Pay equity audits: why the gap you are not looking for is the one that ends up in a lawsuit

A departing employee mentions a number in her exit interview, almost in passing, and the HR director doing the interview realizes it is lower than what a man hired into the same role eighteen months later is making. Nobody set out to pay her less. No single decision explains the gap. It is the accumulated result of a starting offer that was never benchmarked against the role, a raise cycle that rewarded the people who asked loudest, and three years of nobody pulling the numbers apart to see what they actually showed side by side.

That is how most companies find their pay equity problem: after it already walked out the door, or after a journalist asked a question nobody in the building could answer confidently, or after a new state law forced a salary range onto every job posting and someone finally noticed the ranges did not match what current employees were actually paid. A pay equity audit run under that kind of pressure is a company doing damage control, not a company managing risk.

Here is what a pay equity audit actually is as a practice, why most companies put it off until something forces the issue, how to run one in a way that protects the company rather than hands a plaintiff's attorney a roadmap, what it changes about the offers a recruiter is allowed to make, where legal privilege actually matters, and how to keep the gap the audit just closed from quietly reopening within a year.

What a pay equity audit actually is, and what it is not

A pay equity audit is a structured statistical comparison of what people are actually paid, grouped by role, level, location, and tenure, to see whether gender, race, or other protected characteristics correlate with pay after accounting for the legitimate factors that should explain a difference. It is not a glance at an average salary by department, which tells a company almost nothing useful, because averages hide the exact gaps a real audit is built to find. A department where women cluster in the junior half of a pay band and men cluster in the senior half can show an identical average while hiding a very real structural problem underneath it.

It is also not the same exercise as a market pay study, though the two get confused constantly. A market study asks whether the company's pay is competitive against other employers. A pay equity audit asks whether the company's own pay is internally consistent and free of unexplained gaps tied to protected characteristics. A company can be perfectly competitive against the market and still have a serious internal equity problem, because the market comparison was never the question that mattered for this particular risk.

SHRM's research on pay equity and compensation strategy frames the audit as a regression-based exercise, not a spreadsheet of averages, specifically because only a regression that controls for legitimate factors like role, level, and experience can isolate whether an unexplained gap tied to a protected class actually exists. Skipping that rigor produces a report that looks thorough and proves almost nothing if it is ever actually tested.

Why most companies wait until something forces the issue

Running a pay equity audit voluntarily means a company might find a problem it now has to fix, and fixing it costs real money in the current budget year, while not running the audit costs nothing visible until the day it does. That asymmetry is the entire reason this work gets delayed at company after company that could easily afford to do it properly. The risk is real either way. Only one version of it shows up on this year's budget line.

Salary transparency has made the avoidance strategy much harder to sustain, because a growing list of states now require a posted pay range on every job listing, and a range that does not match what current employees in the same role are actually paid is no longer a quiet internal inconsistency. It is a public data point sitting next to the job posting that anyone, including a current employee, can compare against their own paycheck in about thirty seconds.

Harvard Business Review's research on compensation and organizational trust has repeatedly found that employees who suspect a pay equity problem exists, even without hard proof, disengage and start job searching at meaningfully higher rates than employees who trust the system is fair, regardless of whether an actual gap exists. Waiting to audit does not just risk a legal problem down the line. It is quietly costing the company retention today, among the exact people most likely to be watching closest.

Where to start: grouping roles correctly before you look at a single number

The single most common mistake in a first-time pay equity audit is comparing job titles instead of actual job content. Two people with the title 'senior analyst' in two different departments can be doing work that differs enough in scope and complexity that comparing their pay directly produces a meaningless result, either hiding a real gap or inventing one that is not actually there. A credible audit groups people by comparable work, not by whatever title happened to land on their offer letter, which usually means building out job families and leveling criteria before the compensation analysis ever begins.

How to write a job description covers the discipline of defining what a role actually requires rather than what a hiring manager happened to call it, and that same discipline is the foundation a pay equity audit depends on. A company with inconsistent, hastily written job descriptions across similar roles is going to struggle to group its own employee data correctly, because the descriptions that were supposed to define each role never agreed on what the role actually was in the first place.

The leveling frameworks that come with a mature hiring system matter here too, because a company that has never standardized job levels across departments is effectively trying to run a statistical comparison on data that was never collected consistently. Fixing the leveling structure is not a side project to the audit. For most companies running one for the first time, it is the audit's actual first phase.

