Renegotiation candidates: outlier spenders at 2x the network median

What this row is

Renegotiation Candidates is a row on the HQ Vendors page that flags vendors where per-franchisee spend on the exact same vendor varies unusually widely across your network. It does not require a formal HQ vendor-approval program, a negotiated rate on file, or any manual flagging. It works off a single statistical signal: for each vendor, Verinode compares the median monthly spend a franchisee pays against the 75th-percentile (p75) monthly spend, and flags the vendor when the p75 figure is at least twice the median.

That gap matters because it means some franchisees are paying meaningfully more than others for an identical relationship: the same software subscription, the same equipment rental agreement, the same subcontractor trade. When a handful of locations are paying twice what a typical location pays for the same line item, that is not normal price variance, it is a sign nobody has gone back to that vendor with the terms your best-negotiated franchisee already has. This row exists to put that list in front of the people who make vendor calls, without anyone at HQ having to eyeball every franchisee's spend by hand.

Verinode surfaces the pattern; it does not negotiate on your behalf and it does not tell a franchisee which vendor to use. HQ reads the list, decides which gaps are worth a call, and works the negotiation.

Where to find it

Open Vendors from the HQ sidebar at hq.verinode.ai/vendors. The page is a single scrolling home, no tabs. Renegotiation Candidates sits below Top Vendors and above Broadest Network Footprint, in this order top to bottom: Hero, Local Network, Off Program, Rate Drift, Top Vendors, Renegotiation Candidates, Broadest Network Footprint, All Vendors.

The row is titled Renegotiation Candidates and renders as a horizontal, scrollable strip of tiles, the same tile-row pattern used everywhere else on the page. It is built from the network data, the same nightly network rollup that feeds every other row on Vendors, so what you see reflects last night's aggregation, not a live query.

How the ratio is computed

Every vendor in the network data already carries two per-franchisee spend statistics, computed nightly across the franchisees actively using that vendor:

• Median monthly spend: the middle value of what franchisees pay this vendor per month. Half of franchisees using the vendor pay at or below this figure, half pay at or above it.
• p75 monthly spend: the 75th-percentile value, the amount that only a quarter of franchisees using the vendor pay more than.

A vendor qualifies for Renegotiation Candidates when both of these hold:

1. Median monthly spend is on file and greater than zero.
2. p75 monthly spend is on file and is at least 2 times the median.

In other words: the top quarter of franchisees paying this vendor are paying at least double what a typical franchisee pays. Verinode uses median and p75, rather than comparing the highest payer to the lowest, because a single outlier data point (a bad invoice month, a one-off setup fee) shouldn't put a vendor on a call list. Requiring the whole top quartile to be at 2x or more is a sturdier signal that the gap is structural, some franchisees are on an old rate, a smaller package, or simply never negotiated, not a one-time blip.

This calculation runs independently of HQ's vendor-approval programs and negotiated rates. A vendor can show up here even if it has never been added to a program in Programs and even if no negotiated rate has ever been captured. It only needs enough spend history across enough franchisees for the median and p75 to be meaningful.

What each tile shows

Each vendor that qualifies renders as a double-width tile, colored in the Maintain signal tone (Hard Hat Yellow) to mark it as worth a look, not a compliance emergency. A tile shows:

• Headline figure: the ratio itself, formatted as "N.N× median" (for example, "2.4× median"). If the ratio can't be computed for some reason, the tile reads "Outlier" instead.
• Vendor name: the canonical vendor name, as the tile's main label.
• Sub line: "p75 $X vs median $Y", the two dollar figures the ratio is built from, each abbreviated (for example, "$4k" or "$180"). This is the same figure formatting used across the rest of Vendors: values a million dollars or more show as "$X.XM", values a thousand or more show as "$Xk", everything else shows as a plain dollar amount.
• Meta line: the franchisee count using that vendor, phrased as "N franchisees" (or "1 franchisee" for a single user, though see the privacy note below, a single-franchisee vendor generally will not reach this row at all).

Clicking a tile opens the same vendor detail overlay used everywhere else on the page, a centered modal on top of the Vendors home. Nothing navigates away from the page; closing the overlay returns you to your scroll position.

Ordering and the tile cap

The row shows up to 8 tiles. It is not sorted by ratio, the vendor with the highest multiple is not guaranteed to be first. Instead, qualifying vendors keep the same order as the underlying rollup, which is sorted by network monthly spend descending. In practice this means a vendor with a very large network monthly spend and a 2.1× ratio can appear ahead of a smaller vendor at 4× median, because the row is surfacing spend-significant outliers in the order the rest of the page already uses, not ranking by severity of the gap. If you want to check every qualifying vendor's exact ratio, open each tile: the ratio and both dollar figures are also shown on the vendor's detail overlay (see below).

The privacy floor

Renegotiation Candidates respects the same cohort-privacy guard as every other row on this page. In an independent-operators network (the default posture for franchise and association groups), a vendor's row, including its p75-vs-median ratio, is only shown once enough distinct franchisees use that vendor for the numbers to be safely aggregated. A vendor used by too few franchisees network-wide is excluded from this row entirely, no matter how wide its spend variance is, because showing "this vendor's spend is 3× its median" for a vendor with almost no adoption would risk identifying the specific franchisee behind the high figure by elimination.

