Accenture Discovers a New Employee Benefit
Accenture has a new employee benefit: not taking vacation.
The company told employees they may roll over unused vacation days this year as it pushes to maximize sales before the end of its fiscal year. Officially, it’s about giving employees more flexibility. Unofficially, it’s hard not to notice that the flexibility only appeared when the company needed everyone to keep working.
Which raises an interesting question: if carrying over unused vacation is such a valuable benefit, why isn’t it always available?
The timing is especially awkward given what’s happening inside the company. Over the past three years, employee sentiment has fallen across nearly every dimension we track. CEO approval is down 15.2 points, positive business outlook is down 13.2, and overall sentiment is down 6.0.
Sometimes the best way to improve employee sentiment is to give people more flexibility. Sometimes the best way is to announce a new flexibility policy that only exists because you need them to skip vacation.
Work-Life Balance Is Out. Work-Life Harmony Is In.
One category held up in the Accenture data. Work-life balance fell 0.7 points over three years, while everything else fell between 5.6 and 15.2.
So perhaps it’s time for a rebrand.
In its 10-K filed yesterday, Peloton introduced us to a new phrase: “work-life harmony.” The company says its compensation and benefits are designed to foster work-life harmony alongside competitive pay and comprehensive benefits.
“Balance” implies a tradeoff. “Harmony” sounds much nicer.
At Peloton, work-life balance is the only sentiment category that has improved over the past three years, up 1.2 points. Everything else moved the other way. CEO approval is down 22.8 points and the share price is down 34.6%.
Then again, it’s probably easier to achieve work-life harmony after shrinking the workforce from 8,662 employees in 2021 to 2,262 today.
Get PeopleReturn insights in your inbox
Speaking of Work-Life Balance…
Amazon has spent years fielding stories about workers urinating in bottles, skipping breaks, and getting tested for COVID on their own time. So we checked what the company had to say about its workforce this year.
Employee demographics were last updated in April 2025 and quietly disappeared from this year’s sustainability reporting cycle. Health and safety, however, received three pages in the sustainability report and its own dedicated page on Amazon’s corporate website.
That is probably the right place to spend disclosure. It is also where the numbers become the hardest to interpret.
On April 6, a 46-year-old worker collapsed on the second level of Amazon’s warehouse in Troutdale, Oregon. Employees later told The Western Edge they were instructed to continue picking and loading orders for more than an hour while he lay on the floor. Amazon said the employee died from a pre-existing medical condition.
That determination does more than explain one tragic incident. It decides whether the death is considered work-related, which in turn determines whether it appears in any workplace safety dataset at all. In most cases, that judgment is made by the company itself and is never revisited.
The result is an uncomfortable reality: nobody knows exactly how many Amazon employees have died on the job. The same is true for Walmart, Home Depot, and virtually every other large employer. That surprised us. We assumed OSHA would have the answer. It doesn’t.
The reasons are structural, not conspiratorial. Injury logs are maintained at the establishment level, so rolling them up requires knowing exactly which facilities belong to which parent company at a given point in time. Contractors, staffing agencies, and delivery partners often employ people working inside the same building without ever appearing on the establishment’s OSHA log. State-plan OSHA programs collect and report data differently from the federal system, while the federal census that most reliably tracks workplace fatalities generally does not identify employers.
This is where we spend much of our time. We map establishments to parent companies, reconcile subsidiaries and contractors, identify duplicate records and reporting anomalies, and calculate dozens of health and safety metrics. Even after doing all of that, the picture is rarely complete. The objective was never perfect certainty. It was to replace anecdotes with evidence.
What that work reveals is that almost every public debate about workplace safety is built around a single statistic. Investors deserve a more complete picture.
Consider the six largest big-box retailers. Depending on which metric you choose, you can reach entirely different conclusions.
No company is green across the board. That’s the point.
Costco illustrates the problem perfectly. It has the highest injury rate in the group and the greatest productivity loss from workplace injuries. It also has the lowest fatality rate and the strongest ratio of restricted-duty cases to missed-work cases. The same company appears to have both one of the best and one of the worst safety records in the industry.
Amazon tells a different story. Its overall injury rate is lower than both Costco and Walmart, making it difficult to support the more extreme claims often made about its workplace safety. Yet one metric stands out dramatically: Amazon reports more than six restricted-duty cases for every case involving days away from work—roughly two-and-a-half times the next highest retailer. The ratio doesn’t tell you whether Amazon is managing injuries better or simply classifying them differently. It tells you where to start asking questions.
Fatalities tell yet another story. Home Depot stands out, while Costco appears remarkably low. Before concluding one company is safer than another, remember how difficult fatalities are to measure consistently. Contractors, delivery partners, corporate structure, and work-relatedness determinations all influence what ultimately appears in the data.
None of these metrics is wrong. Each answers a different question, and each has meaningful blind spots. That’s why we don’t think our value lies in collecting more OSHA records. It lies in helping clients understand what the numbers actually measure, where they can mislead, and how to use them in company engagement, stewardship, and investment decisions.
The goal isn’t to eliminate uncertainty. It’s to help investors make better decisions despite it.
What we’re reading:
- I Helped Run Lululemon. Companies Need to Stop Kidding Themselves About A.I. Every inflated announcement, every time a company blames A.I. for layoffs teaches people to distrust the technology itself. Distrust builds resistance. A.I. is a genuinely powerful technology that still needs people, and putting it to use is slow, human work. The sooner we are honest about that with our boards, our employees, our investors and the public, the sooner we can stop performing A.I. and start building something with it. (New York Times) and Your AI strategy isn’t failing because of bad design. It’s because your team doesn’t believe in you Most AI strategies fail in the same place. Not in the design phase, not in the technology selection, and not in the rollout plan. They fail the moment a leader stands in front of their team and realizes — sometimes immediately, sometimes months later — that the people nodding in the room have already decided not to follow. (Fortune)
- Glassdoor Employee Confidence Index: Record low in July Employee confidence fell to a new record low in July, beating the previous record low set just a few months ago in May, according to the latest data from the Glassdoor Employee Confidence Index. The share of employees reporting a positive 6-month business outlook fell to 43.5%, a decline from 44.4% (revised up from 44%). As energy prices rebounded in July and job security remains a top concern for workers, employee confidence continues to sour. (Glassdoor)
- Black moms are leaving the workforce in a mom-cession that hides a deeper crisis Black mothers with young children are leaving the workforce at an alarming pace. Labor force participation among Black mothers aged 25-54 with children under 5 has fallen 11.5 percentage points in just three months, dropping to its lowest level in over 30 years. (Los Angeles Times)
- The $10 trillion reason to rethink your RTO strategy Low employee engagement is a widespread problem that is having a profound and costly impact on workplaces. According to Gallup, the analytics and advisory company, low engagement resulted in approximately $10 trillion in lost productivity last year. That’s the equivalent of 9% of GDP. Gallup also reported that in 2025 employee engagement was at its lowest level since 2020, and has been on a steady decline globally the last two years. (Fast Company)
- More Jobs, Higher Bar: The 2026 AI Employer Report AI’s introduction to the labor market got off to a rocky start, with speculation that it would be a massive job killer. The data tells a different story. Instead of cutting headcount to save money, companies are using AI to reshape their teams, speed up hiring processes, and raise the bar on what they expect from employees. AI adoption is ramping up, and so far, it looks like a net positive for job creation. (ZipRecruiter)