Why external evidence quality decides private debt approvals
A private debt lending approval rests on an external evidence pack: a structured, source-grounded body of evidence about the borrower, its market and its covenant package, assembled before credit committee sits. The pack has five components: borrower financials, sector benchmarks, covenant benchmarks, market-position validation and downside scenarios. Everything a committee votes on should trace back to one of those five, and every material claim inside them should trace back to a document. When the pack is thin, the committee is not weighing risk, it is guessing with extra steps.
The stakes are structural, not cosmetic. Private credit has grown to an estimated $1.5-2.0 trillion in assets as of end-2024, borrowers typically lack public ratings, and available evidence indicates they carry lower credit quality and higher leverage than borrowers in comparable public markets. The Financial Stability Board also flags valuation opacity, growing reliance on private ratings and limited fund- and loan-level data, alongside concentration in technology, healthcare and services, sectors where credit risk in private credit is being reshaped by technology substitution. In other words, the market's default condition is opacity, and the lender's evidence pack is the corrective.
Competition sharpens the point. Global private credit fundraising increased again in 2025, even as the asset class's annual growth rate looked set to cool, according to S&P Global Market Intelligence. When capital is abundant and spreads compress, the temptation is to win deals by relaxing evidence standards. The lenders that compound returns over a cycle do the opposite: they treat evidence discipline as the differentiator that lets them move fast without moving blind, which is why evidence-backed diligence research has become a competitive rather than administrative concern.
- Borrower financials: audited statements, quality of earnings, management accounts and the debt schedule
- Sector benchmarks: leverage, pricing and performance comparables for the borrower's industry
- Covenant benchmarks: market-standard covenant structures, cushions and testing mechanics
- Market-position validation: external proof of the commercial thesis behind the loan
- Downside scenarios: stress cases that test debt service under adverse assumptions
The five-part external evidence pack for a lending approval
Each component of the pack answers a different committee question, and each has a defined standard of good. Borrower financials anchor the capacity to repay; benchmarks anchor the comparison; market-position validation anchors the commercial thesis; downside scenarios anchor the stress case. A pack that is strong on three and silent on two is not two-thirds complete, it is structurally biased toward approval, because the missing components are usually the ones that would have argued against the deal.
| Component | What it covers | What good looks like |
|---|---|---|
| Borrower financials | Audited statements, quality of earnings report, management accounts, existing debt schedule financial model checks | Reconciled EBITDA with every add-back tied to a source document; debt schedule that matches the draft facilities agreement |
| Sector benchmarks | Leverage, pricing and margin comparables for the borrower's sector and geography | Sourced comparables with defined peer sets, not anecdotal market colour |
| Covenant benchmarks | Market covenant structures, cushion levels, testing frequency and carve-outs | Current market evidence on cushion ranges and cov-lite prevalence, mapped to the proposed documentation |
| Market-position validation | Customer concentration, pricing power, competitive position, revenue durability | External corroboration of management claims, ideally from customer or supplier evidence rather than the data room alone |
| Downside scenarios | Stress cases for revenue decline, margin compression, rate resets and refinancing risk | Explicit debt-service math under stress, with the covenant cushion tested against each scenario |
Market-position validation deserves particular care in private debt because the loan is underwritten on cash flow durability, not on an exit multiple. A borrower can show three years of audited growth and still be a weak credit if its two largest customers are unprotected or its pricing is being compressed by a new entrant. Evidence here means contracts, churn data, customer references and independent market analysis, assembled with the same rigour as a commercial diligence workstream commercial diligence checklist. Downside scenarios then close the loop: they convert the validated thesis into concrete debt-service tests, so the committee sees not just whether the borrower can pay, but how much has to go wrong before it cannot.
What credit committees should test in the evidence
Raw documents do not decide deals; tested evidence does. Between the data room and the approval sits a testing layer, and three tests matter most in private debt: the integrity of EBITDA, the mechanics of the leverage ratio, and the calibration of the covenant package. Each test exists because the headline number flatters the credit more often than borrowers or sponsors intend.
EBITDA definitions and add-backs
Average closing leverage in Proskauer's 2025 private credit data was 5.1x, up from 4.9x in 2024 and 4.6x in 2023, but the chapter's authors caution that trends in closing leverage must be read alongside how consolidated EBITDA is defined, because more borrower-favourable formulations make closing leverage appear lower than it is. That gap is the add-back stack: restructuring costs, transaction expenses, projected synergies and run-rate adjustments layered onto consolidated net income. Some adjustments are legitimate; most of the risk concentrates in the projection-based ones. The test is simple and unforgiving: every add-back should be traced to a source document, checked for historical realization, and capped where the documentation caps it. The same data shows that discipline slipping: among borrowers with EBITDA below $50 million, only 49% of 2025 deals capped the add-back for general non-recurring expenses, against 66% in 2022 and 67% in 2023, while among borrowers above $50 million of EBITDA the share fell to 26% from 61% in 2023 and 34% in 2024. Normalising earnings this way is the core of financial due diligence in a lending context.
