5 Myths About Leading Indicators That Boards Still Believe

The International Labour Organization estimated in 2023 that nearly 3 million people die each year from work-related accidents and diseases, yet many boards still read leading indicators as if a rising count automatically meant stronger prevention. It does not. A count can go up while the work stays unchanged, which is why the real question is whether the metric changed a decision before harm appeared.

Leading indicators are only useful when they reveal whether control health, decision quality, and exposure management are improving in the field. If a metric only rewards activity, it may look busy, yet it tells leaders nothing about whether the risk moved, whether the control held, or whether the team learned anything useful from the last signal.

Across 25+ years leading EHS at multinationals, Andreza Araujo has seen the same error repeat in different sectors. Leaders trust the number they can display, even when the number no longer describes the work. In more than 250 cultural transformation projects supported by Andreza Araujo, the strongest board conversations started when someone asked what changed in the field, not what changed on the slide.

That is the thesis of this article. Leading indicators fail when they measure motion instead of decision change. Andreza Araujo's books, especially Safety Culture: From Theory to Practice and A Ilusão da Conformidade, make the same point from different angles, because culture and compliance both drift when leaders confuse paperwork with control.

Key Takeaways

• Leading indicators matter when they change a decision, not when they add another line to a dashboard.
• A green chart can still sit on top of stale data, weak controls, or a task nobody has changed in the field.
• Boards should read leading indicators beside control verification, serious exposure, and lagging outcomes.
• EHS cannot own every signal alone, because operations owns the work and therefore owns the exposure.
• Metrics need freshness rules and expiry dates, or they quietly become furniture.

Why these myths cost dearly

Leading indicators are not a mood board. They are a governance device that should tell leaders what the work is doing before the injury record catches up. That matters because boards do not need more comfort. They need a shorter path from weak signal to corrected decision.

James Reason remains useful here because latent conditions can sit in planning, supervision, maintenance, or design long before the visible event happens. Patrick Hudson's maturity model adds another useful lens, since maturity is a trajectory of better decisions, not a badge that stays bright forever.

The market often sells leading indicators as if the number itself were the control. It is not. A metric only matters when someone with authority changes the job, the verification routine, or the escalation path after reading it. That is why the myths below are not cosmetic errors. They are decision errors.

Myth 1: More indicators always make the dashboard better

This myth sounds sensible because more data feels safer than less data. Boards see more charts, more colors, and more trend lines, then assume they are getting a fuller picture. In reality, a crowded dashboard often hides the few signals that deserve attention.

The trap is cognitive overload. If the board must scan fifteen measures to find the one that changed a critical control, the meeting starts managing the report instead of the risk. A long list can also create false equality, where a weak housekeeping count sits beside a serious exposure signal as if both carry the same weight.

Andreza Araujo's safety culture work points in the opposite direction. The strongest leaders use fewer measures, but each one has a clear owner, a clear review cadence, and a direct link to a field decision. If you want a practical filter, compare the metric with safety metric freshness and ask whether it still reflects the current work.

What to do instead is simple. Keep the board set small, tie each indicator to one risk family, and retire any metric that cannot change a control, an approval, or a stop decision. If the measure cannot alter behavior, it is decoration, not governance.

Myth 2: A rising count means prevention is improving

This myth survives because activity is visible. More observations, more audits, more walkthroughs, and more meetings feel like progress, especially when the graph moves upward in a clean line. But an increasing count can also mean the team is counting more motion without changing the underlying exposure.

A better leading indicator asks for a decision outcome. Did the issue stop the task, change the sequence, trigger a design request, or force an escalation? If the answer is no, the metric may be telling you that the reporting habit improved, while the work itself stayed the same.

In more than 250 cultural transformation projects, Andreza Araujo has seen this pattern often enough to trust the warning. Activity metrics rise fastest in organizations that are good at reporting themselves. Field change rises only when the leader is willing to make the job different tomorrow. The article on leading indicator quality audit shows how to test that difference.

The practical rule is to measure decision conversion, not just activity volume. If a signal did not cause a field check, a control change, or an escalation, the count is only noise with a deadline.

Myth 3: Low TRIR proves the leading indicators are working

Low TRIR is pleasant to look at, and that is exactly why it can mislead. Boards often assume that a clean lagging record proves the prevention system is healthy, but TRIR only tells you what already crossed the reporting threshold. It cannot prove that the serious exposure was reduced, detected, or even visible.

