Starting From the Job, Not From the Catalogue

Most assessment processes are assembled from what is available: the platform the organisation licenses, the interview format everybody uses, the test somebody liked. The sequence below inverts that.

The principles in “Starting From the Job, Not From the Catalogue” become easier to maintain when ownership and time spent are visible. Teams evaluating time analytics software can use it to coordinate the administrative work around assessment, identify stages that repeatedly expand and plan capacity, while keeping the hiring decision tied to job-relevant evidence instead of raw activity totals.

For an independent benchmark, compare the process with O*NET occupational information; the useful test is whether the local method remains job-related, proportionate and explainable.

The sequence

One: the job analysis. Five to eight requirements, each with why it matters and what evidence would demonstrate it.

Two: assign a method to each requirement, using the previous note.

Three: check the total burden — hours, stages, cost to the candidate.

Four: write the criteria and rubrics.

Five: pilot on somebody already doing the job well.

Six: run it, recording scores.

Seven: look back at whether the scores related to anything.

Most organisations do steps one, six and stop.

Why starting from the catalogue fails

The available instruments determine what gets assessed, and they were built for the average role rather than yours.

Requirements with no matching product go unassessed and are quietly dropped.

And the process acquires stages that exist because the platform includes them.

The pilot step

Run your assessment on two or three current staff who are good at the role, before using it on candidates.

If they do not score well, the assessment is wrong. This finding is common and saves considerable embarrassment.

It also calibrates what a good score looks like, which is otherwise guesswork in the first round.

Half a day, and it is the step most worth adding to an existing process.

Budgeting the burden

Add up what the whole process asks of a candidate: application, tests, exercises, interviews, travel.

Compare against what the role pays and how many candidates you expect.

Above about four or five hours in total, you are selecting on availability, and the note on candidates who cannot take time off covers who that excludes.

Writing the rubrics

For each criterion: what a strong response contains, what an adequate one contains, what an inadequate one looks like.

Written before anybody is assessed.

Specific enough that two assessors reading it would agree, which is testable and covered in the calibration note.

Reviewing it afterwards

Record scores. Revisit them when you know how the hire worked out.

Small numbers will not prove anything and will show you whether your assessment is obviously disconnected from performance.

Almost nobody does this and it is the only way a process improves.

When to redesign rather than adjust

When the role has changed substantially.

When assessors consistently disagree despite calibration.

When your hires are not working out at a rate that concerns you.

And when the process has accumulated stages nobody can justify, which happens over about three years to every process.

What to check

Did your process start from the job or from the available tools?

Has it been piloted on anybody already doing the role?

What is the total hours burden on a candidate?

And when was it last reviewed against how hires actually performed?

The point

Design from the job rather than from the catalogue, and pilot on somebody already doing the role.

If they score badly, the assessment is wrong.

Underlying all of this

Almost everything in this collection reduces to one discipline: write down what the job requires, assess that thing directly, record the evidence, and look at your own outcomes afterwards. None of it requires buying anything, and organisations that do those four things consistently outperform ones running longer processes built from instruments chosen before the requirements were known.