Ask most organisations why they have a job evaluation scheme and you tend to get a version of the same answer. It is there to protect us. It gives every role a rating. It shows we applied a consistent method. It means that if a claim ever lands, there is something credible to put in front of a tribunal.
Job evaluation, in this reading, is a legal artefact. You commission it, you run it, you sign it off. The salary structure is then poured on top of it like a concrete slab. Grades, bands, ranges, midpoints, all resting on an evaluation exercise that gets treated as finished the moment the last role is scored.
I understand exactly how the profession arrived here. I do not think it is a safe place to stand any more.
Where the defensive mindset came from
The framing is not irrational. It is historical. UK equal pay law has, since 1970, given employers a specific advantage where a valid job evaluation study exists. Work rated as equivalent is one of the three routes to establishing equal work under the Equality Act 2010. A defensible, analytical, non-discriminatory scheme genuinely does change the shape of a dispute. It shifts the argument away from an open contest about value and onto ground the employer has already mapped.
Reward teams learned the lesson well. Evaluate the roles, keep the paperwork, sleep better. The scheme became something organisations defended rather than something they actively managed.
The trouble is that the legal advantage only exists while the evaluation is an accurate description of the work. An evaluation is evidence about a role at a moment in time. Roles do not stay still. Restructures, automation, hybrid working, team mergers, headcount freezes that quietly redistribute accountability. Every one of these changes the demands on a role without anyone updating a single document.
The speed problem
Here is the mismatch that should worry Heads of Reward more than it currently does.
A traditional evaluation cycle takes months. Job descriptions are gathered. Panels are scheduled. Consultants are briefed. Committees convene. Outputs are moderated, appealed, re-moderated. By the time a scheme is signed off, a meaningful share of the roles inside it have already moved. The organisation then treats the output as settled for three to five years because the cost and disruption of doing it again are so high.
Meanwhile the business is making pay decisions daily. Offers go out. Counter-offers get approved. Retention uplifts are agreed in a corridor conversation. New roles are created and slotted into a grade by analogy because there is no time to evaluate properly. None of this passes through the evaluation framework. The framework is a photograph. The pay bill is a live feed.
This is the point I would most want reward leaders to sit with. Job evaluation does not become obsolete because the methodology was wrong. It becomes obsolete because the organisation has no governance process for detecting when the work has moved. The question worth arguing about is therefore not which scheme is most rigorous. It is how an organisation keeps its understanding of work current between the moment a role is scored and the moment somebody has to justify the pay attached to it.
The organisation therefore ends up with two realities. There is the documented structure, which is coherent, defensible and out of date. There is the actual distribution of pay, which is current, messy and largely undocumented. The gap between the two is where the exposure sits. Nobody is measuring it.
Why the exposure is now much larger
For years an organisation could live with that gap. Equal pay litigation was slow, expensive to bring and mostly concentrated in the public sector. That has changed on three fronts at once.
First, the market forces argument has weakened considerably. In the Next case, the Employment Tribunal found in 2024 that predominantly female sales consultants were doing work of equal value to predominantly male warehouse operatives. Next argued market rates, recruitment difficulty and commercial viability. The Tribunal held that those factors were indirectly discriminatory and could not be objectively justified, partly because market rates themselves can carry historic assumptions about what counts as men’s work and women’s work. Around 3,500 claimants were involved. The estimated cost exceeded £30 million before any forward-looking correction to pay rates. Next has appealed.
Second, the scale of these claims is enormous and the courts are not making them easier to defeat. Similar litigation is running against Asda, Tesco, Sainsbury’s, Morrisons and Co-op, with claimant groups running into tens of thousands. In May 2026 Tesco failed in its attempt to overturn the approach being taken to assessing the relative value of shop and distribution roles. Back pay in England and Wales can reach six years. A structural error therefore does not cost you this year’s payroll. It costs you six of them, plus the permanent uplift, plus the legal spend, plus the disclosure exercise that pulls your reward team off everything else for a year.
Third, the burden of explanation has shifted. The EU Pay Transparency Directive passed its transposition deadline on 7 June 2026. Only a small number of member states legislated on time. Others have confirmed delays into 2027. That fragmentation has been widely misread as breathing space. It is not. The obligations arriving include individual rights to pay information, published criteria for pay and progression, gender pay gap reporting from 2027 for larger employers and a joint pay assessment triggered where a gap of five per cent or more in a category of workers cannot be justified by objective, gender-neutral criteria and is not corrected within six months.
