Job Drift Is Now a Compliance Risk, Not Just an HR Problem

Every organisation has roles that no longer resemble their job descriptions. A coordinator who quietly absorbed line management. An analyst now owning a regulatory return. A team leader carrying responsibilities that were never re-graded. This slow, silent expansion of a role beyond its original scope is job drift. It happens through small, repeated additions rather than any single decision, which is precisely what makes it so hard to detect.

Job drift was once treated as a manageable irritation. Under the EU Pay Transparency Directive, it has become a source of legal and financial risk.

Why re-evaluation frequency now matters

The Directive identifies four objective, gender-neutral criteria for assessing the value of work: skills, effort, responsibility, and working conditions. Where an employer uses a job evaluation or classification system, that system must rest on criteria of this kind. Roles assessed as comparable in value are treated as work of equal value, even where titles and functions differ. Where a gender pay gap of 5% or more appears in any category of workers and cannot be objectively justified, the employer must carry out a joint pay assessment.

The EU's own guidance on gender-neutral job evaluation, published by the European Institute for Gender Equality in March 2026, is clear that these systems should be reviewed regularly, because roles evolve, new positions appear, and requirements shift. A pay structure that was defensible two years ago is not automatically defensible today.

This is where the problem begins. Traditional job evaluation was never built for that cadence.

The economics of standing still

Legacy job evaluation is slow and resource-intensive. Frameworks can take months to apply and demand significant specialist effort simply to maintain. RoleMapper recently made the point that evaluation frameworks needing months to run, then left untouched for years, inevitably describe an organisation that has already changed.

That cost profile forces a compromise. Because a full re-evaluation is expensive, organisations do it rarely, then allow roles to drift in between. The result is a pay structure that looks orderly on paper and quietly diverges from reality underneath. One workforce survey found that only around 7% of employees who took on expanded duties received additional pay in return, which tells you how routinely drift goes unacknowledged.

Software changes the economics. When the marginal cost and time of a re-evaluation fall sharply, regular re-evaluation stops being a special project and becomes a governance discipline. That single shift, from occasional to routine, is what closes the gap between the structure you can evidence and the work people are doing.

A living job description database

The second change is structural. A platform holds job descriptions and evaluation records in one place, version-controlled and continuously updatable, rather than scattered across documents that age the moment they are signed off.

This is about more than administrative tidiness. In the recent Tesco equal pay litigation, the Court of Appeal examined how an employer's own documentation could be used as evidence of what roles actually involved. Job information that is current, consistent, and traceable is no longer simply good practice. It forms part of the evidential record if a pay decision is ever challenged. A living database gives you the audit trail: what a role was, what it became, when it was re-evaluated, and on what basis.

Grouping and dimension-level analysis

The third capability is analytical. A platform lets you group employees, by function, by grade, by location, by any category relevant to the equal-value question, then compare those groups at the level of individual evaluation dimensions rather than a single overall score.

This is where drift becomes visible before it becomes a liability. An overall score can mask movement underneath it. Comparing one group against another dimension by dimension surfaces the specific factor where a role has quietly grown. It also reveals where two groups doing work of equal value are being scored inconsistently. That granularity is what the Directive's worker-category logic demands, since pay gap analysis has to happen at the category level, not only across the organisation as a whole.

A worked example

Consider a Customer Support Specialist role scored on RDCI in 2023 and again in 2026. Over three years the team absorbed escalation handling and took ownership of a compliance log, while routine diagnostics were automated away. The role's formal position, its title, grade, and reporting line, did not change. A traditional evaluation anchored on those markers would register almost nothing.

Illustrative RDCI dimension scores for a single role over three years. Demand dimensions are scored 1 to 7, Contribution dimensions 1 to 10. Figures are hypothetical.

The Contribution side is broadly flat. The Demand side is not. Relational Complexity and Asymmetric Clarity, the two dimensions most prone to being underweighted, have risen as the team took on conflict-laden escalations and more ambiguous, incomplete information. RDCI records that as a measurable increase in role size. If it pushes the combined index across a band boundary, the role has grown without any formal re-evaluation recognising it. A system that reads only formal authority sees no change at all.

The same logic applies across groups. Suppose a predominantly female support team and a predominantly male technical team sit at a similar combined index. Comparing dimensions, the support team's Relational Complexity scores well above the technical team's, without a matching difference in grade or pay. That difference would not, by itself, demonstrate unequal pay. It would, however, flag an area that warrants closer examination. It is the kind of demand-heavy, lower-authority work the EU guidance identifies as a common source of gender bias. RDCI is built to make exactly this visible, where systems anchored on authority or output tend to miss it.

Reward Logic is built on the Role Demand-Contribution Index, which scores each role across nine dimensions split into Demand and Contribution, making routine dimension-level comparison across groups practical rather than a bespoke consulting exercise.

The shift in mindset

The underlying point is not really about software. It is about frequency. Job drift is inevitable in any organisation where work evolves faster than paperwork. What changes the risk profile is how quickly you can detect and correct it.

For most of the history of job evaluation, the honest answer was that it happened neither quickly nor often. Pay transparency has removed that as an acceptable position. The organisations that manage job drift well will be the ones that treat re-evaluation as continuous rather than periodic, keep their job information genuinely live, and can interrogate their structure at the level of detail where drift actually hides.

Sources