Questions to resolve before close

What this section is

On an engagement's detail page in Verinode Diligence, once the target has shared at least one category of data and Verinode has enough of a market read to compare it against, a section titled Questions to resolve before close can appear. Each line in it is a single, plain-language prompt: something to go check with the target before you rely on a number, not a verdict on whether that number is good or bad. Verinode never tells you the deal looks strong or weak. It tells you which figures sit far enough off the market that a competent buyer would want to understand why before closing, and leaves the reading to you.

This article is the deep dive on that one section: exactly what makes a line appear, what it does not flag, and how to work through it. For the rest of the engagement detail page, including the header, the deal profile, the Data access rows, and the Target / Your network / Market comparison table underneath, see reading the data-access state on an engagement.

Where it sits on the page

Open Diligence from the HQ sidebar (route /diligence), then click into an engagement to land on its detail page at /diligence/[id]. Reading top to bottom, the page runs: header, deal profile, Data access, then, only when applicable, Questions to resolve before close, then the numeric comparison tables (one per shared category), then the Resolve controls.

Two things about that placement are worth noting. First, the questions sit above the comparison tables that produced them, so you read the prompts before you read the numbers behind them. Second, the section is a single flat list, not grouped by category the way the tables below it are. A question about net margin from Financials and a question about cycle time from Jobs & Claims can sit right next to each other in the same list, in this fixed order when they occur: Financials (net margin, gross margin, operating-expense ratio), Jobs & Claims (cycle time, days-to-pay), AP & Procurement (materials ratio), then Team & Compensation (labor ratio, revenue per employee). Only the metrics that actually trip the check appear; the rest of that order is simply skipped.

How a question gets flagged

A question is generated per metric, not per category, and only when all of the following are true:

1. The category is shared. Verinode only computes a comparison for categories the target has actively granted (marked Shared in Data access). A category still reading Awaiting consent produces no comparison and therefore no question, regardless of what the target's real numbers might show.
2. The target has a usable value for that metric. The target's own account needs enough history for Verinode to read a representative figure. No value, no question.
3. The market has a usable median for that metric. The "Market" column is the median across operators outside your own network, gated by the same peer-count floor every Verinode benchmark uses before it will publish a figure. Until enough outside operators clear that floor, the market column reads as still forming rather than showing a premature number, and with no market figure to compare against, nothing can be flagged.
4. The target's figure sits more than 15% off the market median, and on the side that works against the target. The direction that counts depends on the metric: for a margin figure (net margin, gross margin), being 15% or more below market is what gets flagged; for a cost or time figure (operating-expense ratio, materials ratio, labor ratio, cycle time, days-to-pay), being 15% or more above market is what gets flagged.

That last point is the one worth sitting with:

Where the target's figure sits against your own network plays no part in whether a question fires. The network column is shown for context in the table below, but the check itself only ever compares the target to the market.

The verification prompts, by metric

Each flagged metric renders as one line of plain text, phrased as something to check, never as a finding. The wording plugs in the actual direction ("above" or "below") and the rounded percentage gap:

| Category | Metric | Prompt (with placeholders) | |---|---|---| | Financials | Net Margin | "Net margin is [X]% [above/below] the market median. Reconcile owner compensation and one-time items before relying on it." | | Financials | Gross Margin | "Gross margin is [X]% [above/below] the market median. Check job-cost allocation and revenue recognition before close." | | Financials | Operating-expense ratio | "Operating-expense ratio is [X]% [above/below] the market median. Separate owner discretionary spend before normalizing." | | Jobs & Claims | Cycle time | "Cycle time runs [X]% [above/below] the market median. Confirm the job mix and whether backlog is masking capacity." | | Jobs & Claims | Days-to-pay | "Days-to-pay is [X]% [above/below] the market median. Check carrier mix and aging before assuming the working-capital need." | | AP & Procurement | Materials ratio | "Materials ratio is [X]% [above/below] the market median. Verify purchasing controls, inventory shrinkage, and any owner add-backs before close." | | Team & Compensation | Labor ratio | "Labor ratio is [X]% [above/below] the market median. Confirm crew mix, overtime, and any owner or family payroll before close." | | Team & Compensation | Revenue / Employee | A generic prompt: "[Metric] sits [X]% [above/below] the market median. Verify the drivers before close." |

Every prompt except Revenue / Employee has a purpose-written line naming exactly what to go check (owner compensation, job-cost allocation, purchasing controls, crew mix, and so on). Revenue / Employee is the one metric in the set without a dedicated prompt yet, so it falls back to the generic wording above; the direction and percentage still reflect the same underlying check.

The Take-Action treatment

Visually, each question renders as a single line of body text with a thin ember-red rule down its left edge, the same visual cue Verinode uses across the platform to mark something that belongs on your list of things to act on. There is no card, no button, and no "Discuss with agent" control on these lines, unlike the decision workspace elsewhere in the platform (see the decision workspace and acting on decisions for how that pattern works when Verinode is recommending an action). Here, the rail is doing one job: telling you this line is a checklist item, not background information, without ever telling you what the answer to the check should be.

Why the section sometimes does not appear at all

There is no "No questions to resolve" message. If no metric across every shared category crosses the 15% threshold in the adverse direction, the entire section, heading included, is simply absent from the page. This happens in a few honest ways:

• Nothing has been shared yet (every category still reads Awaiting consent).
• What has been shared is close enough to the market median on every metric that nothing clears the 15% bar.
• The categories shared so far do not have a mature enough market median behind them for the check to run.

A missing section is not a clean bill of health from Verinode; it means nothing in what has been shared and can currently be checked crossed the line worth flagging. It is not the same as a verified, audited target. The completeness note that runs under Data access (documents provided, not audited or confirmed complete by Verinode) still applies to every number this section could ever reference.

Reading a flagged question correctly

1. 1Read the Data access rows first. A category with no comparison underneath it (still Awaiting consent) cannot generate a question, so an empty-looking deal on paper may just mean the target has not shared enough yet, not that everything checks out.
2. 2For each line under Questions to resolve before close, find the matching row in the comparison table below and look at the actual Target and Market figures, not just the rounded percentage in the prompt.
3. 3Treat the prompt as the starting point of a conversation with the target or their advisors, not as the answer. "Reconcile owner compensation" is an instruction to go do diligence work, not a statement that compensation is being misreported.
4. 4Remember the check is one-sided. A target with no flagged questions at all may simply be performing at or above market on everything measured so far, or it may mean too little has been shared to check much of anything. Cross-reference against how many categories actually read Shared before treating "no questions" as reassuring.
5. 5Work through every line before you rely on the numbers to model the deal. None of this replaces your own accountants, lawyers, or operational diligence, it tells you where to point them first.

Best-practice example

You open an engagement where the target has shared Financials, Jobs & Claims, and AP & Procurement, holding back Team & Compensation. Questions to resolve before close shows three lines: net margin running well below the market median with the prompt to reconcile owner compensation and one-time items, days-to-pay running above the market median with the prompt to check carrier mix and aging, and no line at all for gross margin, operating-expense ratio, cycle time, or materials ratio, because none of those crossed the 15% mark. You scroll down to the Financials table, confirm the actual net margin figure against market, and raise owner compensation directly with the seller's accountant before you build your model off that P&L. Meanwhile, because Team & Compensation was never shared, no question could ever have been generated for labor ratio or revenue per employee, whatever those numbers might actually be, so you follow up separately to request that category rather than reading its silence as a clean result.

Data sources