10 Steps to Increase Interviewing Accuracy into the 90% Range

There are two huge problems when hiring is viewed as an end-to-end process. The first one involves sourcing. Most companies are terrible when it comes to advertising, recruiting, and attracting the best. Of course, as a recruiter, how I make my money is by finding top people that others can’t. And, in today’s Internet age, this is actually quite easy. However, this is a big waste of time if you or your hiring managers don’t know how to accurately assess candidate competency.

I’ve seen many good people get overlooked and underperformers get hired because the recruiter or someone on the hiring team didn’t properly assess competency. With this in mind, I would like to offer a 10-step process that will increase your company’s ability to accurately determine a person’s ability to perform on the job with 80-90% accuracy.

Article Continues Below
  1. Know the job. As odd as this seems, the primary reason most people aren’t very good at assessing competency is that they don’t know what the person really needs to do on the job to be successful. Recruiters are equally at fault here. When you don’t know the job, you over-rely on experience, skills, and qualifications with some combination of intuition, gut-feel, personality, and technical depth. I use performance profiles to determine real job needs, but the key is to focus more on results and deliverables to define the work rather than qualifications.
  2. Emphasize comparable results, not skills and competencies. Dig deep into a candidate’s major accomplishments to find out how their skills, behaviors, and competencies collectively were used to achieve success. Then, compare these results to what’s required on the job. By trending these accomplishments over time, you’ll also be able to observe consistency and growth. This “combo-behavioral” interview is far more effective than looking at individual traits separately. (Here’s the one-question fact-finding interview I use for this.)
  3. Give a very restricted “no” vote that must be proven with facts, not feelings. It’s too easy to say “no” about someone for a minor issue (late, unprepared) or incorrectly assess someone on an important trait that requires skill and training to measure, like meeting performance objectives on time. To minimize these false conclusions, require written documentation with substantive proof for all “no” votes.
  4. Don’t give anyone a full “yes” vote. There is no way someone can assess complete candidate competency across all job needs in a 45-60 minute interview. You might be able to determine if someone is an abject failure in this time, but that’s about it. Instead of allowing a full “yes” vote, assign interviewers a sub-set of factors to assess during the interview (here’s a generic competency model you can use for this.) When interviewers “own” a specific trait (e.g., job-specific problem solving), they tend to be more focused and more accurate.
  5. Require candidates to give a PowerPoint presentation of their background. Errors due to lack of good interviewing skills on the part of the candidate or the manager can be reduced by having candidates present their background in a more structured way using PowerPoint. During the interview, allow candidates to talk through their 6-8 page printed summary with the interviewer getting details and asking for clarification. The presentation consists of a work-history overview, major accomplishments and recognition received at each job, and a summary of strengths and developmental needs. This structured interview approach forces both the candidate and interviewer to stay on point and prevents misunderstandings. The written component is especially valuable in overcoming language gaps.
  6. Conduct more panel interviews. Positive and negative emotional reaction to a candidate is one of the root causes of interviewing errors (lack of job knowledge is the other). Structured panel interviews with 3-4 interviewers are extremely useful in minimizing errors due to first impressions, personality style, and preconceptions. During the session, one interviewer must lead, with the others only allowed to ask for clarifying information. The worst type of panel interview occurs when panel interviewers compete with each other asking their “favorite questions.” Here’s an article on organizing panel interviews you might find useful.
  7. Use job-simulation or problem-solving questions. For years I’ve been asking candidates to present their analysis of a business-related issue in a panel interview as part of a take-home project. As long as the problems are job-related, important traits are uncovered, such as thinking, creativity, communications, confidence, interest, decision-making, and analytical skills. You can also ask a similar question during an interview by asking the person how she would handle a realistic job-problem. Then, get in to a give-and-take conversation. This will provide a rough sense of these same traits. Job-simulations like these, or anything else that demonstrate a person’s ability to handle real job needs, improves the overall predictability of the assessment.
  8. Conduct multiple interviews. If the hiring manager is serious about hiring the best, more than one session should be spent with the candidate. For staff positions I’d recommend at least two meetings, and for mid-management at least three. For executive-level spots a minimum of 5-6 hours spread over multiple sessions is essential. You can’t learn much about a person in the first meeting since everything is scripted and the candidate is prepped. The real truth comes out in the second and third sessions.
  9. Use a formal group debriefing process to reach consensus across all job factors. First, assign interviewers a sub-set of the traits in your competency model (here’s a free generic one with a results-based ranking system) and require them to provide detailed evidence to support their assessment. Then review these in a formal group debriefing session. The hiring manager and more senior people should make their comments last. Also, start off with the positives, before getting into the negatives, to increase group objectivity. Then, foster argument about each trait. At the end, you’ll know your assessment is reasonably accurate if there is little variation in opinion on each factor. Wide variation on each factor is indicative of a superficial or biased assessment.
  10. Implement a multi-step validation process. A multi-step interview process as described is not enough. You’ll need to include some type of cognitive skills testing to assess verbal and numeric reasoning. This will increase assessment accuracy by about five percentage points. A personality DISC-like test is helpful as a confirming indicator when used to guide the second set of interview questions. (I like the Profile XT instrument for both of these.) Of course, you must have drug testing (this weeds out a lot of under-performers) and conduct a formal background check (Hireright.com is the one I recommend). In-depth reference checks must also be conducted, with the hiring manager personally involved. Checkster.com is a new tool that we’re just checking out which provides a 360-degree, anonymous means to conduct more effective reference checks.