Leverage ratio mechanics
The covenant ratio is only as meaningful as its definitions. Debt definitions vary on earnouts, guarantees and hybrid instruments; cash netting is often permitted but usually negotiated, with questions over caps, which entities' cash counts and whether a control agreement is required; and pro forma rules determine how projected synergies and run-rate savings enter the calculation. Two loans can quote the same headline leverage covenant while one affords the borrower materially more effective headroom. Committees should require the underwriting pack to show the ratio computed under the actual draft definitions, not under a simplified headline convention. For a deeper treatment of how AI disruption changes borrower risk in these structures, see Plausity's analysis of private credit borrower diligence and covenants borrower covenants.
Covenant calibration and testing cadence
A typical leverage covenant in a direct lending transaction is set with a 25-35% cushion to the EBITDA projected in a sponsor or borrower model delivered before closing, and compliance certificates typically arrive 45 to 60 days after quarter end. Both numbers should shape the underwriting evidence. The cushion defines how much deterioration the covenant tolerates before it speaks; the reporting lag defines how long a breach can remain invisible. Market context matters too: average closing leverage in Proskauer's 2025 deal sample increased to 5.1x, with average debt capacity of 6.3x, meaning lenders were permitting roughly 1.2 turns of additional debt after closing. The committee question is not whether a covenant exists, but what it actually monitors, when it monitors it, and whether the cushion survives contact with the downside scenarios in the pack.
- Recompute headline leverage with every add-back unwound, then compare against market data: average closing leverage in Proskauer's 2025 middle-market sample was 5.1x, with 68% of deals closing between 4.00x and 6.99x
- Test the covenant ratio under the draft agreement's debt definition, cash netting caps and pro forma rules
- Check the cushion against sponsor-projected EBITDA and against each downside scenario, not only the base case
- Diary the compliance certificate timeline and model what 45-60 days of reporting lag means for remedy rights
Evidence checklist and red flags
A working checklist keeps the pack complete; a red-flag discipline keeps it honest. The checklist below covers the documents a per-deal underwriting file should contain before committee. The red flags that follow are the patterns that most often signal the evidence is being managed rather than disclosed.
- Audited financial statements for at least three years, with auditor letters and any qualifications
- Quality of earnings report reconciling reported EBITDA to adjusted EBITDA, add-back by add-back
- Customer and supplier contracts, including change-of-control and exclusivity provisions
- Management accounts for the most recent interim period, bridged to the audited statements
- Existing debt facilities, security packages, intercreditor arrangements and any amendment history
- Sector and market data supporting the benchmark and market-position components of the pack
Red flags in the numbers
The clearest numeric warning signs cluster around cash interest and adjustments. PIK toggles appeared in roughly 10% of transactions in Proskauer's deal data in each of the last two years, a liquidity-bolstering feature that has steadily gained traction. The Financial Stability Board likewise notes that some private credit borrowers appear to be relying more on payment-in-kind loans, which can signal deteriorating credit conditions. Add to that the declining prevalence of add-back caps noted above, and the pattern is consistent: adjustments are expanding faster than the documentation is constraining them. An evidence pack that shows rising projected add-backs alongside rising PIK features is describing a borrower that may already be managing a cash flow problem the base case does not acknowledge.
Red flags in structure
Structural erosion is quieter but equally telling. Covenant-lite structures accounted for 21% of deals in Proskauer's 2025 sample, and 91% of those cov-lite deals involved borrowers with EBITDA above $50 million. In these springing structures the financial covenant is tested only when a negotiated share of the revolving commitments is drawn, typically 40% or higher, rather than every quarter. Combined with covenant ratchets, covenant holidays and uncapped adjustment baskets, these features can hollow out the covenant while leaving the term sheet looking conventional. The test for the committee is whether the covenant package would actually surface deterioration early enough to matter, or whether it mainly documents terms that will be renegotiated when stress arrives. Systematic red-flag reporting of this kind is exactly what a structured risk register is for risk register automation.
Structuring the pack for credit committee
A complete evidence pack can still fail at committee if it is organised as a document dump. Committees act on findings they can verify, rank and map to terms, which is the same discipline that governs version control in committee memos. Three structuring principles convert raw diligence into a decision-ready pack: source traceability, materiality tiering and mapping evidence to the transaction terms they inform.
