Benchmarks tab: peer-relative safety standing

What the Benchmarks tab shows

The Benchmarks tab is the full peer comparison for Safety: one table, six metrics, your number next to an anonymized peer number, with a delta that shows which direction you are ahead or behind. Where the Standing tab folds a peer line into four operational cards (training depth, policy currency, insurance, OSHA 300 log), Benchmarks is the dedicated readout built for one question: how does my safety program compare to the network on the metrics that actually move risk and cost.

Verinode does not grade you against the network or decide what "good" looks like here. It computes your own value from the incidents, certifications, and policies already on file, resolves an anonymized peer figure from operators contributing comparable data, and lays the two numbers side by side. You decide what the gap means and what, if anything, to do about it. No individual peer operator's data is ever exposed, benchmark values are drawn from a cohort, never from any single operator's record, and this data is never sold to carriers.

Where to find it

Open Safety from the sidebar at iq.verinode.ai/safety. The Safety card slider opens with five tabs across the top: Findings, All Safety, Standing, Exposure, and Benchmarks. Benchmarks is the fifth and last tab. Its header reads "Benchmarks," and its subtitle names the cohort directly once one exists for at least one metric: "Peer-relative standing, cohort of [N] operators." Until a cohort resolves anywhere, the subtitle instead reads the generic "Peer-relative standing across the operator network."

The six-metric table

The table has four columns: Metric, You, Peer, and Delta. Each row is one safety metric, your value on the left, the peer figure beside it, and the difference between the two on the right.

| Metric | What it means | Direction | |---|---|---| | Incident rate (per 100k hrs) | Your OSHA-recordable incidents so far this calendar year, scaled to a standard 100,000-hour base | Lower is better | | Days since last recordable | Days since your most recent OSHA-recordable incident | Higher is better | | Near-miss ratio | Share of your reported safety events that were near misses rather than recordable injuries | Higher is better | | Training compliance | Share of certifications on file that are active or expiring soon, rather than lapsed | Higher is better | | Corrective-action closure days | Average days between an incident being logged and its corrective action being closed | Lower is better | | Policy review currency | Share of your active safety policies still inside their review cycle | Higher is better |

How each is read.

Incident rate (per 100k hrs). Verinode counts the OSHA-recordable incidents logged since the start of the calendar year, then scales that count to what it would look like across 100,000 hours worked, the standard basis safety metrics use so a five-person crew and a fifty-person crew read on the same scale. The hours side of the math assumes 2,000 worked hours per active team member per year (a standard full-time-equivalent assumption), so this figure is a working estimate rather than a count built from timesheets. It shows as a plain number to two decimal places.

Days since last recordable. A running count from your most recent OSHA-recordable incident to today. If you have never logged a recordable incident, this reads as a full year (365 days) rather than zero, since there is no incident to count from. Shown as days.

Near-miss ratio. Near misses divided by (near misses plus recordable incidents), shown as a percentage. A higher ratio means more of what your team reports are near misses caught before anyone was hurt, rather than injuries that already happened, generally the healthier pattern for a reporting culture. If you have logged neither near misses nor recordables, this reads 0%.

Training compliance. The share of certifications on file, across your whole team, that carry an active or expiring-soon status rather than a lapsed one. This is a simpler read than the certification-by-certification detail on the Certifications section (see certifications and peer position), which weighs specific certifications against the specific job types you run. Here it is one blended percentage across every certification you have on file. With no certifications on file at all, this reads 100%, there is nothing lapsed because there is nothing recorded yet.

Corrective-action closure days. For incidents where a corrective action has actually been marked closed, the average number of days from when the incident was logged to when the corrective action closed. Only closed corrective actions count toward the average, so this figure reads as a dash until you have closed at least one.

Policy review currency. The share of your active safety policies whose review date has not yet passed. With no active policies on file, this reads 100%, again because nothing has drifted out of cycle when nothing is on record yet.

