A growing number of companies have come to see that behavioral interviewing is the most reliable approach to selecting and hiring the right candidates. But what’s the best approach to getting the questions right?

In his 2015 book, Work Rules!, Laszlo Bock, former Senior VP of People Operations at Google, confirms what we should already know – that questions like, “What can you tell us about yourself?” and “What are your greatest strengths/weaknesses?” are, as he puts it, worthless. Bock also disdains more commonly-used interview strategies – here’s a brief excerpt:

Equally worthless are the case interviews and brainteasers used by many firms. These include problems such as: “Your client is a paper manufacturer that is considering building a second plant. Should they?” or “Estimate how many gas stations there are in Manhattan.” Or, most annoyingly, “How many golf balls would fit inside a 747?” … They have little if any ability to predict how candidates will perform in a job.i

Another hugely successful company that attracts thousands of job applicants, Facebook, has given up on those sorts of tactics. Such questions make it seem like they are helping to evaluate intelligence and problem solving while also making the interviewer look smart. However, they do nothing to tell you whether or not the candidate will be successful in the job.

What is the solution?

Many of you already know the answer, which is to return to the same old “boring” types of behavioral questions that have been working for years. These are the ones that generally start with, “Tell me about a time when….” They aren’t sexy, but they work.

We know that the best predictor of how a person will behave in the future is how he or she has behaved in the past, so why not ask candidates to describe what they did and how they did it? The research shows that structured behavioral interviews are the most predictive method of evaluating candidates – aside from seeing a candidate do the actual job.ii

Behavioral Interview Questions: Best Practices

The heart of a good behavioral interview is getting detailed stories about what the candidate did in specific past situations. These stories should give you a “you are there” feeling, as if you could envision what the candidate did, thought, and even said in retelling the events. Unfortunately, many commonly-asked questions that seem like they should get good stories actually don’t tell you much about whether the candidate has the necessary competencies for the job.

Let’s examine a few of these types of questions and see how they fool interviewers into thinking that they are getting good information.

Asking for Typical Situations

Suppose, for example, that you were interviewing a candidate for a sales manager position and you wanted to learn how the candidate on-boards new salespeople. You might ask:

  • “How do you train new salespeople?”

You might get an answer like the following:

  • “I give them as much responsibility as they can handle. For example, I put them in front of customers as soon as I can and I coach them as they give their pitch.”

Questions that start with “How do you (usually do something)?” give you the candidate’s idea of his/her “theory” of how he/she handles situations like this. But did we learn anything about the types of responsibilities assigned to new salespeople or what they were? Or anything about the candidate’s coaching skills? You might assume that the candidate has what it takes, but you have no real evidence – and in fact you haven’t learned anything.

Instead, you could ask something like the following:

  • “Tell me about the last time you trained a new salesperson. Walk me through the story and tell me what you did.”

This gives you a lot of opportunities to follow up with questions like, “Who was this new salesperson?” “What was one of the things you did?” “Give me an example of a specific situation or conversation you had with the new salesperson.” “How did he/she respond?” “How did it all turn out?” Although you get only one story out of many possible stories, at least you get a real story and some real data which you can make an assessment around.

Asking for Hypothetical Situations

Now suppose you wanted to learn how a candidate handles on-the-job stress. You might ask a question like:

  • “What would you do if you had to work with an especially difficult boss?”

Questions about hypothetical situations that start with “What would you do if…?” can be mistaken for behavioral interview questions, except for two things. First, it’s easy for candidates to make up what they think are the “right” answers, which may or may not correspond with what they would actually do in the future when dealing with similar situations. Second, questions like this don’t leave you any opportunity to ask for a story, since the answer would be conjecture, not based in reality.

Instead, ask for an actual situation from the candidate’s past that is relevant to the kinds of challenges he/she might encounter on the job:

  • “Tell me about a specific time when your boss was being especially difficult? Set the scene and tell me how you handled it.”

This gives you a lot of opportunities to follow up with questions like, “What was the situation?” “What was your reaction?” “What was the first thing you did?” “Then what did you do next?” You will not only get a more colorful and detailed story, but you will get a better reflection of the candidate’s attitudes, skills, and reactions in handling stress. Probing for lots of detail is absolutely critical to the effectiveness of a behavioral interview. It ensures that candidates don’t stay at a superficial (and less meaningful) level, and also makes it a whole lot harder for them to “fake” their way through the interviews.

Asking Leading Questions

These questions are the subtlest and most insidious of the bunch. They are often asked when reviewing the candidate’s resume, which may contain something relevant to the job the candidate is being considered for. However, here is where the interviewer can make wrong assumptions about how the candidate handled a responsibility or achieved success in a previous activity.

For example, suppose you were interested in learning how well a candidate sets priorities and manages his/her workload, and you see project management responsibilities somewhere in the resume. You might be tempted to ask a question like the following:

  • “According to your resume, you were responsible for a major project in your current job. Can you tell me how you organized it?”

Here the interviewer is making several errors: first, assuming that the candidate personally managed the project; and second, that the candidate organized the project when it might have been someone else. You are also telegraphing the idea that “organizational skills” are what you are looking to assess, thereby inviting the candidate to make up a story about his/her exceptional organization skills.

A much better question would be something like this:

  • “It says here you were responsible for a major project. Tell me about this project. What was the project about and what did you do?”

By asking the question in this way, you are not “showing your hand” on what skills you are trying to evaluate – you are just trying to learn how the candidate handled the responsibilities of project management and looking for evidence that he/she has the organizational skills you are looking for.

Context is Key

Another thing we have learned in many years of assessing candidates for leadership roles and training clients in the art and science of behavioral interviewing is this: Questions that target specific hiring criteria are much less effective than questions that ask about specific job situations and challenges that require the criteria you are looking for. This is the opposite of what intuitio