Targeted Selection Interview Guide

What Targeted Selection Actually Is

Targeted Selection is a structured behavioral interviewing methodology developed by the Peterson Group starting in the late 1980s and early 1990s. It isn't a general interview tip sheet or a collection of suggestions. It's a rigid framework designed to eliminate interviewer bias and produce hire-or-no-hire decisions backed by consistent scoring. The core mechanism is simple on paper and frustratingly unforgiving in practice. You predefine competencies, write Behavioral Event Questions for each one, and every candidate gets asked the same questions in the same order while interviewers score responses against predetermined anchors. The reason companies adopt it is usually because they got tired of hiring managers making decisions based on gut feeling, the "likeability factor," or the candidate's confidence rather than demonstrated behavior. Studies consistently show that unstructured interviews have reliability coefficients around 0.20, which is worse than most selection tools. Targeted Selection pushes that number higher through structure and rater training. That's the entire value proposition. Everything else is implementation noise.

How to Build a Targeted Selection Interview Guide

Here's the practical workflow. Start by identifying the job family you're hiring for. A senior engineer guide is not going to look like a customer success guide. Pick three to five competencies that actually predict performance in that specific role. Not company values. Not cultural fit. Competencies like problem-solving under ambiguity, conflict resolution, technical communication with non-technical stakeholders. The distinction matters because people confuse them constantly and build guides that measure the wrong things. Competency selection is where most organizations fail before they even write a question. For each competency, draft two or three Behavioral Event Questions. The question format has to ask about specific past behavior, not hypothetical reasoning. You want "Tell me about a time when..." or "Describe a situation where..." Questions that start with "What would you do if..." are useless in Targeted Selection. They measure reasoning ability, not demonstrated behavior, and that's a different construct entirely. You might get a good answer to a hypothetical question, but it doesn't tell you whether the person has actually done the thing you're asking about.

Then you write behavior indicators. These are the observable actions that signal someone demonstrates or doesn't demonstrate the competency. For conflict resolution, a positive indicator might be "acknowledged the other person's perspective before presenting their own." A negative indicator could be "dismissed concerns without explanation or justification." These indicators become your scoring anchors. When you score a response, you're matching what the candidate said against these behavioral descriptions. The anchored rating scale typically runs from one to five, with each point tied to a behavioral description rather than a vague label like "good" or "excellent." Interviewers need training on how to probe. Behavioral questions invite short answers. Candidates will tell you what happened but won't volunteer specifics about their personal role. The interviewer's job is to ask follow-ups that isolate the candidate's individual contribution from the team's collective work. "What was your specific responsibility in that situation?" "What did you decide to do?" "What was the outcome?" This probing discipline separates people who understand the method from people who think they understand it after reading a blog post.

A Real Problem I Ran Into

About four years ago, I was rolling out a Targeted Selection Interview Guide for mid-level product manager roles at a company that had never used structured interviewing before. We'd written the questions, trained the panel, and started conducting sessions. Everything looked correct on paper. The first six interviews produced nearly identical scores across all raters. Then we brought in a candidate who had worked at a well-known startup and had an exceptionally polished way of describing their experience. Every rater gave them higher marks than any other candidate received, including candidates whose documented accomplishments clearly exceeded the startup candidate's track record. That shouldn't have happened. The scoring anchors are supposed to prevent exactly this kind of halo effect. But what I discovered was that our behavior indicators were too vaguely written. The anchor for a score of four on the "cross-functional collaboration" competency said something like "demonstrates strong influence across teams with minimal conflict." When a charismatic candidate described a smooth project launch, the interviewer mentally checked that box without realizing the candidate hadn't actually described any specific influence tactics or how they handled pushback. The question was technically answered correctly but the evidence was thin. The fix was brutal. We rewrote every behavior indicator to require a specific action verb paired with a measurable result. Instead of "demonstrates strong influence," we wrote "proposed and defended an alternative approach with data, resulting in adoption of the proposed approach by stakeholders who initially opposed it." The revised anchors forced raters to listen for concrete behavioral evidence rather than impressionistic comfort. The inter-rater reliability jumped from about 0.45 to 0.72 after that change, which is the kind of improvement that doesn't happen through any other intervention. It just takes honest rewriting of the indicators.