Source traceability comes first. Every finding in the pack should point to the document, page and passage that supports it, and the pack should carry an explicit evidence-gap log listing what could not be verified and why. Data provenance is what separates an evidence-backed finding from a well-written assertion, and it is the property that lets a committee member check any claim in minutes rather than re-running the diligence data provenance in diligence. An honest gap log also protects the lender: it records that the committee approved the deal knowing exactly which assumptions rested on management representation rather than verified documents.
Materiality tiering comes second. Findings should be ranked by financial impact, legal exposure and deal relevance so the committee reads a ranked risk picture rather than an alphabetical file. This is the discipline behind materiality-weighted risk intelligence Risk Radar: the ten findings that could move the credit decision belong on page one, with the long tail of minor observations available but not competing for attention, the same reframing that turns diligence into committee decision logic. Mapping comes third: each finding should be tied to the covenant package, pricing grid or condition precedent it informs, so that diligence output lands in the credit agreement rather than in a side deck. A leverage-ratio finding that never reaches the covenant definition is a finding the deal quietly ignored.
- Tag every finding to its source document and page, and log evidence gaps explicitly
- Rank findings by financial impact, legal exposure and deal relevance
- Map each material finding to the covenant, pricing or CP it should influence
- Circulate the structured pack to underwriting, legal and advisory workstreams before committee
Underwriting diligence versus ongoing monitoring
The evidence pack described in this article supports one decision: the lending approval. Underwriting diligence is a point-in-time exercise, bounded by the information available before closing, designed to give the committee a verifiable basis for approving or declining a specific transaction on specific terms. Ongoing monitoring is a different discipline that begins after funding: continuous surveillance of covenant compliance, borrower performance and portfolio health across the life of the loan. The two share source documents but differ in cadence, tooling and questions asked, and conflating them leads teams to under-invest in one while assuming the other covers it.
Regulatory frameworks reinforce that both stages demand rigour. In the UK, the FCA's rules for alternative investment fund managers set out dedicated risk management system requirements for firms managing these strategies, and European frameworks such as AIFMD similarly expect robust, independent risk management and due diligence processes. But regulatory expectations at the portfolio level do not collapse the distinction at the deal level: the pre-approval evidence pack is an underwriting artefact, not a surveillance system, and it should be presented to committee as exactly that.
- Underwriting diligence: point-in-time, decision-bound, produces the evidence pack for the approval
- Ongoing monitoring: continuous, portfolio-level, tracks covenant compliance and borrower health after closing
- Separate tooling and cadence: a strong approval pack is not a monitoring system, and vice versa
For clarity about scope: the evidence pack supports the approval decision only. It does not provide continuous portfolio monitoring, automated credit approval or KYC, and nothing in a diligence workspace substitutes for the professional judgment of the credit committee and its advisers.
How to use this in your next diligence workflow
The five-part pack becomes repeatable when the workflow that builds it is repeatable. In practice that means four stages per deal, each with a defined output that feeds the next.
- Ingest early. Data Room Ingestion connects to the VDR and processes PDFs, spreadsheets, contracts and financial models within minutes, so the evidence base exists before analysis starts rather than accumulating in parallel with it.
- Analyse against the framework. The AI-Analysis Engine reads and cross-references the ingested documents, producing findings with source traceability, while Risk Radar ranks them by materiality, financial impact and deal relevance, which is how the add-back, leverage-mechanics and covenant-calibration tests from section three get executed consistently.
- Structure for committee. Report Builder drafts the committee-ready pack with full source traceability, turning the ranked findings into the five-part structure this article defines, including the evidence-gap log.
- Socialise and align. Collaboration Hub keeps underwriting, legal and advisory workstreams aligned on the same findings, so the version of the evidence the committee sees is the version the workstreams tested.
Teams that work this way spend less time assembling documents and more time testing them, which is where the credit decision is actually made. Plausity is built for today's investment and deal teams. Trusted by >200 firms. Credit and private capital funds run the same workflow across diligence workstreams. If your next private debt approval deserves an evidence pack that stands up to committee scrutiny, the workflow above is where to start.
How Plausity accelerates this workflow
Plausity is an AI-native due diligence and deal intelligence workspace that helps M&A advisory firms, VC and PE funds, corporate development teams and investment-banking teams structure evidence, findings and questions across a data room. Plausity supports evidence extraction, source grounding, findings management and IC preparation — it does not replace human analysts, advisers or investment professionals, does not provide legal, tax, audit, regulatory or investment advice, and does not make autonomous investment decisions. All findings require human review. Built for today's investment and deal teams. Trusted by >200 firms.
To explore the underlying capabilities, see the Plausity AI analysis engine and the findings and risk intelligence product page. For team-level workflows, see how VC and PE funds and M&A advisory firms use Plausity across live deals.
