FindRisk Logo
Near Miss Reporting: Report Example and System Setup Guide
All articles
near miss reportingnear miss reportnear miss reporting systemnear missincident reportingsafety cultureleading indicatorsISO 45001

Near Miss Reporting: Report Example and System Setup Guide

March 29, 202621 min readFindRisk Team

Near miss reporting is the practice of recording and reviewing workplace events that could have caused injury, illness or damage but did not. Each report shows a hazard, or a failed control, before anyone is hurt. The warning has already happened for free, so the fix can come before the injury does.

Many organizations only learn from events that hurt someone. A near miss reporting system gives workers a quick, blame-free way to say "that almost went wrong," gives the safety team data to find patterns, and gives management a leading indicator that moves before the injury rate does.


What Is a Near Miss? Definition and Examples

A near miss is an unplanned event at work that did not cause injury, illness or damage but had the potential to. The definitions used by the main reference bodies agree on that core:

Source Definition
ISO 45001:2018, 3.35, Note 2 An incident "where no injury and ill health occurs, but has the potential to do so, may be referred to as a 'near-miss', 'near-hit' or 'close call'."
HSE, HSG245 (UK) "An event that, while not causing harm, has the potential to cause injury or ill health."
National Safety Council, Near Miss Reporting Systems (2013, OSHA–NSC Alliance) "An unplanned event that did not result in injury, illness, or damage – but had the potential to do so."

Other names you will hear are close call, near hit, narrow escape and, for moving objects, near collision. In ISO 45001 terms, both a near miss and an accident are incidents: the standard defines an incident as an "occurrence arising out of, or in the course of, work that could or does result in injury and ill health." That matters, because every ISO 45001 requirement for incidents (Clause 10.2) applies to near misses too.

Near Miss vs Accident vs Hazard

Term What it is Example
Near miss An event happened; no one was hurt, but someone could have been A pallet falls from a rack seconds after a picker walks past
Accident An incident that causes injury or ill health (ISO 45001 calls it an incident "where injury and ill health occurs") The pallet falls and strikes the picker
Hazard Something with the potential to cause harm; no event has happened yet A damaged rack beam with no load notice
Unsafe condition A physical state of the workplace that raises risk Wet floor with no warning sign
Unsafe act A behavior that raises risk Walking under a suspended load

Hazards, unsafe acts and unsafe conditions should be reported too, but they are not near misses: nothing has happened yet. One reporting channel for all of them, with a type tag, keeps things simple for workers. Our guide to unsafe acts and unsafe conditions has 30 classified examples.

10 Near Miss Examples From the Field

  1. A reversing forklift enters a pedestrian aisle; a worker steps back just in time.
  2. A tool falls from a scaffold and lands in the barricaded zone, where no one stands.
  3. A guard is opened on a machine that was not isolated; the interlock stops it.
  4. A sling slips and the load drops where a rigger stood a minute earlier.
  5. A worker trips on an unmarked cable and catches a handrail.
  6. A solvent container overflows during decanting; the splash misses the worker's face.
  7. A gas detector alarms at a tank entrance before anyone enters.
  8. A van drives through a loading bay while a worker is behind the truck.
  9. A ladder slides on a polished floor as a worker steps onto it.
  10. A worker reaching for a panel notices exposed live parts behind a missing cover.

Why Should Near Misses Be Reported?

Near misses should be reported because they reveal the same weaknesses that cause injuries, at no human cost. The NSC's Near Miss Reporting Systems case study, developed through the OSHA–NSC Alliance, states that "history has shown repeatedly that most loss producing events (incidents), both serious and catastrophic, were preceded by warnings or near miss incidents," and that a faulty process or management system "invariably is the root cause" of the increased risk.

