The office leaderboard and composite ranking

What the office leaderboard is

The office leaderboard is one ranked view of every office in your network, standing side by side on a single composite score built from every benchmark your network computes for its own locations. It answers the question a regional operations lead asks every week: which offices are pulling ahead, which ones need a coaching conversation, and on what, specifically. It is an internal, office-vs-office view: your locations measured against each other, not against the industry.

This is a different number from the composite on the Network Health hero. That one is a single network-wide health read (compliance rate, recent-activity share, and margin position, blended into one 0-100 score for the whole network). The office leaderboard's composite is per-office: each location gets its own 0-100 standing, built from its rank across every benchmark it has enough data to appear in. See the Network Health composite score explained for that other number; this article is about the per-office one.

Verinode never reaches into a single location's underlying jobs, invoices, or day-to-day records to build this ranking. Every input is a rollup your locations' own data already feeds into the network's benchmarks: a rank, a percentile, a median. Locations own their data; HQ sees the standings.

Where to find it

Two paths lead here:

• The Leaderboard tile, in the Explore row on Network Health (hq.verinode.ai/network), the row of overview tiles below Take Action, Best Practices To Propagate, and Network Flow.
• The Leaderboard item in the sidebar's Network group (labeled Franchisees, Locations, or Network depending on how your network is set up), which opens the full ranked board directly at hq.verinode.ai/leaderboard.

The Leaderboard tile on Network Health

The tile in the Explore row shows:

• Value: the count of offices currently ranked, or an em dash if none are ranked yet.
• Sub-label: "Offices ranked" once there's a board, or "Awaiting data" before there is one.
• Preview: a small bar chart of your top six offices' composite scores, largest first, so you get a shape of the standings before you even click in.

Clicking the tile opens the full interactive leaderboard.

How the composite score is built

Every office gets one composite score, 0 to 100, where higher means stronger standing across the network's own benchmarks. It's built in three steps:

1. Per metric, convert rank to a percentile. For any single benchmark (margin, cycle time, incident rate, whatever it is), the best-ranked office scores 100 and the worst-ranked scores 0, with everyone else spaced evenly between. This already accounts for direction: a metric where lower is better (like days to pay) ranks the fastest office highest, not the largest number.
2. Average the percentiles across every metric the office qualifies for. An office's composite is a straight average of its percentile on each benchmark it has enough data to appear in, not a weighted formula and not a penalty for metrics it hasn't reported yet.
3. Two data-quality floors, both structural, neither about hiding a specific number. A metric only counts toward an office's composite once at least two offices have a value for it (there's no rank without a comparison), and a metric only appears anywhere in the network's benchmarks at all once at least three offices report it (a distribution across one or two offices isn't a distribution, just a number). Sparse metric families, safety incidents before most offices log any, fleet data before equipment is tracked, simply aren't in the average yet; they join automatically as more offices start reporting them.

Job mix is excluded on purpose. Metrics like the share of an office's jobs that are water, fire, mold, or commercial work never feed the composite. Running more of one job type than another isn't better or worse, it's just a different kind of book, so it has no business pulling a standing score up or down.

What feeds the composite

Nearly every per-office benchmark the network computes counts toward the score, once it clears the two floors above: gross and net margin, revenue per employee, days to pay, labor/material/subcontractor/equipment cost ratios, cycle time, average job value and marketing spend as a share of revenue, the Verinode Score and its nine dimensions (Margin, Cash Flow, Operations, Compliance, Team, Reputation, Sales, Process, Vendor), collection rate, supplement approval rate, time to on-site, jobs per employee, lead close rate, referral share, return on marketing dollars, cost per acquired job, the four safety metrics (incident rate, days since last recordable, near-miss ratio, training compliance), equipment ownership and calibration compliance, reputation score plus Google and Yelp ratings, certification currency, facility and fleet economics, team headcount and tenure, commercial client mix, margin by job type (water, fire, mold, biohazard), vendor count and spend concentration, and loaded labor rate. As your network's data coverage grows, new categories light up and start contributing on their own; you don't have to turn anything on.

