Vendor detail: the Verinode Score dimension breakdown

What this section is

Every vendor tile you open from the Vendors page carries a Verinode Score, a single number out of 10 built by Verinode's research layer. This article covers the drill-down beneath that number: the Verinode score breakdown, the per-dimension detail, the Provisional banner that tells you whether the number is final, the Also scored in row for vendors researched under more than one category, and the Sources & evidence trail showing exactly what fed the score and where it came from.

All of it is catalog intelligence, not any franchisee's private business. It comes from the benchmark data, the same research record Verinode maintains for this vendor across the whole platform. It reads identically whether you are looking at it from HQ or an individual operator is looking at the same vendor from their own IQ account. Nothing here is scoped to your network specifically, and nothing here exposes what any one franchisee pays, rates, or reports. That data lives in a separate section further down the same overlay, described in Vendors: your network's spend and procurement leverage, in aggregate.

Where to find it

Open Vendors from the HQ sidebar at hq.verinode.ai/vendors. Click any vendor tile, in any row on the page (Off Program, Rate Drift, Top Vendors, Renegotiation Candidates, Broadest Network Footprint, or All Vendors), to open that vendor's detail overlay.

Inside the overlay, scroll past the header (logo, category, program status, and the average Verinode Score), the three headline figures (annual spend, franchisees using, average satisfaction), the Cost distribution figures, and, when a negotiated rate is on file, the Negotiated rate vs network median panel. The Verinode score breakdown is the next section down, followed immediately by Also scored in (when applicable) and Sources & evidence (when applicable).

This section only appears when Verinode's research layer has actually computed a score for this vendor. If it hasn't yet, the section is simply absent, no placeholder or "coming soon" text takes its place. It is also absent whenever the vendor itself doesn't clear HQ's network privacy floor: a vendor used by too few franchisees to aggregate safely shows the aggregate-only banner instead of the entire overlay content, including this section, even though the score breakdown itself carries no franchisee-specific data. That's a deliberate simplicity trade-off, one privacy gate for the whole overlay, not evidence that the score breakdown is risky on its own. See What HQ sees vs. what stays private for the full rule.

The Provisional banner

Directly above the dimension list, a tan-bordered banner appears whenever this vendor's score is not yet final:

Provisional: final once the Operator Advisory Council weights this category and peer ratings accumulate.

A score only drops the Provisional label once two things are both true: the Operator Advisory Council has formally weighted this vendor's scoring category (the dimension weight profile behind the number has been reviewed and set, not just applied by default), and the vendor's Peer Intelligence dimension is backed by real, validated ratings from enough distinct operators across the platform, not a single operator's own estimate carried forward. Until both conditions hold, the score displays with the Provisional badge regardless of how well-researched it otherwise is. A vendor can show a well-evidenced 8.2 Strong score and still carry the Provisional badge; provisional is about how the score reaches finality, not about how thin the underlying research is (that's the separate data-density figure covered next).

Newly scored vendors, and most of the catalog while the Operator Advisory Council works through its category-weighting review, will show this badge. That is expected, not a data problem.

The header line: category, data density, confidence

Above the dimension list sits a right-aligned meta line, for example: saas · 62% data density · assessed.

• Scoring category is one of the nine high-level buckets Verinode groups every catalog vendor into for weighting purposes: SaaS, Equipment, Subs, Insurance, Fleet, Materials, Rental, Estimating Services, or Services. It determines which dimensions apply to this vendor at all and how heavily each one counts. It is shown exactly as the platform stores it, lowercase, with any internal underscore rendered as a space (for example "estimating services").
• Data density is the percentage of this category's total possible scoring weight that currently has real evidence behind it, the weighted share of dimensions Verinode has actually been able to score for this vendor versus every dimension that theoretically applies. A vendor scored on most of its applicable dimensions reads a high percentage; a vendor Verinode has only begun researching reads low. A thin-data score is also capped so a vendor with only a few data points scored can't leapfrog a category leader Verinode has fully researched, the number you see already reflects that guard.
• Confidence is the composite read on that same density, in four bands: Verified (70% or more of the possible weight scored), Assessed (40 to 69%), Estimated (15 to 39%), or Directional (under 15%). This line is omitted only when no confidence value has been computed at all.

