Skip to content
Kosek Advisory Group

Recruiting & Selection

See the full process before you buy it: how we rebuild a role from the selection science up, document every step, and put a credentialed name behind the decision.

Selection science

Hiring at this size scales further when it runs on a repeatable structure instead of instinct and a job post copied from the last one. This page lays that structure out in full: exactly what the Recruiting and Selection engagement does, step by step, before you commit to it.

Kosek Advisory Group (KAG) rebuilds the role from the selection science up. That means the job architecture, the criteria that actually predict performance, and a structured process an Industrial and Organizational (I/O) Psychology practitioner authors and personally backs. Every step traces to job-relatedness, every step is documented, and a named, credentialed professional signs the work.

One thing we say plainly. We do not promise a hire will survive a challenge. We make it a decision you can explain, on the record, to a board or to counsel.

In short

  1. The role is rebuilt from the work itself, not from a copied posting or the last person who held it.
  2. Every screening criterion links to a task the role actually performs. Requirements that predict nothing get removed.
  3. The whole process is written down as it happens, so any decision traces back to the job requirement beneath it.

01 / Job analysis

Rebuild the role from the work, not the last posting

We start by separating three things that usually get collapsed together: the role, the person who last held it, and the posting that was copied off the internet. Only one of them tells you what the job actually requires.

The work is a structured job analysis. We interview the manager the role reports to and the people whose work connects to it, review the artifacts the role produces, and observe the workflow where that is possible. The questions are narrow on purpose: what does someone in this seat actually do, how often, and what happens when it goes wrong.

From there we separate the essential functions of the role from the marginal ones, and rank tasks by frequency, importance, and the consequence of an error. Public occupational libraries, such as the U.S. Department of Labor's Occupational Information Network (O*NET (opens in a new tab)), give us a validated starting vocabulary that we adapt to your business rather than impose on it.

The output

A written role architecture: the essential tasks, the conditions they happen under, and the outcomes the role owns. Every step that follows traces back to this document.

02 / The criteria

Hire against what predicts performance, not proxies for it

For each essential task, we derive the Knowledge, Skills, Abilities, and Other characteristics (KSAOs) required to perform it well. These are the four families measurement science uses to describe what a job actually demands of a person.

Knowledge

The facts, procedures, and domain understanding the work requires a person to already hold.

Skills

The proficiencies a person applies: the things they can demonstrably do, not just describe.

Abilities

The more enduring capacities that make the skills learnable and repeatable over time.

Other characteristics

Work styles, dispositions, and any credential or license the role genuinely requires.

Here is the discipline that makes this selection science rather than a wish list: every criterion has to link to a task in the role architecture. A requirement that cannot be traced to the work does not become a screening criterion. This is where "five years of experience because the last post said so" gets cut.

We separate the must-haves from the preferences, and we define what strong looks like at each level with behavioral anchors. That is how a vague ask like "good communicator" becomes something two people can actually agree on and score the same way.

03 / The evaluation

The same yardstick for every candidate

Every candidate moves through the same process, evaluated against the same criteria, in the same order. Consistency is not bureaucracy here. It is what makes the comparison fair and the result more predictive than a conversation that drifts on rapport.

  • Structured interviews

    Questions written in advance and tied directly to the KSAOs, each with rating anchors so two interviewers score the same answer the same way.

  • Job-relevant exercises

    Work samples where the role warrants them, so you observe the work being done rather than only hearing it described.

  • Independent ratings first

    Interviewers record their scores before they discuss, which keeps the loudest voice in the room from setting the anchor. The panel's reasoning is documented, not averaged into a single number that hides the disagreement.

04 / The documentation

A decision you can walk backward

Every step is written down as it happens, not reconstructed afterward from memory. The record is a deliverable in its own right, as much as the hire is. That is what lets any finalist call trace back to the job requirement beneath it.

The trace

Essential task Required KSAO Structured question or work sample Rating Decision

What you hold at the end is the full chain: the criteria, the scoring records, the rationale for the finalist, and the reasoning for the candidates who were not advanced. It turns "we went with our gut" into a documented, job-related basis you can put in front of someone and explain.

What it is not is a legal opinion. The boundary of that claim is section 06 below, and we hold it carefully.

05 / The authorship

A name on the work, not an anonymous process

The method is authored and signed by a named, credentialed professional, not generated by a tool and left unowned. Brian Kosek, an Industrial and Organizational (I/O) Psychology practitioner, holds the Senior Professional in Human Resources (SPHR), the Professional in Human Resources (PHR), and PROSCI change-management certifications, with more than fifteen years building workforce and selection systems at enterprise scale.

AI accelerates the analysis, and we are open about using it. What it does not do is make the judgment or carry the accountability. A person with measurement-science standing sets the criteria, reviews the evidence, and puts a name to the recommendation.

That is the layer an off-the-shelf hiring tool does not reach, and it is exactly where your highest-stakes calls, the ones about people, get made.

06 / The boundary

What we promise, and what we do not

We do not promise a hire will survive a challenge. We make it a decision you can explain, on the record, to a board or to counsel.

What we promise is a process that is job-related, consistent, documented, and authored by a credentialed professional. What we do not do is give legal advice or guarantee how any dispute would resolve. Where a question crosses into legal territory, we say so and route it to your counsel rather than answering it ourselves.

The value is not a guarantee. It is that when someone asks why you hired the person you hired, you have a clear, job-related, documented answer instead of a shrug.

07 / What you keep

The system outlasts the engagement

When the engagement ends, the role architecture, the KSAO criteria, the structured interview guide with its rating anchors, and the documentation trail are yours. The next time you hire for that role, you start from a validated foundation instead of a blank posting.

That is the firm's standing intent on every engagement: you keep the operating systems, and the business gets better at the work without staying dependent on us.

References

Have a hire you cannot afford to get wrong?

A short consultation is the fastest way to find out whether this engagement fits the role you are filling.

Book a consultation (opens in a new tab)