In practice:

  • You fix the cause before someone is hurt. The aisle layout that nearly put a picker under a forklift today is still there tomorrow.
  • Investigation is easier. HSE's HSG245 says investigating near misses "is as useful, and very much easier than investigating accidents": no one is injured, and witnesses "will be more likely to be helpful and tell the truth."
  • Luck is not a control. HSG245 notes it is "often pure luck" whether an event becomes a near miss or an accident.
  • You get data, not anecdotes. The NSC lists data for "statistical analysis, correlation studies, trending, and performance measurement" as a main benefit of a reporting system.
  • Workers take part. The NSC calls employee participation "a basic component of a successful safety management system."
  • It is a leading indicator. Near miss reports, and how fast they are closed, show the state of your controls before your injury rate moves. See our guide to leading vs lagging safety indicators.

The EU-OSHA OSHwiki article on near misses (Finnish Institute of Occupational Health, updated December 2025) reports that studies show promoting near miss reporting decreases occupational accidents. It also warns that the number of reports "should not be the key objective": the value is in what you do with them.


Is Near Miss Reporting Required by Law?

In the US and the UK there is no general duty to report every near miss to the regulator, but there are specific duties for high-potential events. In Türkiye the employer must investigate near misses and keep reports on them, and ISO 45001 treats them like any other incident.

Jurisdiction / standard What applies to near misses
United States (OSHA) The injury and illness recordkeeping rule (29 CFR 1904) covers injuries and illnesses, not near misses. Under Process Safety Management, 29 CFR 1910.119(m) requires employers to investigate "each incident which resulted in, or could reasonably have resulted in a catastrophic release," starting no later than 48 hours after the incident, and to keep the report for five years. OSHA encourages investigating close calls in general.
United Kingdom (HSE) Under RIDDOR 2013, certain dangerous occurrences, "incidents with a high potential to cause death or serious injury" listed in Schedule 2 (for example the failure of lifting equipment or plant touching overhead power lines), must be reported to HSE even when no one is hurt. Other near misses are not reportable to HSE, but HSG245 recommends investigating them.
Türkiye Law No. 6331, Article 14(1)(b): the employer must investigate events that did not cause injury or death but damaged, or had the potential to damage, workers, the workplace or work equipment, and prepare reports on them. The Risk Assessment Regulation (2012) defines a near miss event (ramak kala olay) and requires the risk assessment to be renewed, fully or in part, after one.
ISO 45001:2018 Near misses are incidents (3.35). Clause 10.2 requires the organization to react to incidents, investigate them with worker participation, determine causes, review risk assessments, take corrective action and keep documented information.

Inside the company, your own procedure sets the duty for workers. This summary is general information, not legal advice; check the rules for your sites.


Near Miss Report Example (Filled In)

A good near miss report is short, factual and specific enough for someone who was not there to act on it. The example below is fictional, written to show the level of detail that works. The fields follow OSHA's near-miss incident report form template (produced under an OSHA grant), with fields added for the follow-up.

Field Entry
Report no. NM-2026-0147
Date and time 12 November 2026, 14:20, B shift
Location Warehouse 2, aisle C4, between racks 12 and 13
Type Near miss (vehicle / pedestrian)
What happened A reach truck reversing out of aisle C4 came within about 1 m of a picker walking along the cross aisle with a hand scanner. The picker stepped back. The driver did not see the picker until after stopping.
What could have happened Picker struck by a reversing reach truck: crush injuries or fatality.
Immediate action taken Driver stopped and reported to the shift leader. Cross aisle at C4 coned off for the rest of the shift.
Conditions observed Reverse alarm hard to hear over the sortation conveyor next to C4. No floor marking separating the pedestrian route from the aisle exit. Mirror at the C4 corner missing (bracket empty).
Witnesses One picker, one driver (names held by the shift leader)
Photo 2 photos: empty mirror bracket; view from the driver's seat at the aisle exit
Reporter's suggestion Replace the mirror; mark a pedestrian crossing with a stop line for trucks at the C4 exit
Reported by Picker (named; could have been anonymous)
Received by / acknowledged Shift safety lead, 12 Nov, 15:05
Potential severity High potential (worst credible outcome: fatality) → full investigation
Actions 1) Replace mirror, maintenance, 13 Nov. 2) Floor marking and stop line, facilities, 20 Nov. 3) Review all aisle exits beside the conveyor, safety lead, 30 Nov. 4) Check the reverse alarm against conveyor noise, engineering, 30 Nov.
Feedback to reporter 14 Nov: picker told what was found and what will change; shared at the B-shift toolbox talk
Closed Pending verification of action 3