Interviewing candidates is an important business process that few companies implement properly, even for those that have competency models and use behavioral interviewing. The problem is weak integration among all of the tools, inappropriate training, and lack of enforcement, including weak metrics. For example, if you’re not measuring predicted candidate performance versus actual performance, you can’t get better through process improvement. (The multi-factor competency model mentioned earlier provides an easy way to do this.)

While some of the steps above are paradigm shifting (e.g., focus on comparable results, group voting, PowerPoint structured interviewing), none are complex. The hard part is in the implementation and monitoring. But if hiring the best is an important strategic initiative for your company, there is no simpler, more accurate, or more effective way to pull it off than what’s described here.

Lou Adler is the CEO and founder of The Adler Group – a training and search firm helping companies implement Performance-based Hiring℠. Adler is the author of the Amazon top-10 best-seller, Hire With Your Head (John Wiley & Sons, 3rd Edition, 2007). His most recent book has just been published, The Essential Guide for Hiring & Getting Hired (Workbench, 2013). He is also the author of the award-winning Nightingale-Conant audio program, Talent Rules! Using Performance-based Hiring to Build Great Teams (2007).

Topics

9 Comments on “10 Steps to Increase Interviewing Accuracy into the 90% Range

  1. Lou offers a valuable set of potent suggestions for maximizing the value of interviewing. However, the research-based upper limit for accuracy of structured, performance-based interviewing, even when combined with a job-related related measure of mental ability, comes in at a population adjusted correlation of .63. While 90% may sound comforting to practitioners, it has no basis in science or reality, and thus undermines moving from excitement rhetoric to provable impact. We will only move out of the shadows and earn seats at the strategy table when we keep it real, numerical, and empirical. I am reminded that ‘the facts continue to be true, whether or not we choose to ignore them.’

  2. Lou –

    Great article… I thought it had some very strong points. Funny enough, we are in the process of implementing a more structured interview process for our company. The process we came up with is similar in a lot of ways? we even use the Profile XT (PXT). It?s been a great tool for us after the initial interview to help the recruiter develop smart questions for the second round of interviews. This part alone has increased the level of talent we are bringing into our company.

    While not all of the steps you outlined apply to every company (I have a fundamental differing point of view with panel interviews), I believe taking a thorough approach to sourcing and accessing talent will go a long way in building any organization looking to become an enduring great company.

    To greater success ~

  3. Lou,

    Once again you nailed it. I find that many of our clients have failed at the very first step. They never really came to and understanding and agreement on the requirements for the job. While we focus primarily on the behavioral side, tools like the Predictive Index PRO give the hiring team the opportunity to individually define the job and then graphically compare and contracts results. I enjoy facilitating the discussion as they bring their individual models together, often for the first time. Nothing makes the process work better than an accurate and agreed upon goal!

