Safety Leadership

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

PepsiCo South America did not lower accident ratio with slogans. Andreza Araujo changed who owned the first control layer, when supervisors checked risk, and how they escalated.

By 6 min read
leadership scene showing pepsico supervisors first control layer case — How PepsiCo South America Cut Accident Ratio 50% in 6

Key takeaways

  1. 01The 50% reduction came from changing control ownership, not from more slogans.
  2. 02The first control layer belongs to the supervisor when the work is still being set up.
  3. 03TRIR and LTIFR cannot prove field verification.
  4. 04Compliance is not enough when drift stays hidden.
  5. 05Training only works when decision rights and escalation are clear.

During her tenure at PepsiCo South America, Andreza Araujo led a 50% reduction in accident ratio in six months, and the detail that matters is not the number itself. The important shift was that supervisors stopped acting like messengers for safety paperwork and started acting as the first control layer, which means they owned the first real decision point before work moved forward.

Across 25+ years leading EHS at multinationals, Andreza Araujo has seen the same pattern repeat: when the field waits for the safety team to validate every move, the operation gets slower and less honest at the same time. In A Ilusão da Conformidade, Andreza argues that compliance can look tidy while risk stays untouched, and this case shows that point in practice.

What changed at PepsiCo South America was not a campaign. It was a leadership decision about ownership, timing, and field proof. That is why the result is useful far beyond one company, because the same logic applies wherever supervisors still inherit risk late, after the work is already half started.

Initial scenario

The operation did not need another poster or another slogan about zero accidents. It needed a cleaner answer to a harder question: who decides whether the job is ready, who sees the drift first, and who can stop the work before the deviation becomes normal?

That question matters because metrics such as TRIR and LTIFR only tell leaders what has already gone wrong. They do not tell a supervisor whether the permit was read carefully, whether a temporary change was discussed, or whether the crew is about to improvise around a control that no longer fits the task.

As Andreza Araujo explains in Safety Culture: From Theory to Practice, culture changes when the routine changes. A meeting that ends with no decision is just ceremony, and a control that is not verified in the field is only a document with a title.

The decision

The decision at PepsiCo South America was to move authority closer to the work. In Make The Difference: Be a Leader in Health & Safety, Andreza frames the supervisor as the first line of care, not as a passive relay for instructions from the office, and that is the logic that makes the first control layer real.

Once that boundary is clear, the supervisor is no longer judged only by whether the meeting was held or the form was complete. The supervisor is judged by whether the work was ready, whether the barrier was still in place, and whether the escalation happened early enough to matter.

This is where many leaders hesitate, because giving the field more authority feels risky until they realize that the current system already contains risk, only hidden inside delay. A leader who postpones decisions to protect harmony usually protects drift instead.

Execution in the field

The execution did not depend on a large new program. It depended on smaller routines that made the first control layer visible every day. The supervisor had to check the work conditions before the crew moved, not after the task had already begun. The team had to state what could change the plan. Escalation had to be faster than improvisation.

That sounds simple, but simple is not easy when the culture rewards speed without proof. The point is to create a rhythm where the leader asks for the control that matters most, not the report that looks most complete. Once that rhythm exists, people stop hiding weak signals behind polite updates.

Andreza Araujo has seen this across more than 250 cultural-transformation projects: field discipline improves when the leader changes the questions, not when the leader adds more slides. A stronger meeting is not the goal. A clearer decision is the goal.

If a supervisor cannot verify the critical control before work starts, then the organization does not have a control system. It has a communication habit.

That is also why this case belongs in safety leadership rather than in a generic culture story. The operational change sat in the supervisor role, where the first decision boundary was moved closer to the work and the field could no longer hide behind delayed approval.

Inline takeaway: if your supervisors are still judged mainly on attendance, completion, and calm meetings, the first control layer has not moved. Only the paperwork moved.

Measured result

During the PepsiCo South America tenure, the accident ratio fell 50% in six months. That result is the proof point, but it is not the whole lesson. The useful part is that the result followed a change in control ownership, not a change in slogans.

Dimension Before the change After the change
Supervisor role Message carrier and meeting host First decision point for the work
Field verification Often late or implicit Explicit before work moved forward
Escalation Filtered through hierarchy Faster and closer to the task
Safety result Higher accident ratio 50% lower in six months

That before-and-after pattern is why leaders should be careful with leading indicators. If the dashboard looks better while the worksite still relies on guesswork, the numbers may be cleaner while the controls remain weak. In that sense, the real improvement is not the metric alone. It is the discipline behind the metric.

What leaders should learn

First, the first control layer is a decision role, not a title. A supervisor who cannot stop work or escalate a mismatch is not leading risk, even if the person owns the checklist.