Good vs Poor Entries

The same event, written two ways:

Field Poor entry Good entry Why it matters
What happened "Forklift nearly hit someone." Who, where, what each person was doing, how close it came The investigator can rebuild the event without guessing
What could have happened "Could have been bad." "Picker struck by reversing reach truck: crush injury or fatality" Sets the investigation depth
Conditions "Driver not paying attention." Alarm inaudible, no floor marking, mirror missing Points to fixable conditions, not blame
Location "Warehouse" "Warehouse 2, aisle C4, racks 12–13" Allows trend mapping by area
Suggestion Blank "Replace mirror, add crossing and stop line" The person on the floor often knows the fix
Feedback Not recorded Date and how the reporter was told Without feedback, reporting stops

The good report leaves out blame: causes come later, in the investigation. For injury events, see our guide on how to write a workplace incident report.


How to Set Up a Near Miss Reporting System: 8 Steps

A near miss reporting system is the policy, channels, people and follow-up process that turn individual reports into corrective actions and trend data. The steps below follow the NSC's best practices and the OSHwiki guidance, with the order we would use on a site starting from zero.

Step 1: Write a Short, Non-Punitive Policy

State on one page what to report, how, and what happens next, and say clearly that reporting will not lead to discipline. The NSC case study says the system "needs to be non-punitive and, if desired by the person reporting, anonymous," with a policy communicated "with the backing of senior management." If the policy keeps an exception for deliberate, reckless violations, write it down and keep it narrow.

Step 2: Define What to Report, With Examples

Workers cannot report what they do not recognize. Give them site-specific examples, like the ten above, and accept hazards through the same channel.

Step 3: Make Reporting Take Minutes, Not a Form Day

Offer more than one channel: a phone form or QR code, a paper card, or a verbal report the supervisor writes down. Ask only what the reporter knows: what, where, when, what could have happened, and a photo. Do not ask them for the root cause.

Step 4: Name the Receiver and Set Response Targets

The OSHwiki article notes that a successful process always states who receives the reports and who handles them, and that reports "should be handled quite soon after reception." Name a receiver for each area and a back-up. Set your own targets, for example acknowledgement within one working day and a decision on the investigation level within a few days.

Step 5: Triage by Potential Severity

Rank each report by its worst credible outcome, not by what actually happened. HSG245 says it is "the potential consequences and the likelihood of the adverse event recurring" that should determine the level of investigation. A dropped spanner in a barricaded zone may need a quick fix; a load swinging over a walkway needs a full investigation. Many companies flag the second kind as a high-potential (HiPo) near miss.

Step 6: Investigate in Proportion and Assign Actions

Low-potential reports get a short check and a fix by the area owner. High-potential reports get a full investigation; our incident investigation guide covers the process. Every action needs an owner, a due date and a check that it worked. Update the risk assessment too: ISO 45001 Clause 10.2 asks for it after incidents, and Türkiye's Risk Assessment Regulation requires it after a near miss.

Step 7: Close the Loop With the Reporter

Tell the reporter what was found and what changed. OSHwiki notes that workers who see no benefit from reporting become unwilling to report. Share good examples at toolbox talks.

Step 8: Track Trends and Review the System

Review reports monthly by area, task, hazard type and potential severity. Repeat reports on one hazard mean the fix did not work. Put near miss data next to injury rates in management review, and never set a target for the number of reports: it produces trivial reports, not fixes.


