The 4.2% Gap and a Full Payroll Budget

Validate before you spend. A pay equity number is a hypothesis until the underlying data survives scrutiny, and acting on a flawed number creates as much risk as acting on none.

Checks.

• Job architecture. Are 'operations coordinator' roles actually comparable across the three states? If coordinators in one facility carry shift-lead duties, the comparison is contaminated. Verify against job descriptions and, ideally, a job evaluation rather than titles.
• Legitimate drivers. Add variables the model omits but that could lawfully explain pay: shift differential (night/weekend), certifications or licenses, relevant prior experience at hire, and performance band versus raw rating. A 4.2% gap is entirely consistent with a night-shift differential if one is not modeled.
• Data quality. Hire date accuracy, re-grade history, and whether the performance rating used is the most recent or a multi-year average; a single year's rating on a 480-person sample is noisy.
• Statistical significance. On a subset of coordinator roles, the sample is small. A 4.2% mean difference among perhaps 60 people can be driven by a handful of individuals. Run the analysis with confidence intervals and also examine the median and the distribution, not just the mean.
• Individual outliers. The scenario says the difference concentrates among pre-2019 hires under a departed manager - that pattern is consistent with both a genuine historical inequity and with a cohort effect unrelated to gender. Check whether men and women hired in that window are actually comparable in shift and site.

What would make me distrust it. Concentration in one facility, a strong shift-differential explanation, or the gap disappearing once hire-level experience is modeled. What would make me more confident. The gap persisting within site and shift, appearing at hiring (starting salary regression by gender), and reproducing in adjacent roles.

Also correct the framing. The 1.1% overall figure is not reassurance - it is a company-wide average that can mask a concentrated problem. Saying 'essentially at parity' at all-hands has already created a communication liability if adjustments follow.

Fund from merit, not from a new line. With 11 months and one analyst, the realistic mechanism is to carve a defined slice of the next merit cycle - for example, 0.5 of the 2.5–3% pool - and direct it to identified individuals as equity adjustments rather than performance increases. This is defensible internally because it is a correction, not a reward, and it must be labeled as such so managers do not read it as a rating outcome. A second source: the promotion and market-adjustment reserve, which is typically underspent. Do not fund by reducing the general increase for everyone; that distributes the cost to the people who were not underpaid.

Sequence.

1. Complete validation (weeks 1–4).
2. Fix starting pay first - the cheapest, highest-leverage action. Publish a hiring range for coordinator roles and require offers to be placed by formula based on experience and shift. This stops the gap from regenerating and costs nothing.
3. Remediate the identified individuals in two tranches: the largest and most defensible gaps immediately, the remainder at the merit cycle. Prioritize by size of gap and by flight risk; a coordinator who is both underpaid and a retention risk is the highest-value dollar.
4. Build job architecture and ranges for the roles that matter most before the transparency law takes effect.

The prior-manager cohort. Do not treat 'a manager who is no longer here did it' as a defense. It is an explanation, not a remedy, and in an enforcement context it often strengthens the case that a systemic practice existed. Remediate on the current gap: if a woman hired in 2017 is 4% below a comparable man, the reason is legally interesting and practically irrelevant. There is, however, a legitimate distinction between adjusting someone who is genuinely below the range for their role versus making a market-level correction - categorize each case accordingly so the spend is defensible line by line.

Internally: explain the system, then the outcome. Do not lead with individual adjustments, which invites comparison and grievance. Lead with a published pay philosophy (what Brightline pays for, how it positions to market, how ranges are set) and published ranges for roles. Employees can then see where they sit and why. Individual conversations about an adjustment happen one-to-one, led by the manager with HR support, framed as 'your pay has been corrected to align with the range for your role and your experience' - factual, no speculation about cause, no commentary on prior management.

Sequence matters. Philosophy and ranges first, adjustments second, manager conversations at the same time as the adjustments take effect. Announcing a correction before employees understand the system generates exactly the comparisons you are trying to prevent.

To leadership: reframe the earlier message honestly. The 'essentially at parity' statement was a company-average claim applied to a role-level question, and reversing it quietly will cost more credibility than correcting it. Recommend a brief, non-defensive correction: the overall number masked a role-specific gap, we found it, here is what we are doing and what it costs. Leaders resist this because it looks like admitting error; the counterargument is that the transparency law will surface the same information within 11 months, and being the source of the disclosure is far better than being the subject of it.

Legal posture. Conduct the validation and remediation under privilege where possible, document the analysis and the business rationale for each adjustment, and have counsel review the communication. Do not characterize the analysis in writing as a finding of discrimination; characterize it as a compensation review with adjustments.

Manager enablement. Brief managers before employees, give them a one-page FAQ, the ranges, and a single escalation path. Managers who improvise explanations about historical pay will create the claims that HR then has to defend.