The Architecture
YOUR DECISIONS AFFECT
Leadership
HR Team
Employee Trust

CA Equal Pay Act · Federal EPA · CO Equal Pay for Equal Work Act

The Architecture

60 days. 14 states. Three different definitions of "equal". Build one framework that satisfies all of them.

Briefing

Tuesday, 7:13 AM. 58 days left. The CEO's email arrived yesterday: "David — you have 60 days to build a compensation framework that works across every state we operate in. Whatever we built before, it's gone. Start from scratch." You are David Park, Director of Total Rewards at Ridgeline Financial. You have three acquired companies with three different pay philosophies, 2,200 employees across 14 states, and a legal team that's stopped returning your CFO's calls. This is not a project. It's a reckoning.

Company Context — Ridgeline Financial

  • Ridgeline Financial — fintech lender, 2,200 employees, headquartered in Denver. Operates in CO, CA, NY, TX, FL, WA, IL, MN, and 6 more states
  • Three recent acquisitions: Apex Analytics (Denver, broad bands), GridPay (NYC, narrow grades), BridgeCredit (San Francisco, fully negotiated — senior engineers vary by 40% at the same level)
  • Current pay gap: 16.6% median, per Priya's preliminary analysis. Data not yet reported to CRD
  • CEO pressure: fallout from the CO/CA/NYC posting violations in Module 1. Board is watching. 60-day deadline is firm
  • Your first meeting is in 47 minutes. Elena Vasquez, Head of Engineering, is already in the conference room.
Activity 1 · The Job Evaluation Criteria

Build the Framework

Before you can build pay bands, you need evaluation criteria — the factors that determine how roles are classified and compared. These criteria will become the backbone of your multi-state compliance defence.

California's standard: "substantially similar work" (skill, effort, responsibility, working conditions). Colorado's: "equal work" (same standard). Federal EPA: "equal work" (equal skill, effort, responsibility under similar conditions).

The criteria you choose will determine which roles are comparable — and which salary differences are defensible. Choose carefully: the wrong combination makes the framework indefensible in court. The right combination makes it a shield.

Select the evaluation criteria for Ridgeline's framework. You can choose up to 6. Some combinations will fail validation — the system will tell you why.

SkillRequired
Education, training, experience, and ability required to perform the job
Required under all three frameworks (CA, CO, Federal EPA). Cannot be excluded.
EffortRequired
Physical and mental exertion required — including cognitive complexity, problem-solving demands, and volume
Required under all three frameworks. Physical-only definitions fail CA's substantially similar standard.
ResponsibilityRequired
Degree of accountability for outcomes, decisions, people, or assets
Required under all three frameworks. Must be applied consistently across functions — not just to revenue-generating roles.
Working ConditionsOptional · Valid
Environment, hazards, time demands, and physical surroundings of the role
Included in Federal EPA and CO. Optional under CA — but including it gives you a broader defence on environmental differences.
Strategic ImportanceOptional
The role's centrality to the company's revenue, growth, or competitive position
FAILS VALIDATION. "Strategic importance" is not a gender-neutral criterion — it tends to rate revenue-generating roles (historically male-dominated) higher than support roles (historically female-dominated). California courts have rejected it. CO investigators have flagged it. Including this criterion puts the framework at risk.
Market RateOptional
External salary benchmarking data from comparable roles in the market
FAILS VALIDATION as a primary criterion. CA Equal Pay Act explicitly prohibits "prior salary" as justification (§1197.5(a)). Courts — beginning with Rizo v. Yovino (9th Cir. 2020, en banc) on prior salary — have applied similar reasoning to "market rate alone", since market data at the senior level embeds documented gender gaps. This area of law is evolving, not fully settled. What is settled: using market data as the primary or sole driver of pay bands creates discoverable exposure. You can reference it as one input in a criteria-driven system, but it cannot be the determining factor or a standalone affirmative defence.
Tenure / SeniorityOptional · Valid
Length of service at the company or in the role
Valid — explicitly listed as a bona fide factor under CA Equal Pay Act and Federal EPA. However: it must be applied consistently and must actually correlate with pay differences. If Brian Kowalski earns more than Keisha Washington and has one fewer year of experience, tenure alone won't justify it.
Documented PerformanceOptional · Valid
Assessed performance against defined objectives, using a consistent rating system
Valid — listed as "merit" under CA and Federal EPA. Critical requirement: the performance system itself must be gender-neutral. If your calibration process has historical bias (and most do), performance ratings can actually make pay gaps worse and harder to defend.
Revenue ContributionOptional
The individual's or role's direct measurable impact on company revenue
FAILS VALIDATION as a standalone factor. Like "strategic importance", revenue contribution tends to rate sales and finance roles higher — roles with historical gender skews. It can be captured within "responsibility" if framed correctly, but as a separate criterion it creates disparate impact exposure.
Education / CredentialsOptional · Valid
Formal qualifications, certifications, and degrees required or preferred
Valid within "skill" — but be careful about requiring credentials that are job-justified. Credential requirements that screen out candidates without business necessity create disparate impact risk.
Day 3 — The Engineering Problem

You share a preliminary version of the framework with Elena. She's read it twice before the meeting starts.

