Per-office benchmarks: ranking your own locations

The frame this page uses

Every other benchmarking product in restoration asks "how does your company compare to the rest of the market." An HQ leader asks a different question first: how do my own offices compare to each other. That is the frame the Benchmarks tab is built around. For any metric Verinode tracks, your network's own offices are ranked against each other inside your own network's own distribution, your network's own p25, median, and p75 are the primary numbers on the card, and the broader industry appears once, as a single quiet reference line, never as a second cohort you are being scored against.

This article is about that ranking mechanic specifically: how a metric's per-office numbers are built, how a benchmark row reads on the tab, and what happens when you click one open. For the full catalog of every metric family and where its numbers come from, see What gets benchmarked: the metric families. For the single ranked list across every metric at once (the "who do I coach first" screen), see The office leaderboard and composite ranking, a related but different number built from these same per-office ranks. For the trust model underneath both, see What HQ sees: the network privacy boundary.

Verinode never opens a single location's underlying jobs, invoices, or financial detail to build any of this. Every number on this page is a rollup: a representative value per office, and the spread those values form across your network. Locations own their data; HQ sees the standings.

Where to find it

Open Benchmarks from the HQ sidebar at hq.verinode.ai/benchmarks. It is the first of seven tabs across the top of the page: Benchmarks, Carriers & TPAs, Materials, Industry Data, Ratings, Analyst Reports, Industry News. Stay on the default Benchmarks pill, this article covers that tab.

How a metric's per-office ranking is built

For a given metric, Verinode works out one representative value per office, then computes the shape of your own network from those values:

1. One value per office. Depending on the metric, this is the office's most recent snapshot (a margin percentage, a Verinode Score), a median across its own rows (a per-job cycle time), or a ratio summed across its jobs (collection rate: total collected over total billed). Only offices with a real, named entry in your network's member directory and a usable value for the metric count.
2. Your network's own p25, median, and p75. Once every contributing office has a value, Verinode sorts them and reads off the 25th percentile, the median, and the 75th percentile, the same three-number distribution shape used everywhere else on the platform. This is your network's own spread, not a comparison to anyone outside it.
3. A rank for every office. Offices are sorted best to worst and numbered 1 through however many contribute, honoring the metric's direction. On a metric where higher is better (margin, collection rate), the highest value ranks 1. On a metric where lower is better (days to pay, cost ratios, incident rate), the lowest value ranks 1.
4. Where each office sits relative to your own p25/p75. Every office is tagged below your network's own bottom quarter, in the middle, or above your network's own top quarter, again honoring direction, so "above p75" always means the good end of the spread and "below p25" always means the end that needs attention.
5. A single industry reference line. For metrics that draw from the same cross-network intelligence tables other operators contribute to, Verinode also computes one more number: the median value among operators entirely outside your network. This is the one place a number from beyond your own offices appears on this page, and it appears as a single median, nothing else, protected by the same anonymity floor every peer benchmark on the platform uses (see the callout below). Metric families sourced purely from your own network's roll-up tables (equipment, reputation, certifications, facilities, fleet, team, vendor and procurement, among others, the full list is in the metric-families article linked above) have no cross-network counterpart table yet, so those show your own office spread only, with no industry line at all.

A metric doesn't render at all until at least 3 of your offices have a usable value for it. A distribution across one or two offices isn't a distribution, it's a couple of numbers, and Verinode does not show a benchmark card that would overstate confidence in a thin sample. As more offices start reporting into a metric family (a safety log, an equipment inventory, lead intake), the card appears on its own the next time the tab refreshes. There is no manual step to turn a metric on.

Reading a row on the Benchmarks tab

Metrics are grouped into category sections (Profitability, Cash Velocity, Cost Structure, and so on), each with a header, a one-line blurb, and a small column header row reading Metric · Peer distribution · You vs peer. Every row underneath follows the same layout:

• Metric name, an info icon you can hover for the plain-language formula, and a bell icon to watch the metric (see Watching a metric).
• Scope label and office count, directly under the name: "Your Offices", followed by the number of offices behind the row, worded as "8 offices" (or "1 office"), never a statistician's "n=8."
• The distribution strip, the middle column and the visual hero of the row. Above the strip, a caption reads "Industry median" with the formatted value, when an industry reference exists for that metric. The strip itself shows a shaded band for your network's own interquartile range (p25 to p75), a vertical tick at the industry median (the reference line), and a colored dot marking your network's own median: green when your network sits at or ahead of the industry median in the metric's better direction, copper/red when it sits behind. When there's no industry reference for the metric, the strip instead shows just your own office spread with no tick.
• The value column, on the right, headed "You vs peer" (the same header IQ's operator view uses; on this HQ tab it always shows your network's own reading). It shows your network's own median, colored the same green or red as the dot, labeled "Network" underneath the number. There's no dollar-impact figure on HQ per-office rows, the "$X/yr" dollar-gap callout you may have seen on an individual operator's benchmark card is an IQ-only feature that needs a single operator's own annual revenue, which a network-level row doesn't carry.

A category section only appears once at least one of its metrics has something to show, and it collapses entirely if none do. Rows you'd expect to see stay hidden rather than rendering as an empty "cohort building" placeholder, per Verinode's label-over-gating rule: an absence of data produces silence, not a ghost row.

