Jon's Job Blog

In the same year that PPG achieved a perfect 100 on the Disability Equality Index—the national benchmark for disability inclusion in corporate America—I disclosed my ADHD diagnosis as part of an internal job application and was terminated eight weeks later.

I want to hold those two facts together, because the distance between them is the subject of this essay. A perfect score on a national inclusion index. A terminated employee whose disclosure preceded his departure by less than two months. Both facts are true. Both are documented. And the system that produced them sees no contradiction—because the score and the story are measured by entirely different instruments, evaluated by entirely different standards, and visible to entirely different audiences.

This is the 100-Point Paradox I introduced in Essay 2, examined now not as an abstraction about corporate values and marketing but as a concrete experience of what happens when the score says "included" and the person says "terminated."

What the Score Measures

The Disability Equality Index evaluates companies on whether disability inclusion policies and programs exist. Does the company have an accommodation process? Is there a disability-focused employee resource network? Does leadership publicly commit to disability inclusion? Are disability awareness training programs available? These questions have binary answers—yes or no—and a company that can document "yes" to enough of them earns a high score.

What the DEI doesn't measure is whether those policies and programs function in practice. Whether the accommodation process actually results in accommodations being granted. Whether the employee resource network has any influence over employment decisions. Whether leadership's public commitment translates into protected employment for disabled workers during budget constraints. Whether the training programs change any behavior or just occupy a line on a compliance checklist.

PPG documented a perfect score. That documentation was accurate—the policies existed, the network existed, the training existed, the leadership commitment existed. Every "yes" was defensible. What wasn't captured by the scoring methodology was the gap between those documented policies and what happened when a specific employee, in a specific team, disclosed a specific disability and sought the accommodation those policies supposedly guaranteed.

This is what organizational sociologists call "decoupling"—the systematic disconnection between an organization's formal policies and its actual practices. The policies serve one audience (ratings agencies, potential employees, the public) while the practices serve another (managers making daily decisions about who stays, who goes, and who gets accommodated). The two can coexist indefinitely because they're rarely evaluated by the same people using the same criteria at the same time.

The mechanism is elegant in its invisibility. The people who submit the DEI application aren't the people who deny accommodation requests. The people who staff the Abilities First network aren't the people who make termination decisions. The people who write the inclusion language for the careers page aren't the people who design the return-to-office mandates that eliminate remote work for employees whose neurodivergence makes remote work a critical accommodation. Each function operates in its own lane, optimizing for its own metrics, accountable to its own stakeholders. The inclusion team delivers a perfect score. The management team delivers headcount reductions. Neither is evaluated on the other's output.

The DEI perfect score and my termination aren't contradictory. They're complementary. The score proves the policies exist. My experience proves the policies don't protect. Together, they illustrate exactly how decoupling works: the organization builds the infrastructure of inclusion and then makes inclusion discretionary at the point of implementation.

Benefits Listed, Requests Denied

The education reimbursement program is the clearest example I can offer of the gap between listed benefits and granted benefits.

PPG's benefits documentation included education reimbursement as a development resource available to employees. This benefit contributed to the company's scores on various workplace excellence surveys—it demonstrated investment in employee growth, career development, continuous learning. The benefit existed. It was documented. It scored points.

I submitted three separate education requests during my tenure. The first was exploring a Doctorate of Business Administration—the natural progression from the MBA that had been a key qualification for my role in the first place. I wasn't even asking for immediate enrollment. I was trying to have a conversation with my manager about what educational path made sense, what the company might support, how continued education could align with my career trajectory. The conversation was shut down. Not relevant to the business, I was told. The logic didn't hold. PPG hired me in significant part because of my MBA. A DBA would have been the obvious next step in that educational arc—deeper analytical capability, stronger strategic thinking, the kind of advanced credential that benefits both the employee and the organization. But the conversation was closed before it started. I also pursued a Python certification and other training relevant to my automation work. Those were denied too.

The denials didn't come with explanations that would appear on any corporate scorecard. They came as budget constraints, relevance questions, priority misalignment—the kind of managerial discretion that operates beneath the visibility of any rating system. The policy existed. The benefit was listed. The requests were denied or deflected. And the gap between the listed benefit and the actual experience was invisible to every external measurement system evaluating PPG's commitment to employee development.