Reading You, Peer, and Delta. The You column is always your own computed value (a dash only appears for corrective-action closure days before you have a closed corrective action). The Peer column shows a dash whenever the network does not yet have a comparable cohort for that specific metric, some metrics will resolve a peer figure while others on the same row of your account still show a dash, that is expected and not an error. Delta is your value minus the peer value, and only appears when both sides have a number. It is colored to match the metric's own direction: green when your number sits on the better side of the peer figure for that metric (lower for incident rate and closure days, higher for the rest), red when it sits on the worse side.

Cohort sizing

Each of the six metrics resolves its peer figure independently, and each pulls from a cohort scoped as tightly as the network allows while still protecting anonymity: your state's operators first, and only when there is not yet a comparable cohort at that level does it broaden to the full national network. That is why one metric on the table might show a state-level peer figure while another on the same row falls back to national, the two draw from different pools depending on how much comparable data exists for that specific metric.

Verinode only shows a peer figure at all once enough distinct operators are contributing comparable data on that metric to protect anyone's individual data from being identifiable. Until that bar is met for a given metric, that row's Peer and Delta columns simply read as a dash, there is no partial or approximate figure shown in its place. As more operators contribute safety data, cohorts fill in over time and dashes convert to numbers on their own, nothing you need to trigger manually.

The OSHA-floor caveat

A caption line under the table reads: "Peer-relative when a cohort exists on the metric; absolute OSHA regulatory thresholds remain the floor regardless."

This is the single most important thing to hold onto when reading this table. The six rows above are all relative, they tell you how your numbers compare to other operators, not what is actually required of you. Comparing well against the peer network on incident rate or training compliance does not change what OSHA recordkeeping and posting rules require, and it does not change what your insurance carrier's workers' compensation pricing is built on. An operator who looks strong on every peer row here can still have an open recordability obligation, a missed OSHA Form 300A filing window, or an Experience Modification Rate (EMR) that is materially above the industry baseline, none of which the peer comparison alone would tell you. Those absolute, regulatory-anchored figures live on the Standing tab (see Standing: where you stand vs OSHA and peers) and in the Findings tab's OSHA-driven checks (see Findings: safety tips and decisions). Read this table as context on top of the regulatory floor, never as a replacement for it.

Empty and loading states

When the Benchmarks tab first opens, before the peer benchmark request has returned, it shows the same header and subtitle as always ("Peer-relative standing across the operator network") with the table replaced by a single italic line: "Peer benchmarks loading…" If the request behind that line does not come back successfully, the tab is left showing that same loading line rather than an error message, since there is nothing else to fall back to until the request succeeds. This is a rare case; on a normal load, the table renders within the same round trip as the rest of the Safety slider.

How to use it

1. 1Scan the Delta column first. A red delta means the peer network is meaningfully ahead of you on that specific metric; that is where a decision is most likely to be worth making.
2. 2For any row still showing a dash in Peer, do not read that as "no peers" permanently, it means the network has not yet built a big-enough anonymous cohort on that specific metric. Check back as more operators contribute safety data.
3. 3Cross-check a red delta against the Standing tab before acting. A weak incident rate delta paired with a climbing EMR (Standing tab) is a different, more urgent conversation than the same delta sitting next to a clean OSHA 300 log.
4. 4If corrective-action closure days shows a dash, that is your own signal to close out an open corrective action, not a peer gap, you need at least one closed action before this row has anything to compare.

Best-practice example

Say your incident rate reads ahead of the peer figure (a green delta), but training compliance shows a red delta twelve points behind. The incident rate result is worth noting but not acting on today, it is already trending the right way. The training compliance gap is the one to open: check the Standing tab's Training Depth card for which certifications are actually expiring, and the Exposure tab for the cert wall over the next 30 days (see Exposure: cert wall, insurance gaps, and past-due corrective actions). The Benchmarks table told you where you stand against the network; the other tabs tell you exactly which record to act on first.

Data sources

Related articles

• Standing: where you stand vs OSHA and peers
• Findings: safety tips and decisions
• Take Action: safety decisions and activation
• How benchmarks work
• Reading a benchmark
• Benchmarks overview
• The decision workspace