Near-Miss Reporting
Also called: near miss report · close call reporting · good catch
A near-miss is an event that could have caused injury or damage but didn't — a slip that didn't fall, a load that shifted but didn't drop, a machine that stopped short. Near-miss reporting is the systematic capture and analysis of these events as leading indicators of future incidents. The data is powerful because it is predictive; the program is hard because frontline workers don't naturally report events where nothing bad happened. Mature safety programs measure near-miss rate alongside incident rate and treat a high near-miss rate as a sign of reporting health, not of unsafe conditions.
Why it matters
The Heinrich ratio (1 serious injury per 29 minor injuries per 300 near-misses) remains directionally true even if the exact numbers vary. Near-misses are the leading indicator that lets safety leaders see the hazard pattern before the injury happens. A site with 50 near-misses reported per month and no injuries is probably safer than a site with 2 near- misses and one injury, because the first site is catching conditions the second site is ignoring. The strategic value of near-miss data compounds over time; the challenge is making the reporting easy enough that workers actually do it.
How it works
Take a 1,200-person food manufacturing plant. Near- miss reporting runs through the employee app: worker sees a condition, taps "report near miss," selects a category (slip/trip, equipment, PPE, chemical), adds a photo and a sentence of context. Submission takes under 60 seconds. Supervisor gets a notification within 2 minutes; investigation and closure happen within 72 hours. Monthly, the safety lead reviews patterns: this area had six PPE-related near-misses; what's changed? Annual report to the safety committee correlates near-miss trends with incident counts.
The operator's truth
The biggest predictor of near-miss reporting rate is not the reporting tool — it is the frontline worker's belief about whether reporting gets them in trouble. If the last person who reported a near-miss got questioned about "why were you doing that in the first place," the reporting rate collapses within weeks. The programs that produce healthy data separate the report from the worker performance assessment and publicly celebrate good catches. The programs that don't produce near-miss numbers that look great on paper because nobody reports anything.
Industry lens
In manufacturing and construction, near-miss programs are mature and regulated; reporting rates are a standard safety KPI.
In healthcare, the equivalent is incident reporting including close calls (wrong medication caught before administration, patient identification errors caught before procedure). The culture challenge is the same — blameless reporting vs punitive follow-up.
In retail and hospitality, near-miss reporting is uneven. Slips and trips get reported; ergonomic near- misses often don't. Programs that include customer- safety near-misses (e.g., close-call on wet floor) generate stronger data.
In the AI era (2026+)
AI changes near-miss reporting in two ways in 2026. First, vision-based systems detect some near-miss conditions (PPE violations, proximity to moving equipment) without the worker having to report them — the data is captured automatically. Second, the AI assistant on the frontline app makes reporting conversational: the worker says what happened, the agent categorizes, asks two follow-up questions, and submits a structured report. Reporting friction drops, reporting rates rise, and safety leads get higher- quality leading-indicator data.
Common pitfalls
- Punitive culture. Workers who feel reporting leads to discipline stop reporting. Near-miss rate will go down, injury rate will go up.
- Complex reporting forms. A 15-field form kills reporting. Keep submission under 60 seconds; ask clarifying questions later.
- No visible follow-up. If reports disappear into a black hole, reporting fades. Close-the-loop is the retention driver.
- Reporting-rate as the only KPI. Gaming the metric (reporting trivial events) destroys the signal. Pair with quality measures.
- Ignoring the pattern. Individual reports matter less than patterns across reports. Monthly pattern review is where the safety value lives.