The dimension list

Below the header, a flat list of every dimension that carries weight for this vendor's category, ordered with active (scored) dimensions first, heaviest-weighted first, and any excluded dimensions pushed to the bottom regardless of weight. A dimension that carries no weight at all for this category (it simply doesn't apply) is omitted from the list entirely, it is never shown at zero.

Up to eleven dimensions can feed a Verinode Score, though no single vendor shows all eleven, only the ones that count for its category:

| Dimension | What it measures | |---|---| | Market Trust | Aggregate ratings from G2, Capterra & BBB | | Peer Intelligence | Satisfaction from operators using this vendor | | Cost Position | Your pricing vs. peers + contract fairness | | Integration & Ecosystem | API quality & key integration depth | | Switching Cost | Ease of exit, contracts & data portability | | AI & Innovation | AI resilience & vendor innovation velocity | | Risk & Compliance | Security certifications & legal exposure | | Industry Alignment | Restoration industry fit & approvals | | Operational Fit | Usability, onboarding & support quality | | Vendor Trajectory | Financial health & product momentum | | ESG & Sustainability | Sustainability & environmental commitments |

Each row shows:

• The dimension name and its one-line description from the table above.
• A confidence dot beside the name, whenever the dimension carries a score: green for Verified, copper for Observed, amber for Estimated, gray for Stale. (A fifth tier, Reported, exists for individual indicators and for the separate Network team scores section further down the overlay, but doesn't appear at the dimension level here.)
• A horizontal bar filled proportionally to the dimension's 1-to-5 score: green at 4 and above, copper from 3 to 4, amber from 2 to 3, red below 2.
• The score itself, right-aligned, out of 5, to one decimal place.

Excluded dimensions render at reduced opacity with a dash instead of a number, and an italic note explaining why, typically "Insufficient data to score." This is Verinode declining to guess rather than forcing a number, an excluded dimension's weight redistributes across the dimensions that do have real data, so a coverage gap never silently drags the composite score down.

Expanding a row. Any dimension with underlying indicators attached shows a small chevron and is clickable. Expanding it lists every indicator behind that dimension's score. Each indicator row shows:

• A status icon: a check for data actually collected, an hourglass for data that probably exists but hasn't been gathered yet, a dash for "not applicable" to this vendor, a warning triangle for a value the vendor claims about itself that Verinode hasn't independently verified, an X for data searched for and not found, and a lock for data deliberately withheld.
• The indicator's label (for example "G2 Rating" or "Warranty").
• The raw value on file, formatted plainly: a number, Yes/No for a boolean, a comma-separated list for an array, or a dash when nothing is on file.
• The indicator's own 1-to-5 score, or a dash if it didn't contribute one.
• A small confidence dot, on the same five-tier color scale as above, this is the one place Reported (blue) can appear.

Also scored in

Some vendors serve more than one line of business, a field-documentation tool that's also researched as an estimating aid, for instance. When Verinode has scored a vendor under a second category, an Also scored in row appears below the dimension breakdown, listing that other category's name, its score to one decimal (or a dash if the secondary score isn't yet computed), and a plain-language label pill (Strong / Solid / Mixed / Weak) beside it when one exists. Rows are sorted highest score first.

This row only lists categories Verinode has actually researched and scored separately for this vendor, not every category the vendor could theoretically be used in. It shows the number and the label only; it doesn't expand into its own dimension breakdown from here. The row is omitted entirely for single-category vendors.

Sources & evidence

Below Also scored in (or directly below the dimension list, when there's no secondary category), Sources & evidence is the transparency trail behind the score: every research source Verinode gathered, the individual data points it contributed, when it was gathered, and how confident Verinode is in it. The section header reads: "Every data point Verinode used to score this vendor, and where it came from."