Cards vs Table view

A Cards / Table switch sits in the top-right of the tab, next to the section header. Cards is the layout above, one visual row per metric grouped by category. Table collapses every metric (except Materials, which lives in its own tab) into a single sortable table with columns Metric, You, Peer, Delta, Cohort, N.

The drill-in: Across Your Offices

Click any row, in either view, to open the deep-dive overlay for that metric. It opens as a full-screen glass slider you can step through left, right, or with the keyboard, same as every other benchmark deep-dive.

At the top, a large number repeats your network's own median for the metric. Below it, a section headed "Across Your Offices" carries the actual per-office ranking:

• A header line giving the office count ("6 offices," "1 office") and, when an industry reference exists for the metric, " · Industry [value]" appended after it.
• A distribution chart: the same p25-to-p75 shaded band, a marker at your network's own median position, drawn across your own offices' spread. (If no office has a value for the metric yet, this area reads "No offices have reported this metric yet," though in practice a card never reaches this drill-in without at least 3 offices already contributing, so this is a defensive fallback rather than something you'll normally see.)
• The ranked office list. Every contributing office, one row each, in rank order: a rank number, the office name, and its formatted value. The value is colored green when that office sits in the good end of your network's own spread for this metric (above p75 when higher is better, below p25 when lower is better), copper/red when it sits in the opposite end, and plain foreground color for everything in between. Office names come from your network's own member directory, the same names used across HQ's Network Health pages; see What HQ sees: the network privacy boundary for how naming works across the platform. A handful of metric families (the equipment, reputation, and vendor rollups) fall back to the label "Unnamed office" if a location hasn't yet been given a name in the directory.

Below the office ranking, the drill-in carries two more panels that are the same on every benchmark card, HQ or IQ:

• "How It's Calculated." For per-office metrics this reads a fixed template: "Median across your [N] office(s) on [metric], with each office ranked in the deep-dive. The band shows your network's p25–p75 spread; the industry line is the median of [N] operators outside your network, a reference, not a competitor."
• "Data Sources." When an industry reference exists for the metric, this lists a bullet reading "Peer cohort, [N] operators contributing under data dividend," where the count is the industry-side operator total, not your office count. When no industry reference exists yet for the metric, it instead reads "Sources will appear once cohort sample size and research coverage land for this metric."

The "Development Over Time" weekly-trend chart and the cohort-scope picker you may recognize from an individual operator's benchmark drill-in do not appear on HQ per-office cards, both are IQ-specific tools (a single operator's own trend line, a national/regional/group scope switch) that don't map onto a network-of-offices view.

What each family draws from, briefly

A quick map of which categories carry the industry reference line and which are office-vs-office only, the full detail (every metric, every source table) lives in What gets benchmarked: the metric families:

• Carries an industry line: Profitability, Cash Velocity, Cost Structure, Sales & Marketing, Service Speed, Productivity, Safety, Job Mix & Specialization, Profitability by Job Type, and Performance Score, all draw from the same cross-network intelligence facts tables individual operators contribute to.
• Office-vs-office only, no industry line: Equipment & Fleet, Reputation, Certifications, Team Performance, Recruiting & Hiring, Facilities & Real Estate, Vehicle Fleet, Workforce, Platform Engagement, Clients & Sales Mix, and Vendor & Procurement, each of these draws only from your own network's roll-up tables, and there is no equivalent table on the cross-network side yet. That's a scope limit on what data exists in the shared intelligence layer today, not a privacy decision.
• A hybrid: Loaded Labor Rate. The office side reads a payroll roll-up that follows your HQ consent settings directly, so every office with payroll data appears even if it hasn't separately opted into peer benchmarking, while the industry reference reads from the same benchmark-consented facts layer as the rest of Cost Structure. Lower is the cost-better direction here.

Empty states

• A metric isn't showing at all. Fewer than 3 of your offices have reported a usable value for it yet. It appears automatically once a third office's data clears that floor, there is no manual step and no way to see it earlier.
• A whole category is missing (say, Recruiting & Hiring). Every metric in that family is below the 3-office floor. Check with the offices you'd expect to have that data connected.
• The industry median caption and reference tick are absent from a row that otherwise shows your own spread. Either this metric family has no cross-network counterpart table (see the list above), or the outside-operator cohort for this metric hasn't cleared the anonymity floor yet. Either way, your own network's p25/median/p75 still render normally, only the external comparison point is missing.
• "No offices have reported this metric yet" inside a drill-in's Across Your Offices section. A defensive fallback for the case where a card has no office values at all; you should not normally see this given the 3-office floor gating the card's existence in the first place.
• An office you expect on the ranked list is missing. It either hasn't been added as an active, named entry in your member directory, or it hasn't reported a usable value for that specific metric (a different office can still show while this one is silent, per-metric coverage varies office by office).

Related help

• What gets benchmarked: the metric families: the full catalog, every metric, where its numbers come from, and which families carry the industry line.
• The office leaderboard and composite ranking: the single ranked view across every benchmark at once, built from these same per-office ranks.
• What HQ sees: the network privacy boundary: the trust model, the k-anonymity floor on the industry line, and how office naming works across HQ.
• Network Health: the network home this tab sits alongside.
• Watching a metric: getting notified when a metric on this tab moves.
• HQ overview
• HQ Programs
• HQ Standards
• HQ Report Library
• HQ Compliance
• Broadcasting to your network
• Discovery Day
• Item 19

Data sources