November 2011 | 

Talent Management: Identifying High-Potential Candidates for Better Execution


True or false: The most valid interviews are designed around each candidate’s unique background.

Wharton management professor Adam Grant recently told a group of executives in Making Strategy Work: Leading Effective Execution, “Talent management is a challenge for every organization, but it becomes critical when you’re tasked with strategic execution. How can you improve your chances of finding the right people? Although research provides strong evidence that many standard practices don’t work well, many managers are not aware of this evidence.

“It’s interesting. There have been so many changes since the economic crisis. Companies understand that they need a new model for mitigating risk, complying with regulations, dealing with competition. And yet most of them continue to use the same methods they’ve used for decades to identify talent.”

Grant cites studies conducted over the past ninety years which show that standard interviews are poor predictors of job performance. “In fact,” he says, “many managers do no better than random chance in selecting high performers.” Most fall prey to at least one of three common interview biases, including: confirmation bias (seeing what we want or expect to see), similarity bias (trying to hire ourselves), and availability bias (paying attention to the most vivid cues, which are not always the most important ones).

Better Interviews Coupled with Ability Testing

The answer to the opening question is false: The most valid interviews are standardized across candidates, not customized to each candidate. Research indicates that for interviews to be effective, they must be structured. That means designing relevant questions in advance, asking the same questions to all applicants, and evaluating responses on a behaviorally anchored response key. However, structured interviews alone are not sufficient to accurately identify top performers. Grant recommends the use of three complementary techniques: work samples, simulations, and situational judgment tests.

Work samples are data or examples from actual work completed, such as a technical report produced by an engineer, revenue generated by a salesperson, or customer satisfaction scores from a service representative. Since past performance is a strong predictor of future performance, relevant work samples provide valuable clues to a candidate’s likely success, and they are twice as powerful as interviews in predicting future performance.

However, many candidates are unable to provide work samples. Some have not worked in a similar job; others have done work that is confidential. In these situations, research shows that simulations are an excellent substitute. Interviewers can create live, hands-on approximations of actual tasks involved in the job and rate how well candidates handle them. For example, many consulting firms conduct case interviews to assess critical thinking and interpersonal skills, and some sales organizations evaluate candidates by asking them to conduct a mock sales pitch during the interview.

In addition to work samples or simulations, Grant supports the adoption of a third technique for evaluating candidates: situational judgment tests. These tests assess job-relevant knowledge and problem-solving skills. Start by creating descriptions of challenging situations that arise in the job at hand. Frame them around desired strategic outcomes as well as current execution challenges. Grant offers an example: “One of your employees doesn’t think he has the resources (budget, equipment, etc.) to complete a task you’ve assigned. What would you do?” Ask top performers and average performers how they would handle the situations, making sure their responses differ. Use their responses to write four or five possible answers, such as:

  1. Tell him how to get the resources
  2. Give the assignment to another employee who doesn’t have the same objections
  3. Tell him to “just go do it”
  4. Ask him to think of some alternatives and review them with you
  5. Provide him with more resources

Score applicants based on how well they match top performers.

Management professor Joe Ryan, who also teaches in Making Strategy Work, agrees that testing for job-related ability makes a difference. “You need to find people who can ‘paint with both hands.’ Too many companies are thinking about sustainable practices; they have ideas, but there’s no execution. They need leadership that brings together smart thinking with smart action.”

Ryan says the need for “ambidextrous” leaders is greater than ever. “There’s a lot at stake. Leadership must be about both meeting today’s numbers, doing well with what you have right now, while experimenting for tomorrow. Finding the talent that can move your initiatives forward has become more complex.”