When the Numbers Don't Tell the Whole Story: Why Organizational Risk Requires More Than a Compliance Lens (A Case Study)

An overwhelmingly male workforce can look like a discrimination problem, and sometimes it is. But what happens when the workforce composition is also deeply connected to an organization's mission, client population, operating environment, documented safety history, job architecture and compensation structure?

That was the question confronting a nonprofit CEO who was concerned about the gender composition of the organization's workforce.

The organization existed to support, rehabilitate, train and, where feasible, employ men experiencing substance addiction and homelessness. As a direct consequence of that mission and operating model, its workforce was overwhelmingly male. Women worked for the organization, but at significantly lower headcount. Viewed through a single metric, the disparity could raise an obvious question: Why aren't more women employed here?

TULIP asked a different question: What conditions within the organization produced this outcome and what risks do those conditions actually create?

That distinction matters. And the multi-disciplinary experience TULIP brings matters when presented with a complex problem such as this.

TULIP didn't start with the conclusion. We started with the operating environment.

The organization had previously experienced multiple, documented incidents of workplace violence involving its client population and employees. So, it was not a hypothetical concern or a generalized assumption about women being unable to work safely around men. The incidents were part of the organization's historical operating experience.

Leadership had also taken steps to address those risks through security measures, policies, protocols and other safeguards. That created a much more complex risk environment than workforce demographics alone could reveal. Therefore, TULIP examined the issue across multiple dimensions:

• workforce composition and job distribution;

• the organization's mission and service-delivery model;

• employee interaction with the client population;

• documented workplace-violence history;

• existing security measures and safety protocols;

• the nature and requirements of different jobs; and

• payroll and compensation data.

The objective was not to find evidence supporting management's preferred conclusion. Instead, it was to determine what the evidence, taken together, actually said about the organization's risk. And the law made the distinction important, especially in the strong ideological DEI climate that was emerging at the time. Defensibility of any conclusion or recommendation regarding this issue was top of mind.

Strong consideration was given to the fact that Title VII generally prohibits employment discrimination based on sex, and the fact that the Bona Fide Occupational Qualification, or BFOQ, exception is intentionally narrow. Given the relevance, the U.S. Supreme Court's decision in Dothard v. Rawlinson was also given consideration because it demonstrates that extraordinary institutional safety conditions can matter when evaluating whether sex is a legitimate occupational qualification for particular positions. But the decision does not establish a general "dangerous workplace" exception allowing an employer to exclude women. The analysis must be specific to the position and supported by facts rather than stereotypes. That distinction became critical and was documented in the findings for further exploration by in-house legal.

Still, upon assessing the data provided, the applied analysis was not guided by the question of whether the facility was dangerous enough to justify an overwhelmingly male workforce. The better questions were as follows:

• Which jobs actually presented heightened exposure?

• What did employees in those jobs do?

• What security conditions surrounded the work?

• Were gender-neutral alternatives available?

• And what did the organization's actual employment practices reveal?

That moved the analysis from demographic observation to organizational risk assessment.

Then the payroll data added another dimension where compensation was examined. This was not a pay-equity analysis, and it should not be interpreted as one. Determining whether men and women receive equal compensation for substantially equal work requires a different analysis of job content, comparators and other legitimate compensation factors. Instead, we asked a broader organizational question: How was economic opportunity distributed across the workforce? With that lens on the problem, an important pattern emerged. Although women represented a substantially smaller portion of total headcount, the average salary for women exceeded the average salary of the organization's total workforce by more than 25%. By contrast, the average salary for men was more than 10% below the average salary for the total workforce. That contrast was significant.

Again, it did not establish pay equity, nor did it offset or cure potential discrimination. Instead, it provided a material contextual factor for understanding how workforce representation, job architecture and economic opportunity intersected within the organization. Women were fewer in number, but they were concentrated in higher-paying positions that also tended to involve significantly less direct exposure to the organization's high-risk client environment and operated within more secure working conditions. Men represented the substantial majority of overall headcount, particularly in positions more directly connected to the population the nonprofit existed to serve. The data therefore revealed an organizational reality that headcount alone could not capture: Women had substantially lower representation by headcount, yet their average salary was more than 25% above the overall workforce average. Men dominated the workforce numerically, yet their average salary was more than 10% below the overall workforce average. Nonetheless, those findings did not answer the discrimination question by themselves. But they challenged a simplistic interpretation of the demographic disparity and warranted a deeper examination of what was actually driving the numbers.

TULIP therefore considered the compensation findings alongside job responsibilities, client exposure, workplace-violence history, security practices, workforce composition and the organization's mission. The story the data told was that headcount identified the disparity; compensation complicated the issue; job architecture helped explain it; safety data contextualized it; and the organization's mission helped us understand the operating conditions producing it. Yet, no single data point was sufficient. The value came from understanding how the data converged and what that convergence revealed about the organization's actual risk.

This is where TULIP is different.

Traditional assessments often begin inside a functional boundary.

• A lawyer may ask: What does Title VII require?

• An HR professional may ask: What does the workforce demographic data show?

• A compensation professional may ask: What does the payroll data show?

• A security professional may ask: How do we reduce workplace violence?

• An operations leader may ask: What staffing model allows us to deliver the mission?

Every one of those questions matters. But leadership has to manage the organization in which all of those answers coexist.

That is the space in which TULIP operates.

Our assessments connect legal risk, workforce dynamics, financial data, operational realities, governance, safety and organizational objectives to understand not simply whether a risk exists, but how risks interact across the enterprise.

In this case, the deeper analysis supported a lower discrimination-risk profile than the topline gender disparity might initially suggest. But it also reinforced an important governance principle: historical violence and safety concerns should never become proxies for assumptions about what women can or cannot do. Employment decisions still require job-specific, evidence-based analysis and appropriate consideration of reasonable, gender-neutral safety measures. That nuance is the value.

Experience changes what you know to look for.

TULIP's differentiated approach is informed by experience spanning law, employment matters, executive leadership, technology, workforce strategy, financial and operational oversight, risk assessment and government too. Those disciplines do not sit in separate boxes when we evaluate an organization. They converge.

• A legal issue may contain an operational signal.

• A workforce disparity may have a financial dimension.

• A safety problem may reveal a governance weakness.

• A payroll file may tell you something important about organizational opportunity that headcount alone cannot.

• And a metric that appears problematic in isolation may look materially different once the operating conditions producing it are understood.

This is why TULIP does not approach organizational risk as a checklist exercise.

Compliance asks whether an organization crossed a line. TULIP’s proprietary frameworks and risk intelligence tools are designed to assess what conditions are forming upstream and what those conditions could mean for performance, people, risk and enterprise value downstream.

The distinction matters because organizations do not always experience risks one function at a time. Oftentimes, they experience them as systems. And sometimes the greatest value an advisor can provide is not identifying another red flag. It is knowing when a red flag requires a much deeper investigation before anyone decides what it means.

At TULIP, we see and solve problems differently because the most important risk intelligence often lives between the numbers, not inside any single one of them.