    Steve Waterhouse
    http://www.predictiveresults.com

  4. I was surprised to see your advocating multiple people interviewing a candidate at once. It takes us down the path of stress interviews. Candidates, particularly those who haven’t interviewed in some time will be thrown by this technique. Not knowing who to direct answers to and minimizing any one on one bonding to reenforce choosing the company over a competitor.

    Sincerely
    Rich Goldberg

  5. My 2 cents – I’ve used panel interviews for over 20 years – as long as they’re well organized a panel interview minimizes errors due to emotions and prejudices. A well-conducted panel interview should be inquisitive, not inquisitorial, so if a candidate get’s stressed during this type of interview, he/she is probably not that strong. Even beter: not only do all interviewers hear the same info in a panel interview, but assessment accuracy has been shown to increase. So if you want the best person to get hired, you should advocate panel interviews. If you want to hide something and get an average candidate placed, you probably should avoid them.

  6. I think Rich brings up a good point. Very recently several I had two candidates interview with several people at once (4 to be precise). Although the candidates had phone interviewed with the hiring manager and done very well, the candidates all got stressed in the face to face (to face to face to face) interview. Towards the end, the candidates started wandering in their answers and ended up finishing with a poor impression.

    End result was we passed on the candidates because they didn’t interview well.

    I can see situations where this might work, but most occasions it has failed in my experiences.

  7. Rich,

    It’s our duty as Recruiters to coach our candidates through stressful interviews…especially group interviews. One of the very best tactics of progressive, strategic recruiting is the group interview! Instead of breaking down the candidate by sending them through 5-6 interviews over the course of a week and a half, or Hiring Managers trying to compare notes; the group interview allows the Hiring Team to be on the same page, not misinterpret information, and gives the candidate a shorter process. Also, The group interview has a much higher positive outcome than normal one-on-ones (I think it’s an increase of 400%). Most of the top agencies have adopted this method. However, this is only effective if the recruiter is a very capable career coach. I’ve coached some of the most nervous candidates through intimidating group interviews at Fortune 500 companies…and they gotten the job.

  8. Obtaining a mutual consensus on a hire where more than three managers must agree in order to put forth an offer is about to happen as often as I win the lottery at my local deli/quick mart.

    I refer to this as the ‘Hung Jury’ syndrome on item #7 in my ‘Maximizing Search Firm Success’ report which was featured on the front cover of TFL this December. (Click
    http://www.searchwizardry.com/maximizing.html)

    Whenever 3 or more managers must agree – disagreement is more likely. Skeletons in the closet, secret axes to grind, vengeance, jealousy, vindictiveness, egos and numerous other ghosts, demons and goblins surface during such interview events.

    If one manager had a lousy review, poor raise, argument with his boss/cooworker (who doesn’t go through this almost every week?) just before the ‘interview’ the ‘interview’ will be where the individual exerts his ability to throw a monkey wrench into the works and garner the attention desired. Sabotage is sweet revenge even if the candidate ‘deserved’ the job.

    I’ve seen all too often. And having actually secured, managed and presented hundreds of millions in offers over more than 20+ years gives me the experience to speak from.

    Limit interviews to the least amount of managers absolutely necessary or nothing will happen for a very long time. Ultimately, one individual only must possess the authority to break a deadlock.

  9. I thought I must have missed something when I read this. Nowhere does Lou Adler say anything about 3 managers, so where did that come from?

    Your point about a disgruntled manager undermining a result might be valid, but that person will present a problem no matter what. I would suggest reading Lou’s additional observations/recommendations (on adlerconcepts) to get a greater sense of the points he was trying to make on accuracy. As usual, they’re very good.

    Incidentally, I agree it is not always easy to reach a consensus. You and the hiring manager may have to do some reasoning to explain a hiring decision and overcome objections, but nowhere is it written that you must have unanimity. The objective should be to have ‘buy-in’ from all major stakeholders, and if 3 of them happen to be managers and they are excluded from the evaluation, deal with it.

    Accuracy is always much preferred over expediency, or you’ll be hiring mediocre candidates just in the interest of making your numbers. In my humble opinion – coming from as much experience in the business as yourself – everyone is much better served by having the major stakeholders involved with a hire. If you avoid that you’re setting your new hire up to have to deal with resentful colleagues who had no say in the hire. Thereby creating the disgruntlement you described.

Leave a Comment

Your email address will not be published. Required fields are marked *