Running the analysis without creating a plaintiff's roadmap

A pay equity audit produces a document that says, in plain and specific terms, which employees are underpaid relative to a protected-class comparison and by how much. That same document, if it exists outside attorney-client privilege and a lawsuit is ever filed, is close to the most useful piece of discovery evidence an opposing attorney could ask for, because the company will have already done the hard analytical work of proving its own unexplained gap. This is not a reason to avoid the audit. It is a reason to structure how it gets run.

The standard practice, and the one most employment attorneys recommend without exception, is to have outside counsel direct the audit and engage the statistician or consultant performing the analysis under attorney-client privilege, rather than having HR commission the study directly. Done this way, the raw findings are protected from routine discovery even as the company uses them internally to find and fix real gaps. Skipping this step to save a legal fee is one of the more expensive shortcuts a company can take, because it turns a tool meant to reduce legal risk into a document that increases it.

Cost of a bad hire quantifies what a single wrong hiring decision costs a company, and a mishandled pay equity audit, one that surfaces a real gap and then fails to fix it or protect the findings correctly, carries a comparable order of risk multiplied across every employee the gap touches. The analysis itself is the easy part. Structuring it so the company can actually act on what it finds without creating new exposure is where the real judgment is required.

What to actually do once the audit finds a gap

Finding a gap and doing nothing about it is worse, from both a legal and a cultural standpoint, than never having run the audit at all, because it creates a documented record that the company knew and chose not to act. The fix for a confirmed, unexplained gap is a pay adjustment, not a note for the next budget cycle, and the companies that handle this well build a remediation budget into the audit process from the start rather than treating the fix as a surprise expense to be debated after the fact.

The adjustments should happen quietly and without drawing attention to which employees received one specifically because of the audit, both to protect the privacy of the people being adjusted and to avoid signaling to the rest of the organization exactly who was found to be underpaid and why. A raise that shows up as a routine mid-cycle adjustment protects the employee from becoming a subject of office speculation about their past pay, which is itself a dignity the process owes them.

Recruiting metrics that matter makes the case that what a company measures should connect to real outcomes, not just activity, and the same logic applies to a pay equity remediation budget. The number worth tracking is not how many audits got run. It is how many confirmed gaps actually got closed within a defined window afterward, because an audit with findings and no remediation budget behind it is a report, not a fix.

How this changes what a recruiter can actually offer a candidate

A pay equity audit that has already standardized job levels and confirmed what a role should pay gives a recruiter something enormously valuable at the offer stage: a number they can stand behind with confidence instead of guessing at what a hiring manager will approve in the moment. How to negotiate a job offer gets significantly easier for a recruiter working from audited, leveled pay bands, because the negotiation becomes a conversation about where in a defensible range the offer should land, not an improvised back-and-forth with no real floor or ceiling.

It also removes one of the most common sources of unintentional pay inequity at the exact point where it usually gets introduced: the initial offer. A recruiter negotiating from a candidate's current salary, rather than from what the role itself is worth, systematically carries forward whatever pay gap that candidate already experienced at their last job, and a company that has never audited its own offer practices has no way of knowing how much of its internal gap started this way, one negotiated offer at a time, long before any of the people involved ever worked together.

Salary transparency and a completed pay equity audit reinforce each other directly: a posted range that was never checked against audited leveling data is just a guess dressed up as a commitment, while a range built on top of real audit work is a number the company can defend to a candidate, a current employee, and a regulator with the same confidence, because it is the same number in every conversation.

Keeping the gap from reopening the moment the audit ends

A pay equity audit is a snapshot, and the gap it closes today can reopen within a single hiring or raise cycle if the practices that created it in the first place are never actually changed. A company that fixes every identified gap through a one-time remediation, then goes right back to letting hiring managers set starting offers with no reference to the audited pay bands, should expect to find the same gap again at the next audit, slightly reshuffled but no smaller.

BLS data on the gender wage gap makes clear this is not a problem that fixes itself with time or general economic growth. It narrows only where specific, deliberate structural changes get made and sustained, which means the audit's real value is not the one-time correction, it is the pay bands, leveling criteria, and offer guardrails that the audit forces the company to finally build and then actually use on every subsequent hire and raise.

The practical discipline that keeps a gap from reopening is a recurring audit cadence, typically annual, paired with a rule that no offer or raise can deviate from the audited band without a documented, defensible reason signed off by someone other than the hiring manager making the request. That second part matters more than the audit frequency. A company that re-audits every year but still lets any manager override the band for any reason will keep finding the same gap, year after year, because the override is where the gap was actually coming from the whole time.