top of page

Why coaching beats compliance in FMCG safety

  • Nov 15, 2025
  • 8 min read

Updated: 6 days ago

There's a moment in most safety programmes where compliance stops working. Not because the rules are wrong. The rules are fine. Signage is up, procedures are documented, training has been delivered, and the audit trail is immaculate. Everyone knows what they're supposed to do. And yet the near-miss rate hasn't moved in six months.


If you've worked in FMCG manufacturing, warehousing, or distribution, you've probably seen this plateau. The initial gains from putting a formal safety programme in place are genuine and significant. Incident rates drop, compliance scores improve, and the organisation feels safer. But at some point, the curve flattens. You're doing everything the regulations ask. You're ticking every box. And the numbers stop improving.


This is the point where the most effective organisations make a decision that changes their trajectory. They stop trying to comply their way to zero harm, and they start coaching their way there instead.




The compliance plateau


Compliance-based safety programmes are built on a straightforward logic: define the rules, train people on the rules, audit adherence to the rules, and discipline non-compliance. It's a framework that works well for establishing a baseline. If your facility doesn't have clear procedures for forklift operation, pedestrian separation, PPE requirements, and exclusion zones, getting those foundations in place will reduce incidents.


But compliance has structural limitations that become apparent once the baseline is established.


Compliance is binary. You're either compliant or you're not. There's no mechanism within a compliance framework for continuous improvement beyond the standard. Once your team meets the requirement, the system has nothing more to ask of them.


Compliance is backward-looking. Audits assess whether rules were followed in the past. They don't tell you what's happening right now, and they can't predict what will happen tomorrow. By the time an audit identifies a gap, the conditions that created it may have already caused an incident.


Compliance is impersonal. A policy document doesn't know that your night shift has three new starters this week, or that the delivery schedule changed last month and created a new traffic conflict at Bay 6. Rules are written for the general case. Risk exists in the specific.


And perhaps most importantly, compliance focuses on what not to do. Don't enter the exclusion zone. Don't exceed the speed limit. Don't operate the forklift without certification. This framing is necessary, but it doesn't build understanding. It doesn't help someone appreciate why the exclusion zone exists, what the consequences of a breach look like, or how their behaviour connects to the safety of the people around them.


Research from Frontiers in Psychology has found that safety compliance and safety leadership are both significant predictors of safety culture in manufacturing, but that the leadership and coaching elements drive the deeper cultural change that sustains improvement over time. The Bradley Curve model of safety maturity makes a similar point: organisations that rely on compliance alone tend to plateau at the "dependent" stage, where workers follow rules because they're told to. The shift to "independent" and "interdependent" stages, where workers make safe choices because they understand and own the outcome, requires something compliance can't provide.





Production floor or FMCG manufacturing environment

Why FMCG environments hit this wall faster


FMCG operations are particularly susceptible to the compliance plateau, for a few reasons.


The pace is relentless. Production targets are tight, shift patterns are demanding, and the pressure to maintain throughput is constant. In that environment, compliance becomes something people do when they're being watched and skip when they're not, because the rules feel like friction against the real job of moving product.


The workforce is fluid. Seasonal peaks, temporary labour, high turnover, and multi-shift operations mean that the people on your floor this week may not be the same people who were trained last month. A compliance programme that relies on annual training and periodic audits can't keep pace with a workforce that's constantly changing.


The hazards are mixed. FMCG facilities combine forklift traffic, pedestrian movement, conveyor systems, loading dock activity, cold storage, and sometimes chemical handling, all in the same building. The interactions between these hazards create risk patterns that a static set of rules can't fully anticipate. A new delivery schedule creates a traffic surge that the compliance programme didn't plan for. A layout change introduces a blind corner that the existing signage doesn't cover.


These are the conditions where coaching becomes essential, because coaching is responsive, specific, and continuous in a way that compliance can't be.




What coaching actually looks like


When I talk about coaching in a safety context, I don't mean pulling someone into an office for a formal conversation about what they did wrong. That's discipline dressed up in coaching language, and it doesn't work.


Real safety coaching is a five-minute conversation at the start of a shift, using a short video clip of a near miss that happened in the team's own work area yesterday. The clip is face-blurred, so nobody is being identified or singled out. The supervisor says something like: "This happened in our zone yesterday afternoon. What do we think caused it, and what could we do differently?"


That conversation does several things that compliance can't.


It's specific. It's not a generic reminder about forklift safety. It's a real event, in a real location, that happened to this team, yesterday. The relevance is immediate and undeniable.


It's constructive. The supervisor is asking the team to problem-solve, not telling them off. The framing is "what could we do differently?" not "who messed up?" That distinction matters enormously for whether people engage or shut down.


It's timely. The event happened yesterday. The conversation is happening today. In a compliance model, this event might appear in a monthly report, if it's captured at all. By then, the context is gone and the learning opportunity has passed.


It builds understanding. When a team discusses why a near miss happened (the delivery truck arrived during shift changeover, the pedestrian walkway was partially blocked by staging pallets, the forklift had to reverse blind because of how the racking was positioned), they develop a deeper understanding of the conditions that create risk. That understanding persists long after the conversation ends, in a way that a policy reminder never does.


And it creates ownership. The team isn't being told what to do. They're being asked to contribute to the solution. That shift from passive compliance to active participation is the cultural change that moves organisations past the plateau.





