Vendor Due Diligence

---
name: Vendor Due Diligence
description: Commissions and controls sell-side due diligence -- scope, provider selection, reliance architecture, and the handling of adverse findings -- when you need a VDD pack that survives a buyer's confirmatory diligence rather than one that collapses in exclusivity.
---

# Vendor Due Diligence

## When to use

Use this skill once the exit route is chosen and before the advisor drafts the CIM, because the VDD databook is what the CIM's financials must reconcile to. It is also the right skill when a known issue exists and the team is deciding whether to disclose it, fix it first, or leave it for the buyer to find. It is worth commissioning where the buyer set is broad or where lenders need a report they can rely on; on a bilateral sale to one known buyer it often is not.

## What it does

Produces: a VDD scope specification by workstream; a provider and reliance plan; a working protocol that keeps the sponsor in control of scope without touching conclusions; a decision framework for adverse findings; and the reconciliation rules that keep the VDD, the databook, the CIM and the model telling one story.

## Method

### Step 1 -- Decide what the VDD is buying

VDD is bought to compress the buyer's diligence and preserve competitive tension, not to market the asset -- marketing is the CIM's job, and a VDD that reads like one is discounted on sight. Set the budget against the price chip you are avoiding and the weeks of timetable you are saving, and be willing to conclude a full pack is not justified.

### Step 2 -- Scope each workstream precisely

Financial and QoE is the workhorse: the databook, adjusted EBITDA with every add-back evidenced, net debt and debt-like items, normalised working capital, revenue recognition. Add commercial, legal, tax structuring, technology and cyber, ESG and pensions as the asset requires, each with periods, entities and the questions it must answer. Scope the databook at the granularity a buyer's QoE will demand -- monthly, by entity, by revenue stream -- because one that cannot be cut the way the buyer wants forces the buyer to rebuild it, which is the outcome the VDD was bought to prevent.

### Step 3 -- Set the reliance architecture before work starts

Agree the addressees, how reliance transfers to buyer and lenders, the liability cap, and the reliance extension fee. Use a QoE provider that is not the company's auditor where possible; buyers discount a report written by the firm that signed the accounts. If W&I insurance is contemplated, remember the underwriter reads the VDD and prices off it, so gaps in scope become exclusions in the policy.

### Step 4 -- Control the process, not the conclusions

The seller controls scope, timetable and access. The seller does not control findings, and any attempt to do so is visible in the drafting. Run a no-surprises protocol: findings are raised at working level as they emerge, the sponsor gets a read-out before the first draft, and management responses sit alongside the finding rather than replacing it.

### Step 5 -- Handle a finding that hurts

Three options: fix it before the process, disclose and quantify it in the pack, or reprice the expectation internally. Concealment is not a fourth option -- it is a deferred and larger price chip. A finding disclosed up front is negotiated as one item against one number. The same finding discovered in exclusivity is negotiated when the buyer has leverage, no competitive tension remains, and it reopens every other item you asserted. Quantify both paths before deciding, and treat the difference as the cost of concealment.

### Step 6 -- Reconcile everything to one EBITDA, and leave the buyer real work

Fix a single adjusted EBITDA definition and enforce it across the VDD, the databook, the CIM, the management presentation and the model, with one owner and version control. Where it differs from the definition used at acquisition, show the bridge -- buyers back-calculate the seller's add-back history, and a silent change of definition is read as a change of story. Then decide what is deliberately left for the buyer's confirmatory work, set the Q&A protocol and turnaround times, and price any top-up scope. A pack claiming to have answered everything invites the buyer to test whether it has.

## Inputs

- Current management accounts, the databook if one exists, and the acquisition-date EBITDA definition
- The exit readiness assessment and any known issues log
- Chosen exit route and indicative timetable
- Advisor recommendations on providers, fees, and reliance terms
- Any prior QoE, audit management letters, or unresolved tax positions

## Output format

Five sections:
1. Scope specification, workstream by workstream, with periods, entities and questions
2. Provider and reliance plan: addressees, caps, reliance extension fees, W&I interaction
3. Working protocol: read-out points, who sees drafts, how management responses are recorded
4. Adverse findings framework: fix, disclose or reprice, each with its quantified consequence
5. Reconciliation rules and the single EBITDA definition, with an owner named

Total length: 1,200-1,600 words. Written for the deal team and the sell-side advisor.

## Example

**Adverse finding, both paths (fictional -- Calder Hygiene Services):**
Management-adjusted EBITDA is $26.0M. The VDD QoE rejects $1.8M of add-backs -- a contract mobilisation credit treated as non-recurring in two consecutive years, and owner costs already run-rated once -- giving $24.2M. Disclosed in the pack, the asset is marketed on $24.2M and bids arrive at 10.5x, an EV of $254.1M.

Left for the buyer to find, the process is marketed on $26.0M and an implied $273.0M. The buyer's confirmatory QoE surfaces the same $1.8M in exclusivity, reprices to the corrected EBITDA and takes a further 0.3x for the credibility loss: $24.2M at 10.2x is $246.8M. Concealment costs $7.3M against the disclosed outcome, plus a specific indemnity and roughly three weeks of timetable -- and that is the cheap version, because it assumes the buyer stays in the process at all.