That means a vendor can have an enormous spend gap between two franchisees and still never surface here, if too few franchisees use it overall. This is expected behavior, not a bug: as more franchisees adopt a vendor, its row becomes eligible for this signal in the next nightly rollup. Enterprises configured as a single legal entity across locations (same_entity) are not subject to this guard, since there is no separately owned franchisee to protect.

See What HQ sees vs. what stays private for the full mechanics of the cohort floor and what happens when you try to open a suppressed vendor directly.

Renegotiation Candidates vs. Rate Drift vs. Off Program

These three rows all point at vendor spend problems, but they answer different questions and none of them require the others to have data:

• Off Program asks: is this vendor even on an HQ-approved program? It needs at least one active vendor-approval program to evaluate anything.
• Rate Drift asks: is the network paying above the rate HQ actually negotiated? It needs a negotiated monthly rate captured in a program's terms before it can flag anything.
• Renegotiation Candidates asks: is there a wide spend gap between franchisees on the same vendor, whether or not a formal program or negotiated rate exists at all? It is the broadest of the three, and often the first signal to appear, because it needs nothing more than spend history.

A vendor can appear in more than one of these rows at once. A vendor with no program and a 2.5× spread would show only in Renegotiation Candidates. The same vendor, once brought onto a program with a negotiated rate, could then also start showing in Rate Drift if some franchisees are still paying above that new rate.

Opening a vendor from here

Clicking a Renegotiation Candidates tile opens the vendor detail overlay. The section most relevant to this row is Cost distribution, which shows:

• Median monthly spend across franchisees using the vendor.
• p75 monthly spend, flagged with a "≥2x median" label when the vendor cleared the renegotiation-candidate bar (which it did, to appear in this row).
• Total network monthly spend across every franchisee using the vendor.

Below that, if a negotiated rate exists on any active program covering this vendor, a Negotiated rate vs. network median block compares the negotiated figure against the network median directly. And further down, Per-franchisee participation lists every franchisee using the vendor with their individual monthly spend, satisfaction rating, and Verinode Score, letting you see directly which franchisees are on the high end of the spread. On independent-operator networks, franchisee names in that list are anonymized to a stable label (for example, "Franchisee #A1B2"); on same-entity networks, real location names show through.

How to use it

1. 1Scan Renegotiation Candidates for vendors with a high ratio and meaningful spend. A 2.1× ratio on a $60/month tool matters less than a 2.4× ratio on a $2,000/month equipment rental line.
2. 2Open a tile and check Cost distribution and Per-franchisee participation together. The gap between median and p75 tells you the size of the opportunity; the participation list tells you which franchisees are actually paying the high figure.
3. 3If the vendor is already on an HQ program with a negotiated rate, check whether the high-paying franchisees are simply not compliant with that rate (a Rate Drift problem you can enforce) versus the rate itself being outdated for current volume.
4. 4If the vendor has no program at all, treat this as the case for bringing it onto one: a network-wide renegotiation using your best-negotiated franchisee's terms as the floor.
5. 5Use the vendor's Compare alternatives panel, if available, to check whether a same-category alternative already has a stronger network footprint and a better Verinode Score, worth raising alongside the renegotiation conversation.

Empty state

Before any vendor's spend spread crosses the 2× bar, the row shows this text instead of tiles:

"No vendors show top-quartile spend at 2× the network median yet. This row populates as outliers emerge."

This is common in small or young networks, where too few franchisees use any single vendor for a meaningful median-versus-p75 comparison to exist, or in networks where vendor pricing happens to be tight across the board. Nothing needs to be configured to make this row populate; it fills in on its own as more franchisees register vendor relationships and spend data accumulates.

Best-practice example

Say Renegotiation Candidates shows a tile reading "2.6× median" for a drying-equipment rental vendor, with a sub line "p75 $3,900 vs median $1,500" and a meta line "6 franchisees." Opening the tile shows the vendor is not on any HQ program yet, so there is no negotiated rate to check compliance against. The Per-franchisee participation list shows two of the six franchisees paying close to the $3,900 figure while the other four sit near or below the $1,500 median. That is a strong case for bringing this vendor onto a preferred-vendor program in Programs, using the $1,500 median (or better) as the target rate, rather than starting a program from scratch with no evidence of what a fair rate looks like. Once that program exists and a negotiated rate is captured, the same two franchisees would then also surface in Rate Drift until their billing catches up to the new terms.

Related reading

• Vendors: your network's spend and procurement leverage, in aggregate: the full page tour, every row in order
• The network vendor headline: the hero panel above this row
• What HQ sees vs. what stays private: the full cohort-privacy mechanics referenced above
• Comparing vendors across your network: the side-by-side compare panel opened from a vendor's detail overlay
• Programs: where preferred-vendor programs and negotiated-rate terms are defined
• Network Health: the HQ command home this page's tiles feed into

Data sources