Common Pitfalls and Counter-Intuitive Truths

Most people treat Targeted Selection as a question bank. It isn't. It's a complete quality control system for the interview process, and removing any piece undermines the whole thing. If you use the right questions but don't score against anchored criteria, you've just created a structured conversation with no accountability mechanism. If you score against criteria but don't train interviewers to probe for specific behaviors, the scores are noise. These components lock together. Another counter-intuitive point that surprises people: less is better when it comes to competencies. I've seen interview guides with twelve competencies and six to eight questions each. That produces a ninety-minute interview session, which means the last half of the interview is just fatigue and repetition. Candidate quality drops because people zone out, and rater accuracy drops because scorers lose track of what they're evaluating. Five competencies, two questions each, ten to twelve minutes per competency with probing. That's thirty to sixty minutes of real interview time. Anything longer and you're not gathering more data. You're gathering redundant data from tired people. There's also a misconception about candidate experience. People assume structured interviews feel robotic or impersonal. In practice, candidates often prefer them because the questions are predictable and fair. Everyone gets the same opportunity to demonstrate their abilities. The experience breaks down when interviewers fail to build basic rapport before launching into the behavioral questions, or when they don't explain the process. A two-minute opening that says "I'm going to ask you about specific situations from your past. There are no right or wrong answers. I just want to understand how you've handled things in real work environments" sets the right frame immediately.

Where Targeted Selection Fails Completely

Targeted Selection works best for roles where past behavior is a reasonable predictor of future performance. That includes most individual contributor positions and many mid-level management roles. It breaks down in a few specific scenarios that people don't always consider. Entry-level roles with limited work history are a genuine problem. If you're hiring recent graduates or career changers, candidates literally cannot describe past behaviors that match your competencies because they haven't had the opportunity to develop them yet. In those cases, using Targeted Selection without modification produces a test that measures luck of birth and opportunity rather than capability. You can partially address this by including academic or volunteer examples, but the evidence quality drops significantly. For these roles, structured situational questions or work sample tests are often more predictive. Creative and R&D roles where innovation depends on unconventional thinking also don't map cleanly onto behavioral frameworks. The competencies tend to measure conformity and reliability, which selects for safe candidates rather than breakthrough thinkers. I've seen this happen multiple times. Teams that ran purely behavioral interview processes ended up with homogeneous skill sets that performed reliably but never pushed boundaries. In those cases, adding technical problem-solving assessments and portfolio reviews alongside the structured interview produced significantly better outcomes.

The method also requires a minimum hiring volume to justify the setup cost. Writing and validating a proper Targeted Selection Interview Guide for a single job family, including competency analysis, question development, rater training, and reliability testing, typically takes two to four weeks of focused work depending on organizational complexity. If you're hiring one person per year into a niche role, that investment won't pay for itself. The return only materializes when you're running repeated hires into the same job family or when bad hires are expensive enough that even a small reliability improvement justifies the upfront effort.

Where to Get a Template or Guide

The Peterson Group still maintains the official materials, including their Targeted Selection Interview Guide resources and certification programs. Their materials are comprehensive but commercial. You'll pay for the templates, the training, and the validation support. Several HR consultancies also offer customized development services that produce interview guides tailored to your specific roles. If budget is a constraint, you can build a functional version internally by following the framework I described above, though you'll need to invest time in developing and refining the behavior indicators yourself. What most people actually end up needing is not a downloaded template but a validated set of competencies and questions for their specific roles. A generic Targeted Selection Interview Guide you find online will have the right structure but the wrong content unless you customize it thoroughly. The structure protects you from bias. The content determines whether you're measuring the right things. Getting both right takes more effort than most organizations plan for, but it's the difference between a process that improves hiring quality and one that just adds paperwork to an already broken system.