Why Near Misses Go Unreported

Near misses go unreported mainly because people fear blame, see no point, or find reporting too much effort. The NSC case study names fear of blame as one problem companies must overcome and calls for a reporting process that is "as easy as possible." The usual barriers:

  1. Fear of blame. If a report leads to "who made the mistake?", the next one is not filed.
  2. "Nothing happened." An event without injury does not feel reportable.
  3. Too much effort. A long form at the end of a shift will not be filled in.
  4. No feedback. Reports that vanish teach people that reporting is pointless.
  5. Normalized risk. A weekly close call starts to look like part of the job.
  6. Incentives that punish reporting. The NSC gives a clear example of a poor incentive: rewarding supervisors and managers on OSHA recordable rates, which "has been shown to suppress reporting." Recognize people for reporting hazards instead.

A behavior-based observation program can complement reporting, as long as it stays blame-free; see our safety observation guide.


Investigating a Near Miss: A 5 Whys Example

A near miss investigation asks why the event was possible, not who made a mistake. For low- and medium-potential events, the 5 Whys technique is often enough. A fictional example:

Near miss: A worker nearly received a chemical splash while decanting a solvent.

  1. Why? The container was overfilled.
  2. Why? The worker could not see the fill level.
  3. Why? The container was opaque and had no level mark.
  4. Why? The container specification did not require a visible level.
  5. Why? The specification had not been reviewed when the solvent type changed.

Root cause: No review of the container specification when the process changed, not "worker carelessness." Corrective action: Update the specification to require level-visible or marked containers, and add container review to the change procedure.

For high-potential near misses, use a structured method such as a fishbone diagram, fault tree or SCAT. Our guide to root cause analysis in workplace safety compares five methods with a worked example.


How to Measure Near Miss Reporting

Measure whether reports lead to action, not just how many arrive. Useful indicators:

Indicator What it tells you
Near miss reports per month, by area Where reporting is happening, and where it is silent
Share of high-potential reports Whether serious precursors are being caught
Median days from report to closed action Whether the system responds
Repeat reports on the same hazard Whether fixes work
Ratio of near miss and hazard reports to injury and damage events How well field reporting works on each site

The last ratio comes from Dave Rebbitt's 2014 Professional Safety article "Pyramid Power," which says it helps show how well field reporting works at different facilities or sites. Rising report numbers after a campaign usually mean better reporting, not more risk; OSHwiki notes that report numbers rise in every workplace after a reporting campaign.


The Safety Pyramid: What Heinrich and Bird Actually Found

The safety pyramid is the idea that serious injuries sit on top of a much larger number of minor injuries and no-injury events. It is a common argument for near miss reporting, and its numbers are often misquoted.

Heinrich 1931 (1:29:300) and Bird and Germain 1969 (1:10:30:600) safety pyramids, with what each ratio measured and its limits

Study Ratio Levels
Heinrich, Industrial Accident Prevention, 1931 1 : 29 : 300 Major or lost-time injury : minor injuries : no-injury accidents
Bird and Germain, 1969 1 : 10 : 30 : 600 Major injury : minor injuries : property damage : no-injury incidents

Heinrich (1931). H.W. Heinrich, who worked for the Travelers Insurance Company, wrote that "in a unit group of 330 accidents, 300 result in no injuries, 29 in minor injuries, and 1 in a major or lost-time case." He did not explain where the no-injury data came from until the fourth edition (1959), which mentions "a study of over 5,000 cases." The often-quoted figure of 75,000 cases belongs to a different analysis, his 88-10-2 split of accident causes (12,000 insurance claim files and 63,000 plant records), not to the pyramid. Heinrich's original data are not available, so no one has been able to check either result. His better-known model, the five-domino accident sequence, is covered in our guide to Heinrich's domino theory.

Bird (1969). Frank Bird led a study at the Insurance Company of North America that analyzed 1,753,498 accidents reported by 297 companies. It produced the 1:10:30:600 ratio. Note that the top of Bird's pyramid is a major injury, not a fatality.

