Once outstanding candidates have been discovered through a robust recruiting process, how do you ensure that the right candidate is ultimately selected?

The standard process that most firms use to assess talent – interviewing each candidate multiple times before the hiring manager makes a final decision – is clearly not working. Much of this is because most people simply aren’t very good at interviewing; they’re untrained, inexperienced, and rely too much on intuition. In fact, there’s ample data showing that most assessment takes place within the first three to five minutes of an interview. The remaining time is spent finding ways to justify and confirm that bias. In other words, most interviewing is a colossal waste of time. Questions such as, “tell me about yourself” or “what is your greatest strength and weakness” or “what is your most proud accomplishment” are essentially worthless.

Instead, organizations should rely on methods that have proven to be impactful. For example, most studies have shown that the best predictor of how someone will perform in a job is a work sample test, not an interview. In fact, work samples have shown to predict future performance with 29% confidence. These tests entail giving candidates a sample piece of work, similar to that which they would do in the job, and assessing their performance at it. Of course, not all roles have nice pieces of work that can be handed to a candidate to complete. But for those that do (engineering, product management, call center, graphic design, etc.), there’s no reason not to leverage this best practice.

The second-best predictors (26% predictive power) of job performance are cognitive ability tests. “In contrast to case interviews or brainteasers, these are actual tests with defined right and wrong answers, similar to what you might find on an IQ test. They are predictive because general cognitive ability includes the capacity to learn, and the combination of raw intelligence and learning ability will make most people successful in jobs.”

Similarly, structured interviews – those where candidates are asked the same series of questions with clear constraints on time and a uniform scoring method – are effective (26% predictive power) because they create a consistent environment where any variation in candidate assessment is a result of the candidate’s performance, not because an interviewer has higher or lower standards, or asks easier or harder questions.

There are two types of structured interviews, behavioral and situational. Behavioral interviews ask candidates to describe prior achievements and match those to what is required in the current job (i.e. “tell me about a time…?”). Situational interviews, on the other hand, present job-related hypothetical situations (i.e. “What would you do if…?) that gauge the veracity and thought process behind the stories told by the candidate.

As you may expect, Google does all of this and much more, including reference checks (7% predictive power), and verifying the number of years of work experience (3% predictive power). As Laszlo Bock, former SVP of People Operations, explains, “the goal of our interview process is to predict how candidates will perform once they join the team. We achieve that goal by doing what the science says: combining behavioral and situational structured interviews with assessments of cognitive ability, conscientiousness, and leadership.”

Thus, without further ado, let’s explore Google’s step-by-step process for interviewing, screening, and selecting:

1. Resume Screened by Recruiter
2. Phone Interview or Google Hangout(video chat) is Conducted by Recruiter – Screened for General Cognitive Ability
3. In-Person Interviews are Conducted by Hiring Manager, Peers, Subordinates, and Cross-Functional Employees
4. Structured Feedback is Compiled and Back-Door ReferencesConducted
5. Hiring Committee Reviews and Recommends to Senior Leader
6. Senior Leader Reviews and Recommends to CEO
7. CEO Makes Hiring Decisions

Although the first few steps are quite standard in most organizations, Steps 3-7 are radically different and are the real secret to Google’s hiring success. Let’s dive deeper and dissect the five major differentiators to their process.

Structured Interviews

To ensure that all interviews are structured the same, Google provides each interviewer with a pre-determined list of questions that are relevant to the role and that assess each key attribute that were highlighted in Step 2. They also limit all interviews to half an hour, which forces productivity and eliminates useless small-talk and meaningless questions.

During the interviews, Google requires all interviewers to include the attribute being assessed, the question asked, the candidate’s answer, and the interviewer’s assessment of that answer. Finally, interviewers rate candidates using the following scale:

1 – we shouldn’t hire this person

2 – I don’t think we should hire, but I could be swayed

3 – this candidate could be good, but I could be swayed

4 – we should hire this person

Although giving a candidate a final score of “3” is fine, Google strongly encourages interviewers to take a stand. They believe that every interview should lead to a final decision. There’s no appetite for being wishy-washy.

Ensure Skilled Interviewers

Yet, to guarantee an effective hiring process, it’s not enough just to ensure that the interviews are structured consistently. Interviewers must also be equipped to perform well and held accountable if/when they don’t. This is so critical to Google, in fact, that their Chairman, Eric Schmidt, once said that effective interviewing is the single most important skill that any business person can develop.

Thus, in an effort to train Googlers in how to interview well, they’ve implemented a program in which an elite team of in-house experts are tasked with teaching their methods to those who want to get better. Participants in the program are given formal didactic training and at least four opportunities to shadow experts as they interview candidates. Employees are then set out on their own and measured against several metrics, including: number of interviews conducted, reliability, promptness and thoroughness of feedback, and effectiveness.

To gauge effectiveness, each interviewer is provided the candidate scores they’ve given in the past along with the history of whether those candidates were hired or not. If interviewers consistently get it wrong than they are asked to stop interviewing all together. If they consistently get it right than they are handsomely rewarded financially.

No More Than Four Interviews 

After just four structured interviews, Google can predict with 86% confidence whether or not they should hire someone. Every additional interview after the fourth only adds 1 percent more predictive power to the decision. With this in mind, they’ve instituted a Rule of Four, which prohibits more than four interviews with any candidate.

This significant change in their process has single-handedly reduced their time-to-fill by half and saved interviewers hundreds of thousands of hours.

Peers, Subordinates, and Cross-Functional Interviewers 

In nearly every organization, candidates interview with the same set of people – hiring managers and a few peers. Google flips this upside down. There, every candidate also interviews with subordinates and cross-functional leaders, those that have little connection to the team for which the candidate is interviewing.

The thinking behind this is straightforward. First, subordinates are the ones who will have to live with this person every day so they should have a voice (a core value – see Step 1) in whom they are reporting to. This also ensures that Google never becomes too hierarchical or bent towards cronyism. Second, a cross-functional interviewer provides a disinterested assessment. A Googler from a different function is unlikely to have any interest in a particular job being filled but has a strong interest in the quality of hiring high.

Hire by Committee

After each interviewer submits his/her detailed feedback, an average score is given to each candidate, and backdoor references have been garnered, an exhaustive hiring packet (includes resume, references, interviewer feedback) is then sent to a hiring committee for review (that’s right, a hiring manager does not get the final say!). Committees typically consist of senior leaders from a team (Product, Finance, People, etc.) plus the recruiter and sometimes a coordinator.

Prospects that have an average of 3 or higher are worth a discussion by the committee. After each candidate is debated, a final yay or nay is given. If the hiring committee rejects the candidate, the process stops there. If they are supportive, their feedback is added to the hiring packet and sent to a senior leader for review.

Finally, if the Senior Leader Review is supportive, then the CEO, will review the file on each candidate and make a final decision. The most common feedback from the CEO is that a candidate might not meet their hiring bar or that the creativity shown in the portfolio isn’t up to snuff. More important than the feedback itself, however, is the powerful message from the CEO to the rest of the company that hiring is taken very seriously at the highest levels.

Although this innovative hiring process is wildly successful, it’s also extraordinarily expensive. Nonetheless, Google believes that most organizations can afford it if they re-allocate resources away from training initiatives and towards re-imagining their recruiting and assessment process. After all, it’s much better, and easier, to hire a 90th percentile performer who starts providing value right away, than it is to train an average performer into a superstar.

Adopt some of these innovative hiring practices now and see the caliber of your talent slowly rise!