Process benchmarks: how your SOPs compare to peers

What this page answers

Every restoration operator runs a version of the same jobs: water mitigation, fire and smoke, mold remediation, contents, reconstruction. What varies is how many steps each company actually documents, who does them, and how long each stage takes. The Benchmarks tab is where Verinode lines your own documented process up against two other things: the anonymized shape of what operators like you are actually running, and the published industry standard for that category. You decide what to change. Verinode never touches your SOPs on its own, and none of this data is ever sold to carriers, insurers, or any other third party.

This article covers the Benchmarks tab itself: what qualifies a process for comparison, what the peer cohort card shows, the privacy floor that gates every peer number, and how a step-adoption gap turns into something you can act on.

Where to find it

Open Excellence from the sidebar (/excellence, which opens straight to /processes). The page is a horizontal row of six cards you swipe or click through:

Findings · Flow · All Processes · By Standard · Coverage · Benchmarks

Benchmarks is the last card, tagged with the teal accent Verinode uses for Monitor-type surfaces. It is where the peer-SOP comparison lives. The other five cards feed it or sit next to it:

• All Processes is where you document and manage your own SOPs (the "My SOPs" kind filter), plus review Pending Confirmations and browse or import shared SOPs from your HQ network.
• By Standard is the reference library: IICRC S500/S520/S770, LEAN, OSHA, EPA, and state-specific frameworks. Each reference SOP there carries a curated "research standard" score and step list, and a one-click Adopt button that copies it into your own SOPs as a draft. That is the industry-standard layer; it does not depend on any peer data.
• Flow is where stage-by-stage timing benchmarks live (see below), separate from the SOP-shape comparison on the Benchmarks tab.

What counts as a benchmarkable process

The Benchmarks tab only looks at your own SOPs (not Pending Confirmations or standard references), and only ones that are:

• Not archived
• Documented with at least one step that has a step type assigned (inspection, extraction, setup, drying, documentation, pack-out, demolition, cleaning, testing, closeout, coordination, or other)

Verinode groups your qualifying SOPs by category (Water Mitigation, Fire & Smoke, Mold Remediation, Storm Damage, Reconstruction, Contents, Biohazard, General) and work type (Residential, Commercial, or Residential & commercial). If you have more than one SOP in the same category and work-type combination, the one with the highest LEAN score represents that bucket. The tab shows up to three buckets at a time so the page stays scannable, if you document processes across more categories than that, the extras simply wait their turn.

The empty state: no SOP yet

If you have not documented a qualifying SOP in any category, the Benchmarks tab shows:

Peer SOPs Document at least one SOP and peer comparisons unlock the moment your cohort hits enough contributing operators. Verinode never sells operator process data to carriers, insurers, or any third party. Your data sharpens the peer picture for operators like you.

The fix is to document a real SOP on the All Processes tab first (or adopt one from By Standard as a starting draft, then refine it). Once you have one qualifying SOP, its bucket card appears here and starts resolving a peer cohort.

The k-anonymity floor: why some cards say "still learning"

Every peer number on this tab, and everywhere else in Verinode, is protected by the same rule: Verinode will not show you an aggregate built from too small a group of other operators. A tiny cohort can accidentally reveal a specific competitor's exact process, which defeats the point of anonymized benchmarking. So each bucket resolves its cohort through a cascade, from most specific to least specific:

1. Same state and same size band
2. Same state (any size)
3. Same size band (any state)
4. National (everyone eligible, any state or size)

Verinode tries the most specific scope first and falls back a level at a time until it finds one with enough contributing operators to publish safely. If even the broadest, national scope does not clear that floor, the card does not render a peer number, it shows a "still learning" message instead. You are also automatically excluded from your own cohort, so if you happen to be the only operator with an SOP in a bucket, you never see your own data reflected back as "the peer average."

When a bucket has not cleared the floor yet, the card reads something like:

Still learning, peer cohort needs more contributing operators in your [Category] · [Work type] bucket. Once the cohort fills in, this card unlocks.

Verinode does not publish the exact contributor count required, since that number is itself part of what keeps a small cohort from being identifiable. What you can rely on is the direction: the more operators in your category and work-type combination who document and share the same kind of SOP, the sooner that bucket lights up, and the more precise the comparison gets over time (state-level, or state-and-size, instead of national).