Supervisor or team leader in conversation with a team member

Why data makes coaching possible at scale


The challenge with coaching has always been scale. A good supervisor with strong safety instincts can coach their team effectively using their own observations. But that approach depends on the supervisor being present, being observant, and having the time to prepare. In a large FMCG operation running 24 hours across multiple zones, that's not realistic for every shift, every day.


This is what computer vision AI changes. When the platform is detecting safety events continuously across every connected camera, the coaching material generates itself. A supervisor doesn't need to witness a near miss to coach on it. The system captures the event, packages it as a short, face-blurred coaching clip, and delivers it to the relevant supervisor ready to use in the next shift briefing.


The heatmap provides the strategic layer. It shows the supervisor (and their operations manager, and the site leader) where events are concentrating, what time of day risk peaks, and whether the coaching conversations are actually reducing event density in the target zones. This closes the loop between coaching and measurable improvement: you coach on a specific issue, then you watch the data to see if the behaviour changes.


For FMCG operations with multiple sites, the same data enables consistency. Every site gets the same quality of coaching material, the same visibility into risk patterns, and the same ability to measure improvement, regardless of whether the local safety manager is experienced or newly appointed. The platform provides the foundation, and the supervisor provides the human connection.





Dashboard or data screen with safety analytics

The cultural shift


The difference between a compliance-led programme and a coaching-led programme isn't just tactical. It's cultural.


In a compliance culture, safety is something that happens to you. Rules are imposed, audits are conducted, and non-compliance is punished. Workers learn to follow the rules when observed and to work around them when not. Near misses go unreported because reporting feels risky. The safety team is seen as enforcement, not support.


In a coaching culture, safety is something you participate in. Events are discussed openly because the response is constructive, not punitive. Workers contribute to identifying problems because they've been asked for their input, not their compliance. The safety team is seen as a resource that helps the operation run better, not a police force that slows it down.


This cultural shift is what organisations operating under New Zealand's HSWA, Australia's WHS framework, or any progressive safety regulation are ultimately trying to achieve. The regulators don't just want compliance. They want organisations to demonstrate that they're actively managing risk, engaging workers in safety decisions, and continuously improving. A coaching-led programme with continuous data to back it up provides exactly that evidence.




Compliance is the floor, not the ceiling


To be clear: this isn't an argument against compliance. Compliance is essential. You need the rules, the procedures, the training, and the audit trail. These are the foundations that keep people safe and keep your organisation on the right side of the law.


But compliance is the floor, not the ceiling. The organisations that achieve genuinely low incident rates, that build cultures where workers actively look out for each other, that sustain improvement year after year, are the ones that build coaching on top of compliance. They use the rules as the starting point and the conversation as the vehicle for continuous improvement.


The shift from compliance to coaching isn't about doing less. It's about doing more, in a way that actually changes how people think and act when nobody's watching. Because as the saying goes, culture is what happens when the supervisor leaves the room. If all you have is compliance, what happens when the supervisor leaves the room is exactly what you'd expect.


If coaching works better, what happens when the supervisor leaves the room is what you'd hope.


If you're ready to see how continuous safety data powers a coaching-first approach, book a demo and we'll show you what it looks like in practice.




Frequently Asked Questions


What is the difference between coaching and compliance in workplace safety?


Compliance-based safety focuses on defining rules, training workers on those rules, auditing adherence, and disciplining non-compliance. It establishes a necessary baseline but is inherently backward-looking, binary, and impersonal. Coaching-based safety uses specific, recent events (such as near-miss video clips) to facilitate constructive conversations that build understanding, encourage problem-solving, and create ownership of safety outcomes. Coaching is forward-looking, continuous, and responsive to the specific conditions on the floor.


Why do compliance-based safety programmes plateau?


Compliance programmes are effective at establishing a baseline by formalising procedures and training. But once that baseline is achieved, the framework has limited tools for continuous improvement. Compliance is binary (met or not met), backward-looking (audits assess the past), and impersonal (rules are written for the general case). The result is a plateau where incident rates stop improving despite continued compliance activity. Progressing beyond this plateau requires a shift to coaching and proactive, data-driven safety management.


How does coaching work in a fast-paced FMCG environment?


In FMCG operations, coaching is typically embedded into existing daily rhythms rather than added as a separate activity. A supervisor uses a short, face-blurred video clip of a recent safety event from their team's work area as a discussion starter during a shift briefing or toolbox talk. The conversation is specific (this event, in this zone, yesterday), constructive (what can we do differently?), and brief (five to ten minutes). Computer vision AI generates the coaching material automatically, so the supervisor doesn't need to witness every event to coach effectively.


Can coaching replace compliance in workplace safety?


No. Compliance is the foundation: the rules, procedures, training, and audit trails are essential for legal requirements and baseline safety. Coaching builds on top of compliance by turning rules into understanding and turning data into conversations that drive continuous improvement. The most effective safety programmes combine a solid compliance foundation with a coaching-led approach to cultural change, supported by continuous data from tools like computer vision AI.


How do you measure whether coaching is working?


Leading indicator data from computer vision AI provides a direct feedback loop. After coaching on a specific issue (such as pedestrian-vehicle near misses at a particular intersection), you can track event density in that zone over the following weeks to see whether the coaching conversation translated into behaviour change. Heatmap comparisons before and after coaching interventions show whether risk is reducing, staying flat, or shifting to adjacent areas, providing evidence that traditional lagging indicators like LTIFR can't deliver at the same speed.


 
 
bottom of page