How PepsiCo South America Cut Accident Ratio 50% in 6 Months by Making Supervisors the First Control Layer

The strongest reading of the PepsiCo South America case is not that a large company got lucky. During Andreza Araujo's tenure, the accident ratio fell 50% in six months because supervisors were made the first control layer, which changed how risk was verified before the EHS function arrived.

Across 25+ years leading EHS at multinationals, Andreza Araujo has seen the same pattern in many operations. Leaders approve the target, then delegate the discomfort. In Safety Culture: From Theory to Practice, she argues that culture shows up in repeated decisions. In The Illusion of Compliance, she shows why a clean record can still hide weak control.

The PepsiCo South America Foods footprint was not small. The regional role covered seven countries, 30 factories, and 168 distribution centers, so any serious change had to survive language differences, production pressure, and the gap between regional strategy and the shift that actually handles the risk.

Why the case matters beyond one company

A 50% accident ratio reduction in six months matters because it compresses the usual safety timeline into one management cycle. The visible result was statistical, yet the operational lesson sits in how leaders changed the conditions under which risk was discussed, decided, and controlled.

The common version of safety improvement says that more training, more messages, and stronger discipline should lower accidents. That version misses the managerial layer. As Andreza Araujo argues in Safety Culture: From Theory to Practice, culture changes when repeated leadership decisions change what the organization treats as normal, not when a slogan is printed with better design.

For readers who want the contrast with cosmetic audits, the companion article on how to diagnose compliance theater before the next audit in 30 minutes explains why a tidy inspection can still leave the field untouched. This case matters because it shows the opposite, a management rhythm that changed what supervisors had to prove before work kept moving.

Initial scenario

The initial scenario was a regional operation with factories, distribution centers, and a risk profile that could not be solved from the corporate office alone. PepsiCo South America Foods operated across countries, sites, and work routines, which meant that any accident reduction plan had to work through plant managers and frontline supervisors.

That scale changes the nature of safety culture. A single leader can influence one site for a few weeks, although that influence fades when production pressure returns. A regional operation needs repeatable routines, shared decision criteria, and evidence that local leaders can verify without waiting for a corporate audit.

Heinrich and Bird still help here, not as a perfect prediction model, but as a reminder that serious injuries rarely appear from nowhere. The precursors are usually visible first in the small decisions that get repeated, tolerated, and eventually normalized. For the metrics side of that same problem, see TRIR vs LTIFR vs DART: which metric should the executive review use?

Decision 1. Supervisors became the first control layer

The first decision was to stop treating supervisors as messengers and make them the first control layer. Operators see the hazard first, but supervisors decide whether the organization has time to control it. If the supervisor treats safety as an EHS department concern, the culture remains decorative because the real production decision still happens elsewhere.

In Make The Difference: Be a Leader in Health & Safety, Andreza describes safety leadership as visible action in the routine of work. That point is practical, because a supervisor's credibility is built in ordinary moments, whether a stop-work concern is protected, whether a rushed changeover is slowed down, and whether a corrective action receives real follow-up.

The usable test is simple. Before work starts, the supervisor should be able to answer three questions. What is the exposure. Which control is supposed to hold it. What proof do we have that the control still works. When those answers are vague, the work is not ready, even if the schedule says it is.

Decision 2. The weekly review changed what leaders saw

The second decision was to change the weekly review so leaders looked at proof, not only at counts. A six-month reduction plan needs indicators that reveal whether the organization is changing the exposure pattern before the injury statistics confirm it.

That means repeat deviations, field verification quality, supervisor response time, and closure evidence matter more than a clean month-end chart. In more than 250 cultural transformation projects, Andreza Araujo has seen companies celebrate a tidy dashboard while weak controls remain untouched. The article on safety decision latency explains why that delay is dangerous.

If the agenda still waits for month-end, the organization has already paid the price. The point of the review is not to admire lagging indicators, but to force a faster decision about what must change in the work before the next shift repeats the same exposure.

Measured result

The verified career fact is direct. During Andreza Araujo's PepsiCo South America tenure, the accident ratio fell 50% in six months. That figure should not be read as proof that one tactic works everywhere, because the case involved a specific regional context, a specific operating footprint, and a leadership system capable of acting on evidence.

The number is still important because it forces a serious question. If a multi-country food operation can move the accident ratio materially within six months, why do many mature companies accept years of flat performance while repeating the same training calendar.

In Muito Alem do Zero, glossed as Far Beyond Zero, Andreza warns that lower numbers can coexist with underreporting, severity blindness, and cosmetic compliance. That is why the result should be audited through near-miss quality, repeat deviations, critical control verification, and stop-work behavior under pressure.

Comparison: campaign safety versus operating safety

The case is useful because it separates campaign safety from operating safety. Campaigns can create attention, but operating safety changes the decisions that sit closest to exposure. That is where the difference between a short-lived improvement and a repeatable result usually appears.

The article on how 250 projects turned executive sponsorship into decision discipline is the closest companion reading here. It shows the same logic at a different altitude, because the boardroom and the shop floor fail in different ways, yet both depend on whether someone turns intent into proof.

What to apply in your operation

If you want to apply the same logic, start with one exposure that repeats across locations, then define the field evidence that proves whether the control is alive. Evidence may include permit quality, supervisor challenge questions, near-miss closure quality, maintenance response time, or repeated deviations after correction.

• Pick one recurring exposure and make it the weekly focus until the pattern changes.
• Name the line owner whose decision changes the risk, not just the department that reports it.
• Define the proof that the control still works, then verify it in the field, not only on paper.
• Keep the weekly review until the same deviation stops returning in the next shift or the next site.

Andreza Araujo's books fit this work because they separate appearance from control. Safety Culture: From Theory to Practice gives the logic, Make The Difference: Be a Leader in Health & Safety gives the supervisor routine, and The Illusion of Compliance warns against confusing paperwork with control.

For teams ready to turn the case into a real diagnostic path, the fastest next step is to use the Andreza Araujo book store as the starting point, then move the discussion from reading to field verification.