Reading a resolved bucket card

Once a bucket's cohort clears the floor, its card fills in with three parts.

Header. The category and work type (for example "Water Mitigation · Residential & commercial"), your own SOP's step count and LEAN score underneath, and in the top right corner the cohort's scope and size, something like "National · N=12." The scope label tells you how specific the comparison is (state-and-size beats national), the N tells you how many contributing peers it is built from.

LEAN vs cohort and step count vs cohort. Two side-by-side stat boxes:

• LEAN vs cohort shows your score next to the cohort median, with a plain-language read underneath: Top quartile, Above median, Below median, or Bottom quartile, based on where your score falls against the cohort's 25th, 50th, and 75th percentiles.
• Step count vs cohort shows your step count next to the cohort median, with its own read: More steps than 75% of peers, Above median peer step count, Below median peer step count, or Fewer steps than 75% of peers.

Neither more steps nor fewer is automatically better. More steps can mean thoroughness or it can mean bloat; fewer can mean efficiency or a gap. Read the step count next to what step types are actually missing, that is the next section.

Steps your peers run that you don't. This is the section that turns the comparison into something actionable. Verinode looks at every step type at least half of the resolved cohort runs, and lists the ones your own SOP does not have. Each missing step shows:

• The step type, in plain language (Documentation, Coordination, Testing, and so on, never the raw database token)
• What share of the cohort runs it (for example, "68% of peers run this")
• The typical time it takes, when available (for example, "~30min")

Next to each one is a Capture as gap button. Clicking it does not silently edit your SOP. It records the gap as a Pending Confirmation, which appears back on the All Processes tab under that same name, where you review it and decide whether to draft your own version of the step. Verinode never copies a peer's actual description into your SOP, only the fact that the step type exists and how common it is. Once you capture a gap, the button changes to Captured so you know it is already on your review queue and won't double up.

If your SOP already covers every step type the cohort runs at that adoption level, the card shows a plain confirmation instead: your SOP covers everything your cohort runs at that level of adoption, no missing-steps list needed.

The other kind of process benchmark: stage timing in Flow

The Benchmarks tab compares the shape of your SOP (which steps, how many, how LEAN) to peers. There is a second, separate kind of process benchmark that compares timing: how many days a real job takes to move from one milestone to the next, for example Completed to Paid, or Interviewed to Hired on the recruiting side. That comparison lives inside the Flow tab, not Benchmarks: open any stage-timing metric there and, where a peer comparison is available, you will see your own median duration next to a peer band (a typical peer range, like "most run 8 to 15 days"), reflecting the same kind of cohort as the SOP comparison above.

Both surfaces draw on the same trust rules: an operator's individual timing never gets attributed to them, only rolled into anonymized cohort medians, and both are gated by the same k-anonymity floor described above. If a stage-timing metric on the Flow tab does not show a peer band yet, it is the identical reason a Benchmarks bucket says "still learning," not enough contributing operators have that stage timed yet.

Best-practice example

Say your Water Mitigation SOP has 9 steps and a LEAN score of 61. The Benchmarks tab resolves a National · N=14 cohort for that bucket: cohort median LEAN is 74, so your card reads "Below median." The step count box shows the cohort median at 11 steps, "Below median peer step count." Under "Steps your peers run that you don't," two step types appear: Documentation at 71% adoption (~30min) and Testing at 64% adoption (~30min). Both are steps your crew likely already does informally, they just are not written into the SOP. Click Capture as gap on both, then open All Processes and turn the resulting Pending Confirmations into real steps. That alone accounts for most of the gap between your 9 steps and the cohort's 11, and the LEAN score typically moves with it on your next SOP review.

Related reading

• Reading a benchmark, the general vocabulary (percentiles, cohorts, trust labels) that this tab's peer numbers follow
• How benchmarks work, the mechanics behind every peer comparison on the platform, including this one
• Benchmarks overview, the operator and vendor benchmark surfaces alongside this process one
• The decision workspace, where a captured process gap can turn into a tracked decision once you act on it

Data sources