This is the false economy of decoupled benefits. Listing the program scores points on surveys and attracts talent during recruitment. Denying the requests saves money in the current quarter. The cost of the gap—reduced capability, declining engagement, the message that stated benefits are aspirational rather than actual—doesn't appear until much later, in metrics the organization doesn't directly control.

PPG's historical strength in career development was once its standout feature. In Computerworld's rankings, PPG earned #2 for career development in 2013 and a Top 5 position in 2020. By 2026, that category was absent from the company's recognition profile entirely. The decline didn't happen overnight. It happened through the accumulation of individual denials—each one rational in isolation, each one invisible to external evaluators, and each one teaching another employee that the listed benefit was a fiction.

The trajectory is worth tracing because it shows how institutional capability erodes. In 2013, PPG was nearly the best in the country at developing its IT professionals. By 2020, it was still in the top five. By 2026, the capability had declined enough that it was no longer a distinguishing feature. Thirteen years from near-best to absent. The decline wasn't caused by a single decision. It was caused by a thousand small decisions—denied training requests, deprioritized development budgets, managers who viewed education benefits as costs rather than investments—each one below the threshold of organizational visibility, and cumulatively devastating to the capability they were supposed to sustain.

My denied and deflected education requests were data points in that trajectory. Individually, they were budget or relevance decisions. Collectively, across hundreds of employees making similar requests and receiving similar responses, they constituted the dismantling of a core organizational strength. And the dismantling was invisible to the scoring systems that continued to give PPG credit for having an education reimbursement program on the books.

The Computerworld Validation

PPG's Computerworld "Best Places to Work in IT" ranking dropped from #18 in 2025 to #38 in 2026. Twenty positions in a single year.

This number matters for a specific reason: it validates individual experience at the aggregate level. When I raised concerns about workplace conditions during my employment—about feedback practices, accommodation gaps, the disconnect between stated values and daily operations—those concerns were individual. They could be dismissed as the perspective of one person, one employee, one disgruntled voice. Organizations are practiced at this dismissal. Every terminated employee is "disgruntled." Every complaint is an outlier. The institutional narrative always has a response to individual experience: it's anecdotal, it's subjective, it's not representative.

The Computerworld survey is none of those things. It's systematic, anonymous, and aggregated across the IT population. The survey for the 2026 ranking closed in July 2025—meaning the data reflects the experiences of employees who were at PPG during the same period I was, who worked in the same IT organization, who experienced the same leadership, policies, and cultural conditions. I was in that survey population as an IT employee during the relevant period.

What the survey captured was the aggregate experience of people living inside the same system I've been describing across this series: the feedback culture of Essay 10, the silence of Essay 9, the masking of Essay 8, the accommodation gaps I'm examining here. These weren't my experiences alone. They were shared widely enough across the IT organization to move a national ranking by twenty positions—a decline significant enough to suggest that the cultural conditions I'd raised as individual concerns were in fact systemic.

The team variability I noted in Essay 10 is relevant here too. Other IT teams—aerospace, logistics, data and analytics—seemed to function well. The dysfunction was concentrated in specific teams, including mine in GCSS. But the Computerworld ranking measures the entire IT organization. For the aggregate score to drop twenty positions, the dysfunction in problem teams had to be severe enough to overwhelm the positive experiences in functional ones. The survey isn't measuring an average. It's measuring a distribution—and the distribution had shifted enough toward the negative to drag the overall result down dramatically.

Twenty positions. Not a minor fluctuation. A signal strong enough to move a national ranking by a magnitude that suggests widespread dissatisfaction across the IT organization. The concerns I raised individually—the ones dismissed as one person's perspective—turned out to be shared by enough of my former colleagues to show up in an anonymous survey fifteen months after my termination.

This is the irony that runs through every essay in this series: the organization terminated the person accurately describing the problem, and then the problem showed up in metrics the organization couldn't dismiss as disgruntled. Individual complaints were expendable. Aggregate data is not.

The Financial Decoupling

The decoupling between stated inclusion and practiced exclusion has a financial parallel that makes the pattern even more explicit.

