New Behavioral Observation Lead in 75 Days: Build Conversations That Improve Risk Perception

A new behavioral observation lead often receives a form, a calendar, and an unspoken expectation to make unsafe choices disappear. That expectation creates the wrong job. In the first 75 days, the lead should build a reliable conversation system that helps people see exposure, question shortcuts, and improve the conditions around the work.

The central test is not how many observations are submitted. It is whether a conversation changes what the worker, supervisor, or process owner does next. Behavioral observation becomes useful when it produces clearer risk perception and a visible response, rather than a score that rewards activity.

What a behavioral observation lead needs to understand before starting

The role sits between individual action and the conditions that shape it. A worker may take a shortcut because of haste, improvisation, exception, presumption, or overconfidence, yet the observation lead should not assume that intent explains the whole event. The task may be poorly sequenced, the control may be awkward, or the production signal may punish a pause.

Andreza Araujo makes this distinction in Far Beyond Zero, where behavior is treated as a response to context rather than proof that a person is the weak link. That position changes the opening question. Instead of asking, “Why did you break the rule?”, the lead asks, “What made this choice appear workable in the moment, and what control would make the safer choice easier?”

The lead is not replacing the supervisor, investigating incidents, or diagnosing a worker. The lead creates a disciplined feedback channel that helps the responsible owner see where the work and the written expectation diverge.

Days 1 to 15: establish the purpose and boundaries

The first two weeks should produce a one-page operating agreement. It should state what the program is for, what it is not for, who receives the information, and how urgent exposure is escalated. If workers cannot predict what happens after they speak, participation will become cautious and superficial.

Define three boundaries before collecting observations. First, immediate danger is acted on at once and does not wait for a weekly review. Second, ordinary observations are used to improve work and controls, not to rank individuals. Third, repeated deliberate disregard for a clear control follows the normal supervisory process, because a learning conversation cannot replace accountability.

Invite operations, maintenance, frontline supervision, and EHS to review the agreement. Their involvement matters because each function sees a different part of the exposure, and the lead needs enough authority to route a concern to the person who can change it.

Days 16 to 30: teach observation that can be acted on

A useful observation has four parts. It identifies the task and exposure, records what was actually happening, explores why the choice made sense at that moment, and names the next action with an owner. A form that captures only safe or unsafe behavior gives the organization a label without a decision.

Train observers to describe behavior without turning the conversation into a verdict. “The guard was removed while the conveyor was moving” is more useful than “the operator was careless,” because the first statement invites questions about access, interlocks, production pressure, and supervision. Those questions reveal which control needs attention.

Run practice sessions with ordinary work, not staged perfection. Ask observers to notice body position, line of fire, hand placement, energy isolation, communication, and the point at which a person must choose between speed and control. The exercise should end with one verified improvement, even when the improvement is small.

Days 31 to 45: move from observation to conversation

The conversation should begin with respect for the work. Ask what the person is doing, what could change the exposure, and which part of the method is hardest to follow. A question that recognizes competence lowers defensiveness, while a lecture about compliance often closes the information channel before the real problem appears.

Use the OPA rhythm, observe, plan, act, as a simple structure for the discussion. Observe the task and the conditions around it. Plan the safest practical adjustment with the person who performs the work. Act by making the change, assigning an owner, or escalating when the lead lacks authority.

Andreza’s Safety Culture: From Theory to Practice describes behavioral observation as a structured conversation of active care rather than the completion of a punitive form. The lead should make that principle visible by thanking people for raising inconvenient information and by returning with an update after the concern is routed.

Days 46 to 60: connect behavior with critical controls

By the second month, individual observations should be grouped around exposures and controls. Three comments about rushed access, for example, may point to a recurring isolation problem, an unreliable permit boundary, or a schedule that leaves no recovery time between tasks.

Choose two or three high-consequence exposures and map the behaviors that show whether the critical controls are present. In lifting work, the indicators may include exclusion-zone discipline, load-path communication, and equipment inspection. In maintenance, they may include isolation verification, stored-energy checks, and clear authority to stop the job.

Do not use the observation program to declare that a control is effective. Compare the reported behavior with field evidence, supervisor checks, and action closure. A repeated observation is not automatically a people problem. It may be a control-design problem whose visible symptom happens to be human behavior.

Days 61 to 75: build the operating rhythm

The final two weeks should establish a monthly rhythm that survives the lead’s absence. Supervisors review themes during their normal work planning, operations owners close control actions, and EHS checks whether the response changed the exposure. The behavioral observation lead coordinates the system instead of becoming its only source of energy.

Use a small dashboard with five measures: observations by exposure, percentage with a named owner, time to first response, percentage of actions verified in the field, and recurring themes that remain open. These measures describe response quality. They do not prove that injury risk has disappeared, so they must remain connected to control verification and incident learning.

Set a rule for stale actions. When an item remains open beyond its agreed date, the owner must either complete it, set an interim boundary, or escalate the decision. That rule prevents the program from becoming a reporting machine that documents the same weakness every month.

Common mistakes that weaken the first 75 days

The first mistake is counting volume as success. High submission numbers can reflect enthusiasm, fear, or a campaign, but they reveal little about whether the organization corrected the exposure. The second mistake is asking observers to fix what only engineering, staffing, procurement, or operations can change.

The third mistake is treating every deviation as a character flaw. James Reason’s work on latent failures remains useful here because the visible action may be the last point in a chain of design, supervision, communication, and workload conditions. The conversation should preserve accountability while still testing the system around the person.

The fourth mistake is closing the loop privately. When workers report a concern and never see what happened, they learn that speaking up consumes effort without changing work. A short update, even when the answer is delayed, protects trust better than silence.

What success looks like after day 75

After 75 days, a credible program has fewer anonymous complaints about “unsafe behavior” and more precise conversations about exposure, control, and ownership. Supervisors can explain what they changed, workers can describe how concerns are handled, and the lead can show which recurring themes reached the function with authority to act.

The strongest sign is not a perfect observation score. It is a team that can interrupt a task without turning the interruption into a personal dispute, because the organization has made risk discussion part of normal work. That capability supports physical safety and strengthens the culture in which people decide under pressure.

Resources to deepen the role

Read Safety Culture: From Theory to Practice for the connection between visible decisions and daily behavior, and consult 100 Safety Objections when common operational responses make a conversation difficult. Both resources support a practical view of safety in which clarity, care, and accountability must work together.

Keep the role close to the field. A behavioral observation lead who spends all day cleaning data will miss the conditions that give the data meaning, while a lead who stays in the work area can help the organization turn a report into a safer decision.

Explore Andreza Araujo’s safety culture resources to continue building the leadership and dialogue skills behind reliable prevention.