This is where Bird and Heinrich remain useful as historical references, because they help leaders remember that minor events and precursor conditions matter. Their models do not mean every small event predicts every serious event in the same way. They do mean that a low injury count is not enough to guarantee control health in high energy work.

That is why leading indicators should sit beside control verification and serious exposure review. If a board wants to understand the risk picture, it should compare the lagging record with control health, TRIR, and SIF exposure, because one number cannot do all three jobs.

The response is not to throw TRIR away. The response is to stop treating it as proof of prevention. A low rate can coexist with weak barriers, late reporting, or a work system that only looks good because luck has not failed yet.

Myth 4: EHS should own the indicators alone

This myth is common because EHS often builds the dashboard and presents it to the board. That arrangement makes sense administratively, but it becomes a trap when the organization forgets that operations owns the work, maintenance owns the condition of the plant, and line leaders own the decisions that shape exposure every day.

If EHS owns every signal, the metric becomes detached from the people who can actually change it. The board then sees a polished report while the production, maintenance, and scheduling choices that created the signal stay outside the room. In practice, the indicator improves only when the owner of the work owns the response.

Across 25+ years of executive EHS work, Andreza Araujo has repeatedly seen stronger results when line managers could answer one question without help: what changed because of the metric? That ownership model is also consistent with Safety Culture: From Theory to Practice, because culture shows up where power and responsibility meet.

Use a shared ownership rule. EHS defines the measure, operations owns the action, and the board reviews whether the action changed the exposure. If those roles are blurred, the metric becomes a reporting ritual, which is the exact opposite of a leading indicator.

Myth 5: Once a metric is green, it stays meaningful

Green has a calming effect, which is why this myth survives. A green line suggests stability, yet many leading indicators decay as soon as the work changes. New contractors arrive, staffing shifts, a permit process changes, a control is bypassed, or the job becomes routine enough that the original signal no longer reflects reality.

Patrick Hudson's maturity model helps here because maturity is not a permanent state. It can move forward, stall, or slide back if the organization stops verifying what the number really means. A green metric without a freshness rule is only a historical claim.

This is where board governance becomes practical. A serious review should ask when the measure was last verified, what would invalidate it, and who would notice if the signal had quietly aged out. The article on dashboard latency is useful because stale measures delay decisions even when the chart still looks active.

The fix is to give every leading indicator an expiry condition. If the context changes enough to alter the signal, the board should stop trusting the old number until someone proves that the measure still fits the work.

What the board should review instead

A useful board pack should compare signal type, decision value, and what each measure can hide. That comparison keeps the board from rewarding the easiest number to display.

Each month that the board reviews activity without decision change, the dashboard looks more active while the work system keeps its old habits.

What to do now

1. Pick five signals and ask whether each one changed a decision in the last 30 days. If it did not, demote it or retire it.
2. Put one operational owner next to each measure, because the owner of the work should own the response to the signal.
3. Add a freshness rule. A green metric that has not been verified recently should not sit in the board pack as if it were current.
4. Review leading indicators beside control verification and lagging outcomes, so the board can see whether the signal is proving anything in the field.

If you want a cleaner executive rhythm, use the next board review to challenge one metric, one owner, and one decision. That is usually enough to expose whether the dashboard is describing the work or only decorating it.

FAQ

What makes a leading indicator useful?

A leading indicator is useful when it gives leaders an earlier read on control health or exposure and then changes a decision. If it only counts activity, it may be busy, but it is not leading anything.

Can a green metric still be misleading?

Yes. A green metric can be stale, detached from the current work, or too narrow to reflect serious exposure. The color only matters when someone verifies that the number still matches the field condition.

Why do boards prefer activity counts?

Activity counts are easy to explain and easy to trend. The problem is that they often reward motion, not risk reduction, which makes them comforting but weak as governance signals.

How should EHS and operations share ownership?

EHS should design the measure and help interpret it, while operations should own the response because operations controls the work. If the response stays in EHS, the metric is usually reporting, not management.

What is the fastest way to fix a stale dashboard?

Remove any signal that has not changed a decision, add an owner to each remaining indicator, and give every measure a freshness rule. That combination usually does more than adding another chart.

Headline Podcast exists for conversations that turn metrics into decisions. If your board still trusts the number more than the work, bring the next dashboard into the room and test it against field reality at Headline Podcast.