During the same fiscal year that PPG maintained its perfect Disability Equality Index score and saw its Computerworld ranking decline by twenty positions, the top three executives received $19.2 million in combined compensation. The CEO's total package was $11.78 million. The CFO received $4.62 million. The General Counsel received $2.75 million. Their bonuses were paid at 175-184% of target.

These numbers existed alongside a 2.44% decline in revenue, a 12.13% decline in earnings, and a 16% decline in stock price. The company's financial performance deteriorated by every meaningful measure, and the leadership team was compensated as though they had exceeded expectations. The adjusted EPS grew from $7.42 to $7.87—the "financial story" that looked good on the earnings call while the human story remained invisible.

The adjusted EPS number deserves particular scrutiny because it illustrates how corporate accounting constructs narratives. "Adjusted" earnings exclude restructuring charges—the $540 million spent eliminating positions and reorganizing operations. By excluding the cost of firing people from the metric used to evaluate performance, the accounting framework allows leadership to claim improvement during a period of organizational contraction. The company shrank. Revenue declined. People were eliminated. And the metric that determined executive bonuses went up because the metric was designed to exclude the consequences of the decisions executives were making. I'll examine this accounting paradox more fully in Essay 17, where the goodwill impairment analysis becomes a central case study.

The restructuring that eliminated 1,800 positions—approximately 4% of the global workforce—was framed as "structural cost reduction" across seven SEC filings spanning 2024 through Q3 2025. In those filings, the word "knowledge" appears only in reference to "knowledge transfer" during transitions. The concept of institutional knowledge loss—the accumulated expertise, relationships, and organizational intelligence that walks out the door with every terminated employee—appears zero times. The company recorded $540 million in total restructuring costs targeting $175 million in annual savings. The people were a line item. The knowledge they carried wasn't worth mentioning.

Here's what that looks like from the other side: $540 million to remove 1,800 people. Zero dollars in goodwill impairment resulting from their removal. PPG carries approximately $5.7 billion in goodwill on its balance sheet—the accounting representation of institutional value that includes, in theory, the people who create it. Eliminating 4% of those people didn't impair a single dollar. Under GAAP, the company was worth exactly as much—actually slightly more, thanks to projected cost savings—without them.

This is the financial expression of the same decoupling that drives the disability inclusion paradox. The formal system says one thing (people are our greatest asset; disability inclusion is a core value) while the financial system says another (people are a cost to be reduced; their removal creates value). The two systems coexist without visible contradiction because they're evaluated by different audiences using different metrics. The DEI evaluator sees the perfect score. The investor sees the improved EPS. The terminated employee sees neither—they see the gap where both systems failed them simultaneously.

The Disclosure Calculation

Every disabled worker in corporate America faces the same calculation: disclose and risk exposure, or mask and risk collapse. The research is clear on the outcome. The ADA National Network reports that workers with disabilities frequently choose not to disclose because the perceived risk of disclosure outweighs the expected benefit of accommodation. Harvard Health research shows that workplace discrimination is measurably higher for mental health conditions than for physical disabilities—the invisible conditions, the ones most dependent on organizational good faith, are the ones most likely to trigger adverse consequences.

The logic of non-disclosure is rational, even if its consequences are devastating. If the accommodation you need is simple—flexible hours, written communication preferences, asynchronous options alongside synchronous meetings—the risk of disclosure might seem manageable. These are low-cost, low-disruption accommodations that any well-managed organization could implement without significant effort. But "low-cost" and "low-disruption" describe the accommodation in isolation. In practice, the accommodation comes attached to a label: this person has a disability. That label changes how you're perceived, how your performance is interpreted, and how your requests are categorized. A request that would have been routine—"can I continue working remotely?"—becomes a disability accommodation request, which triggers different evaluation criteria, different approval processes, and different organizational calculations about cost and precedent.

My experience confirms the research. Disclosure in February 2024 preceded termination in April 2024. I was seeking confirmation that my remote work arrangements—the closest thing to genuine accommodation I'd experienced—wouldn't be affected by the company's return-to-office push. I was asking, in effect, whether the inclusion policies that scored 100 on the DEI would protect me in practice.

The answer was eight weeks long.

The company's legal position—that decision-makers were unaware of my disability—doesn't survive contact with the speaker series invitation that arrived about a month before my termination. People within PPG identified my mental health challenges as visible and relevant enough to build programming around. The awareness existed. The protection didn't. And the system that produced a perfect inclusion score had no mechanism for bridging that gap.