Elena Vasquez

"David, if we apply these criteria consistently, my senior engineers in San Francisco are going to be classified in the same band as the senior legal counsel in New York. That's absurd. The work is completely different."

David Park

"Under 'substantially similar work', California doesn't require the work to be identical — just that the skill, effort, and responsibility are comparable. A senior engineer and a senior counsel both require advanced credentials, independent judgment, and manage significant outcomes."

Elena Vasquez

"The engineers are building the product that generates our revenue. The lawyers are managing risk. That's not the same contribution. The market pays engineers 30% more than counsel at this level — are we ignoring that?"

Priya Mehta

"The market also has a documented gender gap at the senior level. If we use market data to justify the difference, we're essentially importing that gap into our pay structure and calling it policy."

Elena Vasquez

"I'm not asking you to bake in discrimination. I'm asking you to acknowledge that a talent shortage in ML engineering is real, and that 'working conditions' doesn't capture it."

Decision 1 of 2 — The Market Data Question

Elena has a point. Market data is real. The talent shortage for ML engineers is real. If Ridgeline doesn't pay competitively, it loses the engineers who run the product.

But Priya is also right. Market data has a 16% gender gap embedded in it at the senior level. Using it as the primary justification doesn't solve the pay equity problem — it imports it.

You need to give the CEO a position on market data. This will become part of the documented framework.

How does Ridgeline's framework treat market data?

Market data determines pay bands — role comparisons are secondary
Efficiency and talent competitiveness take priority. Pay bands are set to market, and equity analysis runs as a separate check. If there's a gap, it's explained by market rates.
Market data informs bands — but role comparison determines the floor
Benchmark roles to market for competitiveness, but set the internal band floor based on job evaluation criteria. Roles with similar criteria get similar bands, regardless of what "the market" says for that specific job title.
Exclude market data entirely — criteria only
Build pay bands purely on the job evaluation criteria. No market references. Clean, internally consistent, legally defensible — but the CFO and Elena are going to push back hard, and you might lose engineers.
Consequence — The Market Defence Fails

The Market Defence Fails

The CFO loves it. Elena loves it. But when Priya runs the numbers, the framework replicates the existing 16.6% gap — because market data at Ridgeline's job levels has a 16% gender gap built in.

Worse: under California Equal Pay Act §1197.5, "prior salary alone" is explicitly not a valid affirmative defence. Courts have extended this reasoning to "market rate alone". If this framework gets audited, Ridgeline will need to explain why roles that meet the substantially similar work standard are paid differently — and "the market told us to" won't be enough.

You document the decision and move forward. The CFO signs off. In 18 months, Rachel Torres will have to defend it in court.

-3 Compliance
Consequence — A Defensible Position

A Defensible Position

Elena isn't thrilled — she wanted market data to lead. But when you show her that the framework still allows market premiums above the internal band floor for demonstrated scarcity skills, she accepts it.

Priya re-runs the gap analysis. The 16.6% median gap reduces to 9.3% once roles are correctly classified under the new criteria. The remaining gap needs explanation — but now you have a documented methodology that shows the criteria, not the market, drove the classification.

You document the decision: market data as an input, not a driver. This is the distinction that survives court scrutiny under CA's bona fide factor defence.

+3 Compliance
Consequence — Technically Clean, Practically Difficult

Technically Clean, Practically Difficult

The framework is legally clean. Priya confirms the criteria are applied consistently. Elena files a formal objection with the CEO, arguing the framework will make Ridgeline uncompetitive in the ML engineering market.

The CEO calls you. He needs a modified version that acknowledges market realities without using market rate as a justification. You're going to have to add market as an input anyway.

No score impact — you'll get a second chance to position this correctly. The underlying CA Equal Pay Act foundation is intact.

0 Compliance
Activity 2 · State Law Risk Map

Which jurisdictions carry the most legal exposure for Ridgeline's current pay practices?

Before finalising the framework, you need to assess where the risk is concentrated. Ridgeline operates in 9 states with active pay equity laws. Not all of them carry the same enforcement risk.

Place each jurisdiction on the grid based on: likelihood of enforcement action (x-axis) and severity of financial exposure if action occurs (y-axis). See the multi-state exposure overview.

Select a state chip below, then click on the grid to place it. Place all 6, then grade.

Likelihood of Enforcement →
Severity of Exposure →
LOW / LOW
HIGH / HIGH
LOW / HIGH
HIGH / LOW

0 of 6 placed

Decision 2 of 2 — Day 42 · What Gets Documented

The framework is almost done. The CFO has one last question before sign-off:

"David — how much of this process do we actually document? I've talked to outside counsel. They say detailed documentation of how we classified roles could be discoverable in litigation. If we find a gap we can't explain, and we've documented the methodology that created it, we've handed the plaintiff's attorney their case."

The CHRO disagrees: "If we don't document how we made these decisions, we can't prove we made them in good faith. A documented process is our best defence."

