A safety incident rarely begins at the moment something goes wrong. Before an injury, equipment damage or serious incident occurs, there may be warning signs: a worker skipping a step, PPE being used incorrectly, a machine guard being bypassed or a procedure being followed differently from how it was written.

The challenge is identifying these warning signs before they become incidents. This is where behavioural observation data can make a difference. Instead of relying only on injury rates and incident reports, organisations can observe how work is actually being performed, identify recurring behaviours and use the resulting data to make informed safety improvements.

The approach is closely connected to Behaviour-Based Safety (BBS), where workplace observations are used to understand behaviours, encourage safe practices and proactively address at-risk behaviours.

Why Workplace Safety is Essential Than Incident Data 

Important measures of safety performance include incident rates, lost-time injuries and near-miss statistics. But primarily they report on what has already occurred.

According to OSHA, leading indicators are proactive and preventive measures that can reveal potential problems in a safety and health programme before they result in injuries or illnesses. OSHA recommends using leading indicators to drive change while using lagging indicators to measure effectiveness.

The need for proactive safety management is significant. The International Labour Organization estimates that 2.93 million workers die each year from work-related factors, while around 395 million workers sustain non-fatal work injuries annually. These figures reinforce an important point: organisations cannot afford to wait for incidents before identifying weaknesses in their safety systems. This is where observation data comes in.

What Is Behavioural Observation Data? 

Behavioural observation data is information collected by observing people while they perform their normal work activities. Don’t try to “catch” employees doing something wrong.

Rather, the intent is to learn about what is occurring in the workplace and why.
Observations can be made on:

  • Correct use of PPE

  • Safe lifting and material handling

  • Lockout/tagout practices

  • Work-at-height procedures

  • Safe use of tools and machinery

  • Vehicle and pedestrian movement

  • Permit-to-work practices

  • Housekeeping

  • Communication between workers

  • Compliance with established procedures

The observation should capture both safe and at-risk behaviours. For example, an observer may notice that most workers follow a machine isolation procedure correctly, but a particular step is frequently skipped during maintenance.

That observation becomes valuable data. Instead of waiting for an incident involving the machine, the organisation can investigate why the step is being missed and address the underlying issue.

From a Safety Observation to a Safety Improvement 

Just taking observations is not enough. The real value of BBS comes from turning observations into actionable insights. A practical process can look like this:

Observe → Record → Analyse → Act → Re-observe → Improve

 1. Observe What Actually Happens 

Start by observing employees during normal work. The focus should be on specific, measurable behaviours rather than personal opinions.

For example:

Instead of: “Worker was careless.”

Record: “Worker entered the designated area without wearing required eye protection.”

The second observation is specific and can be analysed.

 2. Record the Observation Consistently 

Observations should be recorded using clearly defined categories.

This might include:

  • Task being performed

  • Safe behaviour observed

  • At-risk behaviour observed

  • Frequency

  • Workplace conditions

  • Immediate feedback provided

  • Corrective action required

Consistency is important because it allows safety teams to identify trends rather than isolated events.

 3. Look for Patterns 

One observation may not tell you much. Hundreds of observations can tell a very different story. Suppose an organisation collects 500 observations over six months. The individual observations may appear unrelated. But analysis could reveal that:

  • PPE-related behaviours account for a recurring proportion of at-risk observations.

  • A particular task generates more unsafe behaviours than others.

  • One shift shows a higher frequency of procedure deviations.

  • The same unsafe practice repeatedly appears despite previous training.

Now the organisation has something it can act on. The data has moved the conversation from “someone made a mistake” to “there is a recurring risk that needs to be understood.”

Don’t Assume Every Unsafe Behaviour Is an Employee Problem 

This is one of the most important principles when analysing observation data. If employees repeatedly perform a task unsafely, it is easy to conclude that they need more training.

  1. But what if the procedure is difficult to follow?

  2. What if the correct equipment is unavailable?

  3. What if the workstation design encourages shortcuts?

  4. What if production pressures make the safe method significantly slower?

  5. What if workers were trained on a procedure that doesn’t reflect the way the task is actually performed?

These questions can uncover systemic causes behind behavioural patterns. A strong BBS programme therefore shouldn’t simply ask: “Why didn’t the worker follow the rule?”. It should also ask: “What is influencing the worker to behave this way?”. That shift can lead to much more effective corrective actions.