This is the cruel efficiency of the disclosure calculation. The same organization that scores points for having a disclosure-friendly environment creates conditions where disclosure is demonstrably unsafe. The DEI measures whether the company encourages disclosure. It doesn't measure what happens to people who take the encouragement at face value. And the non-disparagement clause in the severance agreement—discussed in Essay 11—ensures that the people who discover the gap between encouragement and protection can't tell anyone what they found.

What this teaches every other disabled employee at PPG—and at every company with a similar pattern—is that the score is decorative. It doesn't protect you. It protects the company's reputation. When the score and the individual employee's interests conflict, the score wins, because the score is what external evaluators see and the employee's experience is what non-disparagement clauses ensure nobody else does.

When the Score and the Story Diverge

The 100-Point Paradox isn't about one company or one employee. It's about the structural gap between measuring inclusion and practicing inclusion—a gap that the entire corporate disability inclusion infrastructure is designed to obscure.

Rating systems measure the presence of policies because policies are measurable. They can be documented, audited, compared across organizations. The experience of an individual employee navigating those policies is not measurable in the same way—it's specific, contextual, and dependent on the discretionary decisions of individual managers operating beneath the visibility of any rating methodology.

This creates a perverse incentive. Organizations are rewarded for building inclusion infrastructure and face no consequences for failing to use it. The DEI scores points for having an accommodation process. It doesn't deduct points for denying accommodation requests. It scores points for having an ERN. It doesn't deduct points for the ERN having no influence over employment decisions. The system measures the inputs of inclusion while ignoring the outputs, and then holds up the input score as evidence that the outputs must be satisfactory.

The incentive structure extends to the people within the organization responsible for maintaining the scores. The DEI submission team has a clear objective: achieve and maintain a perfect score. They accomplish this by ensuring policies are documented, programs are active, and training is delivered. Their success metric is the score itself. Nobody on that team is evaluated on whether the accommodation process they documented actually resulted in accommodations being granted. Nobody is accountable for the gap between the education reimbursement program on the benefits page and the three denied requests in my file. The incentive is to build the infrastructure and maintain the documentation. What happens inside the infrastructure is someone else's problem—and "someone else" is usually the individual employee navigating the system alone.

The Computerworld decline suggests the outputs were not satisfactory. The 20-position drop tells the story the DEI score doesn't: that a significant number of employees experienced conditions inconsistent with the organization's external reputation. The DEI measured the infrastructure. Computerworld measured the experience. And the two measurements diverged by enough to suggest that the infrastructure wasn't just incomplete—it was decorative.

I don't think PPG set out to build a system that uses disability inclusion as marketing while failing disabled employees. I think the system evolved that way because the incentives point in that direction. Score well on the DEI, and you get recruitment advantages, positive press, and a shield against criticism. Accommodate individual employees in ways that might cost money, create precedent, or require managerial flexibility, and you get... nothing. No score improvement. No external recognition. No measurable return. The system rewards the appearance of inclusion and is indifferent to its practice.

When the score and the story diverge—when the DEI says 100 and the employee says terminated—the question is which one you believe. The system is designed to ensure you believe the score, because the score is public, quantified, and externally validated. The story is private, individual, and constrained by the non-disparagement clause the company purchased along with its silence.

But aggregate metrics eventually tell the story that individual employees can't. Twenty positions on a national ranking. A career development category that went from #2 to absent. A survey population that included the people who lived through exactly the conditions I'm describing. An adjusted EPS that rose while the human costs were excluded from the calculation.

The 100-Point Paradox isn't a failure of one company's values. It's a feature of a measurement system that rewards the appearance of inclusion while remaining structurally indifferent to its practice. Until the scoring methodologies close the gap—until they measure what happens to employees who disclose, not just whether a disclosure process exists—the paradox will persist. Organizations will continue to score perfectly while the employees those scores are supposed to protect continue to discover, at the worst possible moment, that the protection is decorative.

The score says 100. The story says otherwise. And the distance between them is where institutional betrayal lives.

← Essay 11: Severance Under Duress Essay 13: Bureaucracy of Disbelief →