Reward strategy and job value · Lesson 3

A Disclosed Point-Factor Comparison

4 min read

Lesson objectives

  • apply a published point scale to job evidence, then test a disputed input before proposing a preliminary ranking.
Reading tools

A point-factor method structures a comparison; it does not turn judgement into an automatic fact. Select work-related factors and define levels before looking at incumbents' names or salaries. This teaching scale has three factors totalling 100 possible points: knowledge needed, up to 40; decision autonomy, up to 35; and consequence of error, up to 25. These weights are not a universal professional standard. A real organisation would need to analyse its roles and check unintended effects before using a scale in pay decisions. A higher job score does not imply every occupant performs better. Put task descriptions, delegation and evidence of consequences in one traceable record, and invite independent challenge of missing evidence before relying on a result operationally.

Our miniature level definitions are explicit. Knowledge scores 10 when the work follows specified steps, 25 when it diagnoses exceptions within one domain, and 40 when it designs a method combining multiple domains. Autonomy scores 5 for following instructions, 20 for selecting a solution within approved policy, and 35 for approving a consequential alternative under known delegation. Impact scores 5 for an error repaired within the team, 15 for service disruption requiring another unit's response, and 25 for a decision capable of changing services across several teams or substantial resources. Do not award 40 because an employee holds a lengthy qualification unless the work needs that integration. Do not award 35 merely because the job description says “leads”; look for an approval boundary and actual decisions. If evidence sits between levels, mark the provisional result and identify the fact that would distinguish the levels instead of inventing an average.

Lesson summary

Apply a disclosed point-factor comparison to case data, then compare the output with the worked check and its limits.