What the critics showed. In "Reviewing Heinrich: Dislodging Two Myths" (Professional Safety, October 2011), Fred Manuele argues that reducing the number of minor incidents does not reduce serious injuries by the same amount. He cites US workers' compensation data in which claim frequency fell sharply while serious claims fell much less. Later studies found the ratios vary widely between industries and companies. The Baker Panel report (2007) on the 2005 BP Texas City refinery explosion reached a similar conclusion: BP "mistakenly interpreted improving personal injury rates as an indication of acceptable process safety performance."

What still holds. Events without injury carry information about failed controls, which is why reporting them is worth it. But the ratios are not a law of nature, so do not promise that fewer near misses mean fewer fatalities. Find the near misses whose worst outcome is serious and fix those first; that is what the HiPo triage in Step 5 is for.


How FindRisk Supports Near Miss Follow-Up

FindRisk helps with the work that follows a near miss report: checking the area, rating the risk and documenting the fix. On site, a safety lead can photograph the scene and mark the hazard on the photo (the empty mirror bracket, the missing floor marking), and the app's AI analysis can point out other hazards in the same picture.

The hazard can then be scored with the Fine-Kinney method, which gives a consistent way to decide whether a near miss is high-potential. For a follow-up inspection of every similar location, FindRisk can generate an inspection checklist with AI, and it produces a professional report in seconds to share with management and the people who reported. FindRisk runs on iOS and Android.


Frequently Asked Questions

Do you have to report a near miss at work?

Usually yes, under your employer's procedure. OSHA's template near miss form, for example, asks all employees to "report and correct any of these potential hazards immediately." Reporting to a regulator is required only for specific events, such as RIDDOR dangerous occurrences in the UK. In Türkiye, Law No. 6331 Article 14 requires the employer to investigate and report on events with harm potential.

Is near miss reporting non-punitive?

It should be. The NSC's guidance says a near miss reporting system "needs to be non-punitive" and should allow anonymity if the reporter wants it. Where a policy keeps an exception, it should cover only deliberate or reckless violations. Without a non-punitive rule that people trust, workers stop reporting and the hazards stay hidden.

How many near misses happen before a serious accident?

There is no reliable number. Heinrich's 1931 ratio was 300 no-injury accidents for every major injury, and Bird's 1969 study gave 600 no-injury incidents for every major injury, not for every fatality. Both are averages from insurance data, not predictions for one site, and later research shows ratios vary widely between industries. The useful message is that warnings come first, not the exact count.

What is a high-potential (HiPo) near miss?

A high-potential near miss is one whose worst credible outcome is a fatality or life-changing injury, even though no one was hurt. Examples are a dropped load landing where a person had just stood, or a machine starting during maintenance. HiPo near misses should get a full investigation, because HSE's HSG245 says the potential consequences, not the actual ones, should set the investigation depth.

Should near miss reports be anonymous?

Offer anonymity as an option, not the only route. Anonymous reporting helps where trust is low, and the NSC recommends allowing it if the reporter wants it. The cost is follow-up: you cannot ask an anonymous reporter for details or tell them what changed. Named reporting under a credible non-punitive policy gives better investigations.

What is the difference between a near miss and a hazard?

A hazard is a condition with the potential to cause harm, such as a damaged rack beam. A near miss is an event that has already happened and could have caused harm, such as a pallet falling from that beam into an empty aisle. Both should be reported, often through the same channel, but a near miss shows that a control has already failed once.


Conclusion

Near miss reporting turns close calls into fixes. A working system needs a short non-punitive policy, a reporting channel that takes minutes, a named receiver, triage by worst credible outcome, proportionate investigation, and feedback to every reporter. The filled report above shows the level of detail that lets someone act.

Use the safety pyramid for what it can carry: a reminder that harm usually has warnings. Do not use its ratios as a forecast. Put your effort into the high-potential near misses, close their actions, and update the risk assessment each time a control fails.

Download FindRisk to photograph and mark the hazard behind a near miss, score it with Fine-Kinney, and send a clear report to the people who need to act.

Try FindRisk

Ready to modernize your safety workflow?

Conduct AI-powered risk assessments, generate reports instantly, and keep your team safe — anywhere, anytime.