I’ve been interviewing and placing job candidates for 40 years and tracking their subsequent performance for almost as long. Based on this and training more than twenty thousand recruiters and hiring managers on how to actually predict on-the-job performance one problem always stands out: The best person for the job is often the one not hired. Instead it’s the best presenter who usually gets the offer.

In this case success is problematic. This mistake is magnified 2X by not hiring the better person who just happened to not be as good an interviewee.

The impact of this double whammy problem is summarized in the Performance vs. Presentation grid. By changing the decision to hire or not hire based on performance – the horizontal arrow - you’ll eliminate both problems. Here’s how:

Define the job before defining the person doing the job.

Most job descriptions look like this list of more than 800 jobs on Indeed.com for mechanical engineers in the Chicago area. Other than the common generic responsibilities the requirements define what the person hired needs to have in terms of skills, education and experience. These are not job descriptions; they’re “person descriptions.”

Since clarifying job expectations has repeatedly been shown to be the number one driver of performance, it’s important to define the work that needs to be done before defining the person doing the work. Most jobs can be defined as a series of 5-6 performance objectives. Here’s an example of one and the instruction manual on how to prepare one for any job.

Getting the job is not the same as doing the job.

Emotions play a big role in who gets hired. Most managers overvalue first impressions, affability, and communication skills. Techies overvalue the depth of technical skills. Most interviewers quickly eliminate those who “just don’t fit,” using some nebulous criteria.

One way to overcome these biases is by using a scripted 30-minute interview for all candidates whether they make a good first impression or not. This delay forces objectivity into the assessment. At the end of 30 minutes you can then determine if it makes sense to seriously consider the person. Using a talent scorecard with specific ranking guidelines quickly separates the objective interviewers from those who over rely on emotions or their intuition.

Recognize that strangers are treated differently than acquaintances and referrals.

In a recent post, I contended that people who are personally connected to the interviewer in some way – even loosely – are evaluated differently than strangers. Strangers are assumed unqualified to start. Under this premise they are judged largely on the depth of their skills, level of direct experience, personality and first impression. These are terrible predictors of performance and fit. The connected person begins with a significant advantage: he or she is assumed competent. The subsequent assessment is slower and based on the person’s track record of past performance and ability to learn new skills. Here’s a simple way to assess everyone the same way.

Managers ask irrelevant questions and assess people on meaningless facts.

Brain teasers were proved ineffective long ago although it took a huge study by Googlebefore these questions were shown to be useless. I had a former general manager client who related strong organizing and planning skills with an orderly desk, and wanted to visit every candidate’s office as part of the assessment. This past year I had a client who assumed people who cancel interviews at the last minute due to a family crisis lack a strong work ethic. Since it’s hard to know when a hiring manager or someone on the interviewing team will go ballistic I suggest using more panel interviews. This way everyone hears the same questions and answers and everyone keeps everyone else honest.

The typical process is too transactional.

Filling jobs with those who are the most skilled is much different than hiring the strongest person possible. The former is largely a box-checking exercise with the price determined by supply and demand. The latter involves spending more time with fewer candidates focusing on their past performance, their upside potential and ensuring their intrinsic motivators map to the real job. When people are hired this way there’s an instant improvement in quality of hire, an increase in job satisfaction and a huge reduction in unnecessary turnover.

When people are hired based largely on their presentation skills and affability, their on-the-job success is likely to be random and erratic. While it’s difficult to assess the invisible cost of not hiring the better person, it’s not hard to justify its importance. After you do it a few times, you’ll recognize why it’s worth whatever effort it takes.