Hiring on evidence

The problem it solves

A bad hire rarely looks bad on the first day. The person is easy to talk to, answers fluently, and leaves everyone with the feeling that they will fit. That feeling carries more weight than anything else in the decision, and it predicts the actual work least of all, especially when each interviewer asks whatever occurs to them and what gets compared at the end is impressions rather than answers.

The search usually starts late and in a hurry. Someone resigns, the work gets spread across whoever is left, and the hiring begins once the pressure has become unbearable. The brief describes the person who left rather than what the role has to produce, you interview whoever is available rather than whoever is needed, and the position gets filled because it had to be filled.

The bill arrives later, spread across places nobody adds up: training weeks that have to be run twice, a supervisor spending half the day correcting, a team quietly absorbing what the person does not do, and a departure that reopens the same vacancy with less patience than the first time. If the cycle is never broken, the company ends up blaming the labor market for what its own process produces: it decides there are no good people, lowers the bar to fill seats faster, and teaches its best employees that the extra load is permanent.

What it covers

Hiring well depends less on the interview than on what gets decided before it. Working with the roles the participants actually have open, the workshop rebuilds the whole process: what that position has to produce, what evidence would show that someone can produce it, and how that evidence is gathered before anyone signs. The interview comes last, because it is the piece least able to rescue a badly defined role.

The core work is turning requirements into tests. A structured interview is not a stiffer questionnaire; it is a conversation designed so that two different people, interviewing separately, reach comparable conclusions. From there the session covers past-behavior questions, short exercises that show how someone thinks, reference checks that go further than confirming dates, and the probationary period understood as the final stage of selection rather than an administrative deadline that quietly expires. It also separates what can be verified before hiring from what only shows up once the person is working, and shows how likeability, similarity and first impressions make the decision for the interviewer unless the process gets in their way.

The intensive format leaves room for practice. Each participant turns the requirements of their own role into evidence and writes the questions that go looking for it, built around what a candidate has done rather than what they say they would do. Those questions are then tested in threes: two participants interview a third who plays the candidate, each scores separately, and they compare notes, which is where it becomes obvious whether the guide produces comparable conclusions or only impressions. The reference call is rehearsed out loud as well.

Every organization leaves with its own interview guide for the role it has open or about to open. No general template is handed out. What gets built is the instrument that company will use next month, together with the evidence each requirement calls for, agreed before the first candidate walks in, the reference questions that actually yield something, and a probation plan whose review dates and decision criteria are written down in advance.

Which legal limits apply to the questions, the references and the probationary period in your case is a matter for your legal counsel. In ALUD's practice, hiring belongs to the People engine but is designed with Operations in view, because no method fills a role well when nobody has defined what it is meant to produce.

What participants learn

• Define a role by the results it has to produce, not by the profile you had in mind.
• Turn every requirement into observable evidence before the search opens.
• Run structured interviews that two interviewers can genuinely compare.
• Check references with questions that yield more than a confirmation of dates.
• Use the probationary period as the final stage of selection, with defined review points.

Agenda

1. 01From the profile you wanted to the results the role has to produce
2. 02What can be verified before hiring and what cannot
3. 03Past-behavior questions versus hypothetical ones
4. 04Short exercises and work samples, used with judgement
5. 05The structured interview and comparing notes between interviewers
6. 06Likeability, similarity and first impressions: the biases that decide for you
7. 07References: who to call, what to ask, and what to listen for
8. 08Probation as the final stage of selection

What the organization leaves with

• A structured interview guide for the role you currently have open.
• A list of the evidence required for each requirement, agreed before anyone is interviewed.
• A reference-check format with the questions that actually yield something.
• A probation plan with review points and written decision criteria.