Read that last requirement carefully. The trigger is not the gap. The trigger is the inability to explain the gap using objective criteria. An employer with a well-evidenced structure and a five per cent gap is in a very different position to an employer with a stale structure and the same number. The regulation does not test your pay. It tests your reasoning.
That is the moment a defensive, slow, periodic approach to job evaluation stops working. You cannot retrofit an explanation in six months for decisions made over six years.
None of these developments create inequity. They simply make it much harder for organisations to rely on assumptions that have accumulated over time.
The better news, which is genuinely better
Everything above describes a problem of design, not a problem of intent. Almost no organisation I speak to is trying to underpay women. What they have is a framework that answers the wrong question on the wrong timescale.
The encouraging part is that the fix does not require choosing between fairness and competitiveness. That trade-off is largely an artefact of collapsing two separate decisions into one number. Internal value and external price are not the same question. Treat them as one and every market pressure permanently distorts your internal logic. Treat them as two and you can respond to the market at speed without corrupting the structure underneath.
How to hold equity and competitiveness at the same time
Five practical shifts, all of which I have seen work.
Sequence the decisions properly. Establish internal role size first. Map roles into bands on that basis. Only then bring market data in as a calibration input for range placement. Market data is a useful input. It should not determine internal value. When market data sets the structure, you inherit whatever historic bias that market carries. The Next ruling shows how badly that can end.
Put market pressure outside base pay, not inside it. Where a genuine shortage exists, use a named market supplement with an evidence base, an owner, an expiry date and a scheduled review. It sits alongside base pay rather than distorting the grade. You stay competitive for the scarce role. You do not permanently re-rate the job. You also have a documented, objective reason for the difference that survives inspection.
Use range width deliberately and publish the rules for movement within it. Two roles of equal internal size can legitimately sit at different points in a range where there are objective, applied and documented reasons, such as demonstrated capability, scope actually held or verified scarcity. What fails is unwritten discretion. Publishing the criteria costs very little. It converts an unexplained variation into a justified one.
Make evaluation continuous rather than episodic. Role drift is now a compliance risk rather than an administrative irritation. An annual role integrity review, run alongside the performance cycle, asks two simple questions of each role. Has the work materially changed? If it has, is the change specific to the individual or has the role itself moved? A successor test helps here. If you replaced the incumbent tomorrow, would you write the same job description? Where the answer is no, you have a re-evaluation trigger and a clean audit trail explaining why the score changed.
Instrument the whole thing. Version the scores. Record who evaluated, on what evidence, against which criteria and when. Capture exceptions as exceptions rather than letting them quietly become the norm. The organisations that will handle transparency comfortably are not the ones with the smallest gaps. They are the ones that can produce the reasoning behind any single pay decision within minutes.
What this buys you beyond compliance
The compliance case is the least interesting part of the argument. A live view of role size pays for itself long before a tribunal is ever in view.
Hiring gets faster. A recruiting manager with a defined internal size, a range and clear rules for premiums does not need a three-week escalation to agree an offer.
Budget goes where scarcity actually is. Named, time-limited premiums are visible and reviewable. You can retire them when a market loosens instead of paying for a shortage that ended four years ago.
Restructures become tractable. When role size is measured consistently, you can model a new operating structure without re-running the whole exercise from scratch.
Managers can answer the pay question. The single most corrosive moment in reward is a manager who cannot explain a salary. Objective criteria give them language that is honest, repeatable and does not require them to blame the system.
Trust compounds. Employees rarely demand to be paid more than everyone else. They demand to understand the logic. Organisations that can articulate that logic see fewer grievances, better internal mobility and considerably less pay compression drama at review time.
Reporting becomes a by-product. When the underlying data is current and structured, gender pay gap reporting and joint pay assessments turn into an extract rather than a project.
The shift I would like the profession to make
Job evaluation as legal insurance is a defensive crouch. It assumes the worst outcome is a claim and that the objective is to survive it. That framing has produced schemes that are slow, expensive, treated as compliance overhead and left to decay between refreshes.
Job evaluation as live infrastructure is a different proposition entirely. It is the layer that lets an organisation understand its own work, price that work honestly against a market it has chosen not to inherit uncritically and explain every decision it makes. Defensibility stops being the goal. It becomes a side effect of running the thing properly.
The organisations that get caught out will not be the ones that never had a scheme. They will be the ones that had a good scheme in 2021 and assumed it was still true. The methodology was never the weak point. The absence of any process for keeping it true was.
That is a fixable problem. The window to fix it is open right now.