Second, compliance is not enough when the field still improvises. In A Ilusão da Conformidade, Andreza Araujo warns that an operation can appear orderly while its risk remains untouched. This case confirms that warning because the system changed only when the leader moved from recording to deciding.

Third, training alone does not create control. Training helps when the work design, the supervision rhythm, and the escalation path already support it. Without that, training only teaches people how to repeat the wrong process more politely.

Fourth, the result is systemic, not magical. James Reason's Swiss Cheese Model helps explain why a drift that reaches the field often passed through earlier layers first. A supervisor who sees the drift sooner can stop the chain before a near miss becomes a serious injury or fatality.

Fifth, board and director language must match field language. If executives talk only about TRIR, LTIFR, or cost, they will miss the fact that the operation is still deciding late. The field does not need a prettier scorecard as much as it needs a faster control boundary.

What to apply in your operation

If you want to apply the same logic in a plant, terminal, warehouse, or project site, start with the supervisor role and make the first control layer explicit.

  • Write down the exact decision that the supervisor owns before work starts.
  • Define the one or two triggers that force escalation instead of workarounds.
  • End the daily briefing with a field verification question, not a status question.
  • Separate what was discussed from what was checked in the field.
  • Track one leading indicator that proves the control was actually verified, not merely announced.

That change matters because it makes field truth visible. When the supervisor has to prove the control before the crew starts, the organization stops rewarding the illusion that a signed form is the same thing as a safe task.

For leaders who want a practical reference, Make The Difference: Be a Leader in Health & Safety gives the supervisor-level discipline this article points toward, while ACS Global Ventures carries the broader transformation work that makes the discipline stick.

FAQ

What is the first control layer?
It is the first real decision point that can block, reshape, or escalate work before exposure grows. In practice, that is usually the supervisor or foreperson when the work is still being set up.

Why does this matter if my metrics already look good?
Because TRIR and LTIFR are lagging indicators. They can confirm improvement after the fact, but they cannot prove that the control was verified before the job started.

Is this only a safety culture topic?
No. It is also a safety leadership topic, because the result depends on who owns the decision boundary, how quickly drift is seen, and whether escalation is rewarded instead of punished.

Can training deliver the same result by itself?
Not usually. Training works when it sits inside a work design that gives people a clear role, a clear trigger, and a clear path to stop or escalate the job.

Close

If your operation still treats supervisors as message carriers, the field will keep delaying the truth. If you want a cleaner discipline for that role, start with Make The Difference: Be a Leader in Health & Safety, then use the book to reset how decisions, verification, and escalation happen where the work actually lives.

Topics c-level safety-leader safety-culture decision-discipline pepsico

Frequently asked questions

What is the first control layer?
It is the first real decision point that can block, reshape, or escalate work before exposure grows. In practice, that is usually the supervisor or foreperson when the work is still being set up.
Why does this matter if my metrics already look good?
Because TRIR and LTIFR are lagging indicators. They can confirm improvement after the fact, but they cannot prove that the control was verified before the job started.
Is this only a safety culture topic?
No. It is also a safety leadership topic, because the result depends on who owns the decision boundary, how quickly drift is seen, and whether escalation is rewarded instead of punished.
Can training deliver the same result by itself?
Not usually. Training works when it sits inside a work design that gives people a clear role, a clear trigger, and a clear path to stop or escalate the job.

About the author

Andreza Araújo

Safety Culture Expert | Senior EHS Executive

Andreza Araújo is a safety culture expert and senior EHS executive with more than 25 years of experience in environment, health and safety. She is a Civil Engineer and Occupational Safety Engineer from Unicamp, holds a Master's degree in Environmental Diplomacy from the University of Geneva, and completed sustainability studies at IMD Switzerland. Andreza has served in Global Head of EHS roles in Fortune 500 environments, leading cultural transformation programs across multinational operations. She has represented Brazil as a speaker at the United Nations in Paris and has spoken at the International Labour Organization in Turin. She is the author of more than 16 books on safety culture in Portuguese, Spanish, English and German. Her work has earned more than 10 EHS awards, including two recognitions from Indra Nooyi, former PepsiCo CEO.

  • Civil & Safety Engineer (Unicamp)
  • M.A. Environmental Diplomacy (University of Geneva)
  • Sustainability Cert (IMD Switzerland)
  • People Management & Coaching (Ohio University)
  • UN Paris speaker representative for Brazil
  • ILO Turin speaker
  • LinkedIn Top Voice
  • Indra Nooyi PepsiCo CEO recognition (2x)

Documentaries

Watch Andreza's documentaries

Three productions on safety culture, organizational failure and the human lessons behind major disasters.

Podcasts

Listen to Andreza's podcasts

She hosts three shows on safety leadership, EHS and organizational culture, in English and Portuguese.

Summarize with AI