Does the modern job interview mean that everyone loses?

How effective are interviews in finding the right people? If you go by what our co-founder and CEO, Sridhar Vembu, believes, not much: "40% to 50% of hires don't actually work out. In an interview process, you really cannot tell culture fit and a lot of things that go into making someone successful at work."

40% to 50% is not a small margin. The conventional wisdom is that interviews help recruit the strongest candidates for a role. But do they really?

Elimination 

It's true to an extent that interviews can help eliminate weaker candidates. If a candidate reveals inadequacies – such as a marketing candidate not knowing basic grammar – it gives the interviewer the ability to filter them out. They can also be helpful in weeding out candidates that display potential red flags during the process.

Duolingo famously had a secret test for interviewees (for top positions) where they assessed if a person was nice or rude to the driver who picked them up for the interview. Founder Luis Von Ahn once rejected a very promising CFO candidate because the driver reported back that they were mean to them.

This is an interesting case where a candidate does not know that the interview process has begun. But 99% of interviews do not work like that. Interviews begin and end in a manner transparent to all parties within well-defined parameters. But this also makes it difficult to get an accurate read on candidates because the interview becomes a performative display. In the interest of securing an attractive position, the focus of the candidates will be on masking their shortcomings. Interviewees might not always fabricate achievements to get them through interviews, but they are likely to exaggerate things to make themselves look good.

The conclusion then is that interviews are not foolproof in eliminating bad candidates.

Identifying top talent 

Consider the famous case of the University of Texas Medical School at Houston from 1979. After the initial intake of students in the university, class sizes were expanded by 50 seats. These seats were filled by students who were initially rejected during the interview rounds. But researchers later found that there was nothing about student performance that differentiated the students who enrolled first from the later entrants. What then was the point of the elaborate interview and screening process?

What does one want an ideal candidate to do?

Perform their duties well, be punctual when it comes to deadlines, be a good culture-fit, be a team player, have the relevant knowledge and expertise for the role, etc.

The trouble is that most of those characteristics are only revealed slowly over time, possibly over the course of the first few months after joining.

What a job interview may be best suited for is to help verify the required knowledge and skill level for the relevant domain. But even here, things do not necessarily become straightforward or less opaque.

Possessing theoretical knowledge and a good awareness of concepts does not equate to being able to apply them in dynamic, ever-changing real-life situations. This difference is even more relevant in a country like India where rote learning is encouraged in educational institutions, meaning that people pass out of schools and colleges with knowledge of concepts but often without an understanding of how to use them.

Interviewer biases 

Is it possible that when interviewers make assumptions about candidates, it reveals more about the interviewers than the candidates themselves? For example, an unwavering smile and effusiveness can cause one interviewer to infer that the candidate is a people person while causing another to think they're a people pleaser, which is not as flattering of an assessment. The takeaways are different according to who is conducting the interview.

Interviewers have a tough time handling or even being aware of cognitive biases such as the confirmation bias, wherein you see what you want to see, and every signal is interpreted the way you want it to be. Did the candidate study at a renowned educational institution? Now everything they say seems more impressive, because confirmation bias is making the interviewer stick to the positive impression their background made on them.

Factors such as physical appearance, gender, social background, etc. can influence a person's assessment whether they want it to or not. Even beyond factors that one might expect to be an influence - such as good looks - and can be prepared for, there are other factors that appeal to interviewers on a subconscious level and are therefore harder to guard against. For instance, a deep voice can evoke impressions of authority and reliability whether or not the candidate possesses those characteristics in reality.

Let's also consider the problem that interviewers subconsciously tend to look for people who are like them even if that is not relevant for the role. This leads to unnecessary homogeneity, which is not great for an organization, given that diverse perspectives create a more resilient and adaptive organization than one with a lot of similar people.

Testing processes 

Okay, so interviews are far from perfect tools when it comes to finding talent. So what is one supposed to do with this knowledge?

Well, just knowing that they're not infallible is a great first step. Too often, interviewers and HR departments are filled with overconfidence in the idea of the interview and their own ability to identify top talent. This is not helpful, and it's good to free oneself of that collective illusion.

The next step might be to re-think the interview process in your organization to eliminate biases as far as possible.

Worldwide, there has been a trend of workplace culture becoming less formal over time, and unstructured interviews can feel as if they better reflect this change in culture. With free-form sessions, interviewers can feel like they are putting candidates at ease without rigid templates while also presenting their organization as a relaxed environment to work. The problem is that while unstructured interviews can sound great on paper, they are rife with biases, far more than other types of interviews. Sure, there may be some individuals blessed with the kind of intuition that can make this approach work but these cases are certainly outliers and the outcomes not replicable.

Meanwhile, structured interviews can minimize biases and noise that may be introduced by meandering discussions into your data. Standardized questions can feel intimidating to some, reminiscent of exams, but at the same time, candidates will also feel like they are undergoing a fairer evaluation.

The best way to evaluate if a candidate is suited for a job is to give them assignments that incorporate tasks from the job. Problem-solving tasks will require them to apply their knowledge and skills in scenarios that resemble the challenges and situations they will encounter every day in the role they're aiming for. At the same time, it can also help candidates understand what is required of them on the job.

Here are some other things interviewers can do to improve the process of hiring:

• Shortening the interview cycle can make the interview process simpler for both parties. Reducing unnecessary rounds of interviews allows for a more efficient evaluation while still ensuring thoroughness.

• Set clear expectations. Interviewers should clearly articulate the role’s responsibilities, performance metrics, and growth opportunities. This helps candidates understand what success looks like and whether it aligns with their career aspirations.

• Share feedback on candidate responses and give them a chance to explain their approach. Even when an interview doesn’t end positively, solicit feedback from the candidate to see if the process can be improved upon.

• Every interaction matters. Interviewers should be mindful of their tone, approachability, and the overall atmosphere they create during interviews. A positive experience can enhance a candidate's perception of the company, regardless of the outcome.

• Creating an environment where candidates feel comfortable asking questions not only helps them understand the role better but also demonstrates that the organization values open communication.

Finding the right fit 

Knowing that the interview is a imperfect tool should prompt organizations to think beyond the conventional notion of what recruitment should look like. And hopefully, to get creative. Some might find issue with Duolingo's 'driver test', but everyone can agree that they were thinking outside the box. With more thought put into them, interviews can be fairer, smarter, and better customized to the domains, teams, and roles that they are designed for. And help the best candidates and the best businesses find each other.