Both are right. The question is what you document and how.

What documentation approach do you recommend?

Document everything — criteria, decisions, rationale, all calibration discussions
Full audit trail. Every role classification decision has a documented rationale. Every calibration meeting has minutes. This is the gold standard for proving good faith — and every page is potentially discoverable.
Document under attorney-client privilege — legal team channels all sensitive analysis
Route the gap analysis, the calibration discussions, and the remediation options through outside counsel. Attorney-client privilege protects the candid analysis from discovery. The final framework is documented; the messy path to get there is protected.
Minimal documentation — record the outcome, not the process
The framework is documented. The band assignments are documented. The criteria are documented. The internal debates, the gap analysis, the calibration meetings — those stay off paper.
Consequence — The Full Record

The Full Record

The documentation is comprehensive. Every classification decision is recorded. The calibration discussions are minuted. The gap analysis is attached.

In 18 months, when Rachel Torres is in court defending the Keisha Washington case, she'll have a complete record of how the framework was built in good faith. That record will also show where the gap is — and Rachel will have to work with it rather than around it.

Full documentation is the high-integrity approach. It's also the highest-exposure approach. But a well-built framework can survive scrutiny. A framework built to avoid scrutiny usually can't.

This approach and the attorney-client privilege approach are both correct. What matters more than which you choose is that you choose one deliberately — and that the framework itself is defensible.

+3 Compliance
Consequence — Privileged Analysis, Documented Framework

Privileged Analysis, Documented Framework

The final framework is documented and defensible. The analysis that got you there — including the gap analysis that showed 9.3% remaining after classification — is protected by attorney-client privilege.

Outside counsel reviews every piece of sensitive analysis before it's committed to discoverable form. The calibration discussions that produce the framework are privileged. The final framework itself is public.

This is the most legally sophisticated approach. It protects Ridgeline's candid internal analysis while still producing a documented, defensible methodology. Rachel Torres will appreciate it in 18 months.

+3 Compliance
Consequence — The Undocumented Process

The Undocumented Process

The framework is clean. The documentation is minimal. The CFO is satisfied.

But in discovery, opposing counsel won't just ask for the framework. They'll ask for emails, Slack messages, meeting agendas, and calendar invites from the 60-day period. They'll reconstruct the process from fragments — and fragments without a coherent narrative often look worse than a documented gap.

A poorly documented process doesn't protect you. It just makes you look like you had something to hide. The good faith defence requires a paper trail — not its absence.

-1 Compliance
The Architecture — Debrief

What You Built

Ridgeline now has a compensation framework. Whether it survives the next 18 months depends on the decisions you made here.

The framework David Park builds in this module is the same framework Rachel Torres will defend in Module 5. If you built it well, she has a shield. If you built it to avoid scrutiny, she has a liability.

Your Compliance Score

0/12

Leadership

50%

HR Team

50%

Employee Trust

50%

Legal Risk (lower is better)

50%

Tier Summary

What happens to David

What happens to Ridgeline

Scoring Breakdown

Framework criteria selection (max 3)
Chose skill, effort, responsibility — excluded market rate, strategic importance, revenue contribution. These are the statutory factors under CA, CO, and Federal EPA. Invalid criteria make the framework indefensible.

Market data positioning (max 3)
Treated market data as an input, not a primary driver. CA courts have rejected "market rate" as a standalone affirmative defence. Using it as the determining factor imports existing gender gaps into your pay structure.

Risk map accuracy (max 3)
Correctly placing California in the high-high quadrant — above Colorado — is the decisive judgement call. Underestimating CA is the most common mistake.

Documentation strategy (max 3)
Full documentation OR attorney-client privilege — both earn full marks. Minimal documentation earns none. Good faith requires a paper trail. Both full and privileged documentation are legitimate, defensible approaches.

Key Learnings

  • Three states, three definitions of "equal work" — design to the strictest standard (CA's substantially similar test).
  • Market data reflects existing gender gaps. Using it as a primary pay justification imports those gaps and creates legal exposure.
  • Documentation is a professional judgment call, not a hierarchy: full documentation proves good faith and requires a defensible framework; attorney-client privilege protects candid internal analysis while still producing a public framework. Both are valid. Minimal documentation — recording the outcome without the process — is the only wrong answer.
  • The framework you build now is the evidence Rachel Torres uses in Module 5. Build it to survive a courtroom, not just an audit.

Regulatory References

CA Equal Pay Act §1197.5 — affirmative defences
Four defences: seniority, merit, quantity/quality of production, or a bona fide factor other than sex. "Prior salary" and "market rate alone" are explicitly excluded as valid defences.

CO SB 19-085 / SB 23-105 — equal work standard
Colorado's "equal work" standard covers jobs requiring substantially similar skills, effort, and responsibility, performed under similar working conditions.

Federal Equal Pay Act — the baseline
The federal standard applies to all employees. It's the minimum floor — CA and CO are stricter. Build to CA, and the other two are covered.

Take the Module Quiz →