BaseTrue Clear — Non-Bank ABL & Factoring Field Exam Desk

Done-for-you, examiner-released Field Exam & Borrowing Base Verification Reports for non-bank asset-based lenders and factoring companies — delivered in days at a fixed per-exam price, replacing multi-week CPA/consulting day-rate engagements, as fraud incidence and non-bank ABL portfolio growth both accelerate.

Final decision

BLUEPRINT

Proceed. Clears the evidence threshold and six-gate rubric (25/30). Zero semantic overlap found across the 594-run manifest for asset-based lending, factoring, borrowing base, or field exam terrain. Service-first, outcome-priced, AI-native collateral verification desk with credentialed field-examiner RELEASE — not a lending decision, not an audit opinion, not credit advice.

Executive summary

Non-bank asset-based lenders and factoring companies are contractually and practically required to commission independent field exams — on-site-turned-mostly-desktop reviews of a borrower's accounts receivable, inventory, and cash records against the loan agreement's borrowing base definitions — before closing a facility and on a recurring quarterly/semi-annual cadence thereafter. Today that work is bought almost entirely from small CPA/consulting boutiques (Young & Associates, Rosenberg & Fecci, LCG Advisors, ABLC, Tanner, Bank Advisors Ltd, Freed Maxick, Moore Colson, CFO Strategies) on day-rate/hourly billing, with multi-week turnaround. BaseTrue Clear sells a done-for-you Field Exam / Borrowing Base Verification Report per loan per exam cycle: AI extraction and normalization of AR aging, GL, bank statements and sales journals; deterministic testing against the lender's specific ineligible/concentration/dilution rules; AI fraud-pattern detection (duplicate invoices, ghost debtors, related-party dilution, sequential-number anomalies); and a credentialed field examiner's reviewed, signed RELEASE. The lender remains the sole credit decision-maker; BaseTrue never issues a GAAS/SSAE audit opinion, never gives credit or investment advice, and never contacts debtors to collect money.

Thesis

Field exams are a document-synthesis-and-rule-application workflow that a small, fragmented cottage industry currently prices and delivers like manual accounting fieldwork — hourly, slow, and inconsistent across examiners. AI can extract and reconcile AR aging, GL trial balances, and bank statements across dozens of incompatible ERP export formats; apply each lender's idiosyncratic borrowing-base ineligible definitions deterministically; and flag fraud patterns (duplicate invoice numbers, new-debtor concentration spikes, EIN/address reuse across borrowers) far faster than a human reviewing spreadsheets line by line. A credentialed examiner remains the trust chokepoint for exception review, debtor confirmation sampling, and RELEASE. As frontier models improve at multi-format document extraction and anomaly detection, unit cost falls and exam turnaround compresses from weeks to days — without the lender ever operating the AI directly.

Discovery rationale

This run began by reading manifest.json in full (594 prior entries) and keyword/semantic-scanning it for saturated terrain: regulatory-filing/completeness desks now blanket real estate, construction, hospitality, restaurants, logistics, education, elder/disability, self-storage/marina, unclaimed property, sales tax, appraisal QC, zoning, property management trust, franchise royalty, ERISA, ketamine-clinic DEA, and FQHC UDS reporting. Per orchestrator guidance, research was steered toward consumer financial services/banking-fintech ops, veterinary, dental, home services, staffing, franchise non-royalty ops, nonprofit/grant compliance, maritime, aviation, cannabis, funeral, stop-loss, dental/vision claims, and screening/FCRA — all cross-checked against the manifest and found either saturated (veterinary DEA x4, dental x7, HOA x10, FCRA x4, credentialing x7, franchise x16) or thematically adjacent to already-taken patterns. Twenty-plus targeted queries across five candidate sectors (below) surfaced non-bank ABL/factoring field exams as the strongest open wedge: a real, actively-outsourced, actively-hired-for function (3,290+ Glassdoor field-examiner listings) inside a fast-growing, fraud-exposed segment (non-bank ABL outstandings +12.6% QoQ) with zero manifest overlap on "asset-based lending," "factoring," "borrowing base," or "field exam."

Candidate comparison

Scoring used the same 1–5 x 20-factor rubric prior runs applied (novelty vs. manifest, MVP clarity, buyer willingness-to-pay evidence, distribution reachability, licensing safety, margin credibility, Sam Altman fit). BaseTrue Clear led on novelty, active-hiring demand evidence, and a document-only desktop wedge that cleanly avoids physical labor.

CODE validation

• Commercial trend (buyer-side of Consumer): Non-bank/independent commercial finance companies are growing faster than bank ABL books (non-bank outstandings +12.6% QoQ vs. bank -5.1% QoQ seasonal paydown, Q4 2025 SFNet) — this segment structurally lacks the in-house field-exam teams large banks maintain, so its exam demand flows entirely to outside providers.
• Opportunity: Field exams are billed like manual audit fieldwork (day-rate/hourly, multi-week turnaround) despite being 70%+ document extraction, reconciliation, and rule application — a textbook AI-native re-pricing and re-delivery opportunity, and the underlying documents (AR aging, GL exports, bank statements) are exactly the multi-format, messy inputs frontier extraction models handle increasingly well.
• Demand: A fragmented cottage industry of named firms (Young & Associates, Rosenberg & Fecci, LCG Advisors, ABLC, Tanner, Bank Advisors Ltd, Freed