People notoriously focus on what is easy to see. Unfortunately, when it comes to hiring this causes significant blind spots that waste time and lead to sub optimal hires.
So many bad decisions in life can be explained by the streetlamp effect (also known as the streetlight effect). The story's origin is lost to time and has evolved with technology, but the core idea is as follows.
Late at night a police officer sees a drunk man stumbling around under a streetlamp looking for his house keys and offers to help search for them. After a while the officer says, “We’ve looked all over, are you sure this is where you lost them?” The drunk replies that he lost them in the park. The officer asks why he is searching here if the keys were lost in the park. The man replies, “I’d never find them in the park, there aren’t any streetlamps there.”
In other words, people typically only choose to look where it’s easy to look, even when it doesn’t make sense to do so.
Metrics can be great, but people often can only get metrics for easy things. If you’re a small business trying to change your brand image you may use social media posts. You can measure the number of posts as well as views, likes, and reshares. What you can’t easily measure is their impact. To understand the brand’s perception, you’d need a more expensive measure tool. NPS (Net Promoter Score) is one, but that involves a large enough survey sample, which a small business may not have access to. Even better would be to measure brand perception. A company like Starbucks will regularly do this, but a small business can’t. Consequently, they stick to measuring what they can and hope it’s a sufficient proxy for what they cannot. Sometimes it works, sometimes it doesn’t.
In education we can easily measure funding per student, student teacher ratios, and test scores. Much harder to measure is how well a teacher engages her students or shows support and encouragement to those struggling. How impactful is that? It’s hard to know because it’s so difficult to measure it.
The measurement problem is also a fundamental problem with interviews (and dating). It’s easy to see if someone has 10 years of experience with X. You can look at the resume and count the years in certain roles or industries. You can also generally assess for knowledge of Y where Y is a specific tool or skill, such as excel macros, Django code development, or Hubspot administration. This is why companies love certifications. If you have a certificate in Y then the company can easily look under the streetlamp and say, “I see the certification in Y!”
There’s nothing wrong with this. However, as many MIT professors like to say, this is necessary, but not sufficient.
Consider for a moment dating apps. Many people have parameters they look for in a partner including age, race, height, religion, and earning potential. All are very easy to spot. Most apps include the first four explicitly in the profile; the last metric is suggested via proxy by listing the person’s job (e.g., consider the earning potential in the US between “schoolteacher” and “equities trader”). You can evaluate someone along these dimensions in mere seconds or even use a filter from the start to only find people who meet your qualifications. Likewise, looks matter to many and again, that can be judged in seconds based on profile photos. Gen X grew up learning to wax on, wax off; Millennials to swipe left, swipe right.
But even if you find someone who is the right age, race, height, religion, and earning potential, that is not sufficient to find a match. There’s a whole big fuzzy parameter called personality which isn’t so easy to access (nor is a likely correlative parameter called chemistry). Profile descriptions will give some indicators. They often let you know if the person is serious, affectionate, spontaneous, confident, or any of hundreds of other traits that may or may not be a fit for you. However, a paragraph or two of prose isn’t sufficient; it may help you rule people out, but it can’t tell you for certain if there’s a match. When dating, we initially filter on what is easy and then use the dates themselves to access the parts that aren’t so easy to filter, those that aren’t readily visible under the streetlamp. When dating it often takes months of extended time together.
Unfortunately, when it comes to dating, people lie. They lie about age, height, income, even marital status, so we have to check for that. Still, most of the dating process is about filing in the blanks not under the streetlamp.
When it comes to interviewing, we also know people lie (or at least stretch the truth). We don’t just accept “10 years of X” and “knowledge of Y” but need to confirm it. It’s easy to evaluate if someone really has that experience and knowledge by digging into their more detailed answers. The problem is we’re still looking primarily under the streetlamp to confirm what’s relatively easy to measure. We don’t go into the park, and unlike dating, we have much less time to evaluate someone.
What is harder is what’s in the park. It’s those other skills, the job interview equivalent of personality (with chemistry perhaps being the equivalent of cultural fit). As you move up the ladder into more senior roles, even as an individual contributor, skills like communication (everything from expression your vision, to how you email, to one-on-one conversations), leadership, team building, negotiating (as a reminder most of the negotiations we do are with co-workers, as opposed to with counterparties outside the company), and other soft skills start to significantly impact your ability to be successful in a role. It also includes specific activities and attributes like management style, how you motivate people, working cadence, values, empathy, integrity, EQ, ability to delegate effectively, creativity, resilience, adaptability, conflict resolution, etc. Additionally, the subtle differences in company culture means someone who can be successful in one organization may be less so elsewhere.
But most companies don’t even know how to define these roles, let alone how to access them, or, in the words of the analogy, where to even begin looking, let alone how to find them without a streetlamp present. Look at the list above. How do you measure someone’s communication? How about their creativity, or ability to motivate people? If you can’t define it and then have a way to evaluate it, your keys will be forever lost while you happily stare at the ground under the streetlamp. Personality assessments like DISC or HBDI can help, but it’s not a full picture.
When swiping on an app, we make snap decisions. In some cases, it’s fine. A Muslim guy may not be a fit for a woman whose profile reads, “Jesus is a big part of my life.” As a 6’1” guy I generally don’t feel compatible with 5’0” women. But sometimes we swipe incorrectly on the margins. And even once we swipe right, trying to assess the fuzzier parts is really hard. Just ask dating coaches, marriage counselors, and anyone who ever realized their partner wasn’t exactly what they thought s/he would be when they got married.
Interviewing suffers from the same problem. We review a resume for a handful of seconds. People with deficiencies in multiple areas can be skipped quickly. But suppose someone meets five out six requirements and on the sixth only has seven years instead of the requested ten; but what if they excel in all those fuzzy areas we don’t instantly see on the resume? It’s easy to reject too quickly because we can’t easily assess if the extra makes up for what’s missing.,
Even when a company brings in a number of candidates who all do meet the requirements (or makes the call to bring in a marginal candidate), they often spend their time under the streetlamp. Like the drunk man, it’s easy to search there. By not looking at the harder to evaluate attributes, they can miss flags on candidates they should skip, and, more commonly, miss out on candidates who would be a better fit under the ideal rubric (if only they knew how to measure against it).
What can a company do? First, start to better define what matters and make sure the entire hiring team is on the same page about what you’re looking for. See the article “Stop Asking for Strong Leadership Skills in Your Job Posts” for some guidance on how to do that. Second, determine how to evaluate it. It may not all be quantitative, it can even be subjective, but it's better than not measuring at all, or doing it unconsciously. Continue to look under the streetlamp, but don’t only look there. (I go into more detail on this in Chapter 3 Interviewing of The Career Toolkit: Essential Skills for Success That No One Taught You.)
Artificial intelligence, particularly large language models (LLMs), may be helpful here. Humans are abysmal at executing what I outlined above. LLMs probably are limited in what they can do, but they can help guide humans in doing this. We will need to address bias issues (see the article “Redlining in the Twenty-First Century: Everything Everywhere All at Once”), but I believe this can be overcome.
Whether or not you use AI to search beyond the streetlamp, you need to start looking there. Otherwise, it's not just the police officer wasting his time, it's your entire hiring team.
It’s critical to learn about corporate culture before you accept a job offer but it can be awkward to raise such questions. Learn what to ask and how to ask it to avoid landing yourself in a bad situation.
Investing just a few hours per year will help you focus and advance in your career.
Groups with a high barrier to entry and high trust are often the most valuable groups to join.
