People are almost always a company’s greatest asset. In fact, most companies spend around three-quarters of their entire operating budgets on workforce expenses.
Spending that investment on high performers can make a critical difference. Research by Bain & Company suggests that top performers are four times as productive as average performers. For example, Bain found in sales roles, the best sales reps sold at least eight times more than the average.
Given how critical hiring the right people is to the health of a firm, it’s notable how few organisations invest serious time in helping their hiring managers to select the best candidates. Even fewer regularly or systematically assess the effectiveness of hiring practices and their correlation to performance.
That’s a problem because there’s strong evidence that some hiring practices are significantly better at predicting future performance than others. So, what are they?
The over-reliance on unstructured interviews and short cut assessments
When it comes to making professional hires, for example in marketing or digital, most companies rely on unstructured interviews. These are interviews that don’t have a consistent set of questions or a clear scoring system. In fact, they’re more like guided conversations or simply just friendly chats.
The problem is, this interview style is open to bias. Interviewers make unconscious, implicit associations or lean on stereotypes, often clouding their judgement. The consequence is that unstructured interviews have been found to be a very poor predictor of future success.
Research has found that unstructured interviews have a predictive power of just 14 percent, meaning that fewer than one in six interviews increases the baseline chances that a high performer will be hired.
It’s not just poorly structured interviews that are a poor predictor of future performance. Research at firms like Google has found that ‘brainteaser questions’ (how many golf balls would fit in a 747? Etc.) also fail to predict success. Nor do multiple rounds of interviews, past exam results, or schools attended.
So; it’s time to seriously think about your interview process with a focus on questions and tasks that are proven predictors of future success.
Work sample tests are twice as predictive as interviews
The best predictor of how someone will perform in a job? Giving a candidate a sample piece of work, similar to something they might do in the job, and assessing their performance at it. This makes sense: you wouldn’t hire a chef without first seeing them cook a meal.
John Sullivan, a business professor in San Francisco, suggests asking candidates to:
- Identify problems on the job: Ask questions like: ‘Please walk me through the steps of the process that you’ll use during your first weeks to identify the most important current problems or opportunities in your area’.
- Solve a current problem: Explain an actual problem the hire will face during their first week. Then ask them to walk you through the main steps they would take in order to solve the problem. Make sure you mark their responses against a predefined list of critical steps, like gathering data or consulting colleagues, and score them accordingly.
- Identify the problems in a process. Briefly, describe a flawed existing process related to their job. Ask them to examine the process and identify the top three areas where they predict serious problems are likely to occur.
We also recommend you ask the candidate to prepare something ahead of a final stage interview. While presentations can be daunting for some, seeing prospective new hires in action is often invaluable. A client once described this stage of the recruitment process to us as a school maths exam: The content or the outcome isn’t necessarily the most important part, (especially as the individual doesn’t work at your company), it’s seeing how the individual tackled the problem and looking at their ‘working out’ which says a lot about them.
Focus on a candidate’s ability to learn, adapt, innovate
How seriously a candidate takes continuing professional development can be a key predictor of their future success. It’s, therefore, critical that you ask questions that uncover how a candidate approaches learning and innovation.
Professor Sullivan recommends asking candidates to outline the steps they’d take to maintain expert status, adapt to a technological change in their field, or respond to radically new competitive forces in their marketplace.
At Google, they’ve found that general cognitive tests — assessments of a candidate’s general ability in, for example, verbal and numeric reasoning — are a strong predictor of a candidate’s capacity to learn.
They’ve found that these tests, which like IQ tests have defined right and wrong answers, are more effective than simply looking at past educational performance when deciding whether a candidate is likely to be a high performer at Google.
The takeaway? Don’t try to wing it
If it wasn’t obvious by now, good recruitment isn’t simple. You can’t simply skim read a candidate’s CV, have a pleasant, somewhat random chat for an hour, and then make a serious judgement about whether they’re going to be a successful employee.
It’s crucial that you plan your interview process in advance by pre-selecting your questions and determining acceptable answers ahead of time. Only by structuring your interviews in this way can you seriously reduce the risk of hiring a bad candidate.
This article first appeared on LinkedIn on the 27th of February, 2017
If you’re keen to be the best recruiter you can be, be sure to check out the Australasian Talent Conference, starting 21st June. Early bird tickets available here.