Common Mistakes When Using Observation Data 

 1. Treating observations as inspections 

An inspection primarily looks for hazards and compliance issues. Behavioural observation focuses on how people perform tasks and the factors influencing those behaviours. Both have value, but they serve different purposes.

 2. Focusing only on negative behaviours 

If employees only receive attention when something is wrong, BBS can quickly become associated with criticism. Recognising safe behaviours is equally important. Positive feedback can reinforce the practices organisations want to see repeated.

 3. Collecting data without analysing it 

A spreadsheet containing thousands of observations doesn’t automatically improve safety. The data needs to be reviewed for trends, recurring behaviours and areas requiring intervention.

 4. Measuring the number of observations instead of improvement 

Completing more observations can demonstrate participation, but it doesn’t necessarily demonstrate better safety performance. Organisations should also examine whether at-risk behaviours are decreasing, safe behaviours are increasing and corrective actions are producing results.

 5. Using observations to assign blame 

If employees believe observations will result in punishment, they may be less willing to participate honestly. BBS works best when observations encourage feedback, learning and improvement.

What Should Organisations Measure? 

There is no single set of metrics that works for every organisation. However, useful BBS measures can include:

  • Number of observations completed

  • Participation by employees and supervisors

  • Safe-to-at-risk behaviour trends

  • Frequency of recurring at-risk behaviours

  • Corrective actions raised

  • Corrective actions completed on time

  • Repeat observations after intervention

  • Improvement in targeted behaviours

  • Employee feedback and participation

OSHA similarly recommends tracking both leading and lagging indicators and analysing performance over time rather than relying on a single metric. This creates a more balanced view of safety performance.

Turning Data Into a Continuous Improvement Cycle 

Behavioural observation shouldn’t be treated as a one-time campaign. It works best as a continuous improvement cycle.

  • Observe what happens.

  • Analyse the information.

  • Understand why behaviours occur.

  • Act on the findings.

  • Measure whether the intervention works.

  • Repeat the process.

This approach also aligns with the broader Plan-Do-Check-Act philosophy used in health and safety management. The UK Health and Safety Executive emphasises monitoring, reporting and reviewing health and safety performance as part of effective management. Over time, this can help organisations move from simply recording safety performance to actively managing the factors that influence it.

How NIST Global Supports Behaviour-Based Safety 

Building an effective BBS programme requires more than giving employees an observation checklist. Organisations need clearly defined behaviours, trained observers, structured observation processes, meaningful data analysis and appropriate feedback mechanisms.

NIST Global’s Behaviour-Based Safety programme supports organisations in developing a systematic approach to observing workplace behaviours, identifying safe and at-risk practices, analysing behavioural trends and using the findings to support proactive safety improvement.

The focus is not simply on collecting more observations. It is about making those observations useful. Because every observation can answer a bigger question: What can we learn today that could prevent an incident tomorrow?

Key Takeaways 

  • Behavioural observation data provides an early warning system by identifying safe and at-risk behaviours before they contribute to incidents.

  • Data becomes valuable when organisations analyse patterns and underlying causes, rather than simply counting observations or blaming employees for unsafe behaviours.

  • A continuous cycle of observe, analyse, act and re-observe can help organisations strengthen safety behaviours, measure improvement and build a more proactive safety culture.

FAQs

Behavioural observation involves systematically observing how workers perform tasks and recording safe and at-risk behaviours. The information can then be analysed to identify trends and opportunities for improvement.

No. A safety inspection generally focuses on hazards, conditions and compliance, while BBS focuses specifically on workplace behaviours and the factors influencing those behaviours. Both approaches can complement each other.

Observation data can provide information about behaviours and safety activities before an injury or incident occurs. OSHA identifies leading indicators as proactive measures that can help reveal potential problems and support preventive action.

No. A balanced BBS programme observes and reinforces safe behaviours while identifying at-risk behaviours that require attention.

Organisations can enhance the effectiveness of BBS by conducting training for observers, identifying observable behaviours, collecting consistent data, conducting trend analysis and root cause analysis, and providing feedback where needed, while measuring if interventions produce a sustained improvement.

Popular Safety Training Programmes