One block per source, separated by hairline rules, no card frames. Each block shows:

• The source name, a friendly label (for example "Vendor Website," "G2," "Capterra," "Better Business Bureau," "LinkedIn," "Reviews," "News & Press," "Google Business Profile," "AI Assessment," "Insurance Assessment," "Rental Assessment," "Materials Assessment," "Estimating Service Assessment," "Service Assessment," or "Verinode Analyst Research"). A source Verinode hasn't given a friendly name to yet still renders as readable Title Case, never as a raw internal key.
• A right-aligned meta line, up to three pieces joined by a dot, whichever apply: the method (a plain phrase, "Verinode research" or "Verinode analyst"; any raw model or infrastructure identifier is deliberately coarsened to "Automated analysis" so this page never surfaces how Verinode is engineered under the hood), when the source was gathered (a formatted date), and a confidence percentage.
• The data points that source contributed, a short list of field label and value pairs, for example "Price tier," "Warranty," "Build quality (1-5)," "G2 rating," "Complaint count," each with its recorded value: a number, Yes/No, a comma-separated list, or a dash when nothing is on file. A source with no structured fields beyond its own metadata reads "No structured data points recorded."

A footer line closes the panel: "Confidence reflects how reliable each source is, combined with how certain the research was. Lower-confidence inputs move the score less. Verinode is independent, these sources are gathered for operators, never sold to carriers."

The whole panel is omitted, not empty-stated, when Verinode has no recorded sources for this vendor yet.

How the research-layer score relates to Network team scores

Further down the same overlay, a separate Network team scores section shows a different thing entirely: a per-dimension median built from your own franchisees' ratings of this vendor, gated by a rater-count floor so no single franchisee's rating is ever exposed on its own. It sits beside the research breakdown so HQ can read both questions in one place, "what does Verinode's catalog research say about this vendor" and "what does our network actually experience with it," without the two being confused for each other. The full behavior of that section, including its own confidence tiers and the "Insufficient ratings" and "All raters excluded" states, is covered in Vendors: your network's spend and procurement leverage, in aggregate.

How to use it

1. 1Check the Provisional badge and the data-density line first. A Provisional badge with low density is early-stage research; a Provisional badge with a high density and Verified confidence is well-evidenced, it just hasn't cleared the Council-weighting and peer-validation bar yet.
2. 2Scan the dimension order top to bottom. The heaviest-weighted, active dimensions lead; anything excluded sits at the bottom regardless of weight, so you can see at a glance where the real coverage gaps are.
3. 3Expand the one or two dimensions that matter most for your own decision, security and compliance for a data-handling tool, cost position for a commodity supply line, and read the indicators behind them rather than stopping at the rolled-up number.
4. 4If the vendor serves more than one function in your network, check Also scored in before ruling it in or out based on a single category's number.
5. 5Open Sources & evidence when you need to know exactly what's behind a surprising score, recent independent research reads differently than a vendor's own unverified claim.
6. 6Cross-reference against Network team scores below it for what your own franchisees report, the two views are meant to be read together, not as competing numbers.

Best-practice example

Say a materials supplier used across several franchisees shows a 6.9 Solid score, Provisional, with the meta line reading "materials · 55% data density · assessed." The dimension list shows Cost Position and Market Trust both scored near 4, with Risk & Compliance greyed out and marked "Insufficient data to score," Verinode hasn't yet confirmed this supplier's certifications. Also scored in shows the same vendor also researched under Estimating Services at a lower 5.1 Mixed, a secondary line worth knowing about but not the reason it's on your network's radar. Sources & evidence shows the Cost Position figures came from a Vendor Website scrape three weeks old at 80% confidence, current enough to trust. Read together: a solidly-priced, well-regarded supplier with an unverified compliance posture, worth a certifications follow-up before treating the score as final, exactly the kind of nuanced read the breakdown is built to give instead of a single number.

Data sources

Related articles

• Vendors: your network's spend and procurement leverage, in aggregate
• What HQ sees vs. what stays private: the vendor aggregate boundary
• Comparing vendors across your network
• The Verinode Score on vendors and entities
• Vendor Detail: Score & Research and Score Composition
• Programs
• HQ Benchmarks
• Network Health