editione1.0.8
Updated August 24, 2022You’re reading an excerpt of The Holloway Guide to Technical Recruiting and Hiring, a book by Osman (Ozzie) Osman and over 45 other contributors. It is the most authoritative resource on growing software engineering teams effectively, written by and for hiring managers, recruiters, interviewers, and candidates. Purchase the book to support the author and the ad-free Holloway reading experience. You get instant digital access, over 800 links and references, commentary and future updates, and a high-quality PDF download.
In technical hiring, especially at startups, this is a common practice: “I just reached out to all the guys I know and hired who was available.” In fact, your most effective recruiting channel is probably internal referrals—even for later-stage companies—and referrals typically share the demographic characteristics of your existing team. Expanding the pool and being intentional about who you are inviting to interview gives a chance to those who may otherwise have been overlooked. Luckily, there are many ways to diversify your initial candidate pool and expand your pool of qualified candidates.
If your organization already has a university recruiting program, consider expanding your efforts to include schools with more diverse populations than Stanford, MIT, and Harvard, whether that’s HBCUs (historically Black colleges and universities), women’s colleges, or state universities and community colleges. You can choose to recruit from colleges where the student populations are closer to the audience for your product. You can look at the schools that belong to BRAID or other schools that have made effective strides in diversifying their engineering programs.
Partnering with bootcamps that have a commitment to diversity is another great way to source candidates. Recurse Center, C4Q, and Fullstack Academy are just a few examples. Online programs like Massive Open Online Courses (MOOCs) also graduate more diverse classes of engineers than do traditional four-year schools.
One of the big pushes of the past five years of D&I efforts in tech has been the formation of self-identifying talent pools and affinity groups from which companies can hire—Code2040, Hire Tech Ladies, Lesbians Who Tech, to name some. Name any given demographic, and there is a high chance you can find an organization looking to connect you to that group. But these groups are not all the same. It’s wise not to put too much faith in an external organization, especially if you are expecting—and they are promising—an easy solution to such a complex issue.
How you screen candidates at the top of the funnel presents one of the biggest opportunities to improve your hiring process.
Decades of studies with resumes have shown that just changing the name on a resume—from traditionally white-sounding names to traditionally Black-sounding names—results in fewer callbacks, even though the qualifications are exactly the same. According to a two-year academic study from 2016, changing Black-sounding names to white-sounding names and otherwise “whitening” resumes resulted in over twice the number of callbacks to Black applicants, while whitening the resumes of Asian applicants resulted in about twice the number of callbacks.*
controversy Taking a cue from the results of blind auditions in the music world, many organizations have implemented identity scrubbing from their applications (also known as blind reviews), with promising results. The practice of removing names from resumes and coding submissions can have many positive effects. However, with identity removed from the equation, hiring teams default to evaluating based on the majority identity.
Rather than erasing identity, consider acknowledging the social barriers your candidates have had to overcome.*
Consider two runners, one who races against a headwind, while the other runs in calm weather. They both average the same mile time, though the first is actively pushing against battering winds. Who’s better?Gregory Walton, Department of Psychology, Stanford*
Considering social barriers means you may need to reexamine your stance on job hopping, alma maters, nontraditional educations or paths to tech, contributions to open source, GitHub activity, and awarding points for side hustles and passion projects. For many URGs, the lack (or presence) of these things in their resumes are due to the systemic challenges they face in their day-to-day lives.
“This is lowering the bar!” you might say. But evidence indicates that things traditionally lauded in resumes, like educational pedigree, have no bearing on job performance. Additionally, it’s all too common for companies to test URGs in ways that other candidates are not. Interview panels and hiring teams are typically unwilling to take systemic challenge into account when it comes to URG candidates—interviewers often require that they have 110% of qualifications to make it through.
The truth is, however, that hiring teams compromise in hiring all the time anyway. Think about your last desperate hire and all the rules you relaxed when you were trying to find that one available person with that specific skill set that was vital to company success: when you hired that SQL expert, did you compromise on some of the other technical skills you’d otherwise require? When you hired that Android engineer, did you compromise on your “everyone is a fullstack developer” credo? When you hired the security expert, did you compromise on how well you might get along with them?
URGs add unique value, and allowances may be made accordingly, especially when it comes to requirements that are far more difficult for URGs to achieve because of systemic racism, like graduating with a four-year degree from a top school; or due to systemic sexism, like having experience in management at a top company. Given that such requirements do not predict job performance, should they be requirements anyhow? Are there things you consider requirements that are actually better learned on the job? This is not about lowering the bar —it’s about expanding your understanding of what makes a candidate a good fit and raising the bar accordingly.
Careful planning may allow you to hire in cohorts* or batches (“we’re expecting four Data Engineer openings next quarter”). Many talented professionals from nontraditional backgrounds have stories of being given a chance that they feel they would not have otherwise received, had it not been for the opportunities provided by batch-hiring.
One more concrete tactic you can try is the Rooney Rule. It’s an approach borrowed from the NFL, an organization that has faced criticism for lack of diversity in its coaching and senior staff (in marked contrast to its players). There’s some variety in implementation, but the rule requires at least one underrepresented person to be interviewed before a hire can be made.
caution This approach has other effects that may not be immediately obvious: underrepresented candidates might feel like they’re just being asked to interview to check off a box and that you are wasting their time. Untrained interviewers and inexperienced hiring managers may see it as a burden and may even display hostility toward the candidates (“Oh, you’re not a real candidate, I’m being forced to talk to you.”), causing more harm to URGs, rather than the opposite. Therefore, how the Rooney Rule is designed and communicated matters a lot.
important Studies have shown that having only one URG in your interview process gives that person essentially no chance of being hired; and having more than one underrepresented candidate dramatically increases the chances of a URG being hired*—something to keep in mind if you’re thinking of implementing the Rooney Rule in your organization. Be sure to think through the first principles, potential implications, and ways to get the team on board, to make sure your efforts are successful.
As we’ve discussed, URGs leave tech at an alarming rate. Microinequities, stereotype threat, tokenism, and lack of acknowledgement of URG identities lead to career stagnation and dissatisfaction with growth opportunities.* Unless URGs are entering an inclusive environment—one where they’re welcomed and set up for success—they are most likely to leave. In other words, improvements that focus on the top-of-funnel numbers are important, but are only a small part of the overall solution to improve D&I.
Evaluating other people based on the scant data we gather from a few hours of interviewing is a lofty challenge. Recruiting with diversity in mind can help you get better signal from the noise—from interpreting a nontraditional career path, to being able to take into account a candidate’s obstacles that are unfamiliar to you—and account for the conscious and unconscious biases of the interviewers. There are a few strategies that we describe here, but cover in more detail in Part V.
Structured interviewing has several benefits. By emphasizing the training of interviewers and the implementation of rubrics and feedback forms, structured interviewing allows for methodical evaluation for the skills required for the job. These strategies go a long way in minimizing bias and evaluative confusion in the hiring process.