I must have looked at 100+ competency models over the past 15 years. Surprisingly, they’re pretty much all the same. Oh sure, they all have different words, but when you break them down into everyday speak, they all say pretty much the same thing (i.e., some combination and variation of leadership, drive, cultural fit, team skills, confidence, ethics, and technical competency). And that’s one big reason they don’t work. Everyone’s looking for the same things, and on this basis, average is all you should expect.

I cannot find one study that suggests that the use of competency models have helped companies hire better people. Now don’t fire back with statistics your OD person claimed to prove otherwise. Instead, ask said OD person to exclude minimizing mistakes from the stats and then demonstrate that better people were actually hired. There’s no question a structured process focusing on these competencies will minimize hiring errors due to emotional biases or superficial intuitive judgments. But it’s incorrect, on a causal analysis basis, to assign this positive benefit to the use of competency models. The benefits in this case are due to minimizing emotionally-based decisions. Taking the person on a 30-minute tour of the factory or office before the interview started would have had the same impact. Even if you’re convinced you’re now hiring better people, consider that it could all be attributed to better advertising, a successful employer branding effort, or better recruiters.

Aside from the lack of proof, here are some other reasons to support my contention that the use of classic competency models has not resulted in better hires:

1. They’re filled with too much vision and hope, and not enough practicality. Everyone wants to hire top people, and top people all possess the competencies described in some form. But having these doesn’t mean the person will be successful in your company doing the actual work required with the resources available, working for the assigned manager and his/her unique style, and working in your company’s unique culture with all of the associated pressures, systems, procedures, and personalities.  
2. They’re too generic. What does cultural fit, leadership, and all of the other competencies actually mean on the job? If you don’t define this clearly in relationship to the job, you leave it up to the interviewer’s judgment to make the assessment. In my mind, this is the fundamental reason why competency models are not as effective as they could be.
3. They’re not complete. Success in most positions requires a person to collaborate with others, meet deadlines and budgets, plan and organize work, meet all commitments without making excuses, produce high quality work, live with the resources available, meet all performance objectives, and train and develop others, to name the most important. I find it surprising that these important performance-based competencies are left out of most standard competency models.
4. They’re difficult to accurately assess. Conducting an interview is hard work. That’s why most people revert to one of three classic approaches: the intuitive, the technical, and the emotional. Each is flawed, and without specific guidance for the interviewer to evaluate the required competency in some way, the competency itself has no practical value.

All is not lost, however. For one thing, here are links to a bunch of articles and webcasts to help you better address this critical issue. In addition, following are some out-of-the-box approaches you can use to get some real value from your company’s competency model:

1. Relate the competencies to the actual job. To do this, you’ll need to translate each competency to the actual job being filled. For example, if leadership is important, ask the hiring manager, “What does leadership look like on this job?” If you persist you’ll get some type of outcome or task, like, “Collaborate with the marketing group to develop a product spec everyone can accept.” Then go through each of the standard competencies in your model and ask the same question. Before you’re done, make sure each of these outcomes is measurable, or else you’re back to square one. In the leadership example, you could add “Ensure that the new spec meets the published performance standards, is approved by the product launch team, and is completed in 120 days.”
2. Expand the model to include competencies that actually drive OTJ performance. These “extra” competencies described in point 3 above, like “meet deadlines” and “don’t make excuses,” while not grandiose, actually are the factors that drive successful OTJ performance. (More)
3. Conduct a performance-based interview instead of a behavioral interview. Rather than asking candidates to provide an example of an accomplishment describing the competency, ask them to give you an example of a major accomplishment for each of the performance objectives required for OTJ success. (Here’s how to conduct this type of performance-based interview.) If you do, you’ll discover that multiple competencies in various mixes are required to successfully complete any major accomplishment. When you look at these accomplishments over time, you’ll observe the person’s growth rate and ability to meet the actual performance standards for the job.
4. Use an evidence-based assessment process. I suggest using a talent scorecard comparable to the one described in my book, Hire With Your Head. This scorecard has 10 core competencies that have been shown to best predict on-the-job success. Guidance is provided to rank the person on a 1-5 scale in comparison to real job requirements. As a bonus, it only takes two basic questions to conduct this type of interview.

Bottom line: generic competency models are no better than using skills-based job descriptions for hiring top people who are looking for career moves. To paraphrase Yogi Berra, “When you come to a fork in the road, take it; otherwise it’s déjà vu all over again.”