Most of us have interviewed lots of candidates over the years for really important positions we were trying to fill. Those interviews often were (and still are) the key determining factor for making a hiring decision.
Sadly, though, many of us have also had to cut ties with employees we ourselves hired after feeling great about the interviews we had put them through. It happens more than most of us would like to admit. Why?
Given the frequently unstructured nature of many interviews, the right questions are often not asked. If they are asked, they may not be asked in the right way to pull out the information we needed to really assess if this candidate would do a good job.
Even when all of that is done well, interviewing has still been shown to be a remarkably poor predictor of actual job success. We often still hire the wrong person whose strong performance in the interview doesn't translate into strong performance on the job.
So what are we missing? And what could we do differently?
Here are four approaches many companies have started to use. Some are easier to put into play than others, but all get you to better predictability of someone's job success:
1. Use cognitive assessments
Early in my career, I took one of these as part of my job selection process for a strategy consulting firm. At the time, I didn't see the link. Why were they testing my intelligence? Didn't that happen a long time ago in my life?
Interestingly, research has found mental ability to be one of the few consistent drivers of job success predictability.
2. Incorporate skills testing
Many companies have incorporated skills testing for those jobs where the fundamental skills are the most important drivers of job success.
For example, many customer service functions now do skills testing on how well someone can talk on the phone to a customer, simultaneously search the knowledge base computer system for customer information, while also then typing in key data.
I'd look like a bumbling fool trying to do all three of those things at once if I had to take that skills test. During an interview, though, I could probably present myself really well as very customer focused and easy to work with, which could be misleading about my lack of other key skills needed to perform the job well. Skills testing helps bridge that gap.
3. Create profiles of the best of the best already doing the job
I recently worked with a small company to develop a job "predictive index". We combined cognitive assessments, skills testing, and competency profiling based specifically on the best performers the company already had doing that job. In other words, we tried to figure out how we could profile the best of the best and search for that profile in candidates.
It started to work, but could we take this to another level?
The biggest driver of job success predictability is...
Job success itself. That may sound ridiculously circular, but the best way to really know if someone is going to be successful in any role is to watch them do that very role. Interviews don't get you there. Cognitive capability assessments don't get you there. Skills tests and predictive profiles get you a lot closer.
4. Go temp to perm whenever you can
In my career, the best hires I've ever made that did the best on the job were those that were initially brought on as temporary workers. Some were consultants who we needed for some key project work. They did such an awesome job and fit into the culture so well that we hired them. Of no surprise, they continued to do great once on-boarded as "real employees."
Others were temps who were brought on to fill gaps and who outperformed many of their employee counterparts. We brought them on, too, and didn't lose any sleep at night about whether they would do a good job. They already had.
To get to this, some companies use short trial periods for some key roles. Some have implemented a mini internship process for many entry level management roles. I worked with one company that hired all key analyst roles as temp to perm.
This concept is getting more traction in many companies even if it makes the process take longer. It would definitely ensure that we were really assessing the thing that matters most when trying to predict job success. And that's job success itself.