction. The two women hit it off. “I see a lot of myself in her,” says Keville.


Over the next several months, Keville offered the support and outside guidance Dobler needed to confidently present to senior executives, to lead international training sessions, and to delegate tasks to managers. “I wasn’t sure what to expect,” says Dobler, “but it has really paid off.”

That kind of feedback is just what the Everwise cofounders hoped for when they launched. They knew that success stories like Bill Clinton, Warren Buffett, and Richard Branson all credited their mentors for charting their career courses, yet mentoring appeared to be a tradition that was dying out. Or, in any event, was due for–that’s right–disruption.

To make the best matches with mentors outside a protege’s own company, Everwise focused its analytics on what made relationships work–drawing on historical data licensed from another mentoring company that looks at 60,000 mentor-protege relationships over 20 years. Engineers pinpointed which characteristics make a good match, such as the best age differences, whether people need to be in the same job function, and how much chattiness or affect mattered. As Everwise began serving proteges and volunteer mentors, the software continued to learn from their feedback to tweak algorithms.

Everwise can now predict which goals a protege might want to work on, based on others like them. Those goals might be, say, improving executive presence or building a personal brand. The software even makes recommendations on “missions” that might help with those goals–whether it’s watching a particular TED talk or reading a certain Harvard Business Review article–all based on feedback from what other mentors suggested. One woman who struggled with commanding a roomful of people, for instance, tried improv comedy because the software identified suggestions by another mentor whose protege had a similar problem.

Despite the data crunching, Everwise executives realized early on that relationships needed a human touch to feel out whether personalities will mesh and to make introductions, identify goals, and keep the relationship on track. “We learned that people don’t open up to systems, they open up to people,” Bergelson says.

Relying on algorithms to determine whether a relationship will be a success presents many challenges, says Harry Reis, a psychology professor at the University of Rochester. He published a paper last year calling into question the algorithms of eHarmony. The same skepticism could be applied to Everwise, Reis says. “The biggest factor in whether a relationship succeeds or not might be called chemistry–how well two people connect with each other,” he says, “and no one has yet come close to providing credible scientific evidence that this can be quantified with an algorithm.”


It’s this reason that Everwise now employs five relationship managers or “concierges” who each work with hundreds of proteges and mentors. They talk by phone with each one for 15 to 45 minutes before the initial introduction to tap into how people are feeling and expectations–as well as uncover commonalities, such as a shared love for Siberian huskies. “They know a lot about each of you, so it’s like a great cocktail introduction,” says Bergelson.

They can then guide the relationship with check-ins via emails and phone calls, offering recommendations with the help of predictive software. “It’s very light touch,” says Keville, who donated a couple hours a month to work with Dobler because she also benefited from mentoring over the years.

Keville says most internal mentoring programs seem to fail after a few months because no one person is focused on keeping the program and relationships on track. “Everwise does that,” she says.

A number of competitors–Insala, Triple Creek, Mentor Resources, and Chronus, among them–offer mentor matching software. Yet all of them draw mentors and proteges from within the same organization. There’s good reason, says Beth N. Carvin, CEO of MentorScout: many HR managers worry that by cross-company mentoring can lead to good people being recruited away or inadvertently leaking company secrets. “It can be risky,” she says.

That has not happened yet with Keville and Dobler. But while the online matchmaking has so far worked out great for the two women, both Keville and Dobler admit that success is only as good as the two people involved.

“It’s really depends on how willing the protege is to own it and guide it,” Dobler says. “But any caution I had flew right out the window upon that first conversation. My relationship was a total surprise.”