The full interactive board

Open the Leaderboard, either from the Explore tile or the sidebar, and you land on a flat, hairline-separated ranked list headed Office Leaderboard, with a note reading how many offices are shown and "Click to drill in."

Ranked by. A picker beside the header defaults to Composite. Click it to re-rank the whole list by any single metric instead, type to filter the list of metrics if you know what you're looking for. Re-ranking by a metric changes every row's subtitle to that metric's rank ("#3 of 40 on this metric") and its readout to that metric's own value, formatted in its own units.

Each row shows:

• Rank, colored green for the top three, red for the bottom three, neutral in between.
• Office name, plus a small gold star badge and count if the office holds any network recognition.
• A subtitle: when ranked by composite, "Strongest: [metric] · Needs work: [metric]", the two benchmarks where the office's percentile is highest and lowest. When re-ranked by a single metric, the office's rank and the total offices measured on that metric instead.
• A bar, sized to the composite (or the selected metric's percentile) relative to the rest of the network.
• A readout, the composite score itself, or the selected metric's value in its native format (percent, dollars, days, or a plain number).

Click any row to open its drill-in card.

The drill-in card

The drill-in opens as a glass overlay reading Leaderboard / [office name], with a counter ("3 of 40") and a close button. Use the floating left/right arrows, the arrow keys, or a swipe to step through offices in whatever order the board is currently ranked, composite or a re-ranked metric.

Inside:

• The headline number: the office's composite (out of 100) and network rank, or, if you arrived by re-ranking a metric, that metric's rank, total offices, and formatted value instead, colored by where the office sits in that metric's distribution.
• Strongest / Needs work, the same two labels from the row, spelled out.
• Recognition, any network-awarded badges the office currently holds. Network admins can award a custom recognition label here, or remove one; every user viewing the leaderboard can see badges an office already holds.
• The Metrics, one row per benchmark the office contributes to: the metric's label, its formatted value (colored green when the office sits above the network's top quarter on that metric, red when it sits below the bottom quarter, neutral in between), and its rank out of the total offices measured on it.

The privacy boundary

Office names on the leaderboard follow the same rule as everywhere else on Network Health: franchise and association networks (independent-operator model) see anonymized labels, a stable "Franchisee #XXXX" per office, while single-owner, multi-location networks (one tax ID operating several locations as one business) see the real location names. Either way, the leaderboard only ever reads network-level rollups: ranks, percentiles, and medians your locations' own benchmark data already produced. It never opens a single location's underlying jobs, invoices, or financial detail to build a rank; drilling into an office from this board shows you where it stands, not what's in its books.

Empty states

• Fewer than three offices total in your directory have a named, ranked standing. The Leaderboard tile on Network Health shows an em dash and "Awaiting data," and the full board (both from the tile and from the sidebar) doesn't render at all, there's no partial or single-office leaderboard. This is expected while a small or brand-new network is still onboarding, not a broken page.
• An office you'd expect to see is missing from the board. It either hasn't been added as an active, named entry in your member directory yet, or it hasn't reported enough data on any single benchmark to clear the two-office floor described above. It appears automatically once either catches up.
• A metric you expect on a drill-in card isn't there. That benchmark hasn't cleared the three-office floor across your network yet, not just for that one office. It's excluded from every office's composite and every drill-in card until it does.

Related help

• Network Health, the full page: the hero, Take Action, Best Practices, Network Flow, and the full Explore row this tile lives in.
• The Network Health composite score explained: the network-wide composite, a different number from the per-office one on this page.
• Network benchmarks: the underlying per-office distributions (network p25/median/p75 plus the industry reference line) each leaderboard metric is ranked from.
• Best Practices to Propagate: turning a leaderboard gap into a rollout recommendation.
• Broadcasting to your network: pushing a leaderboard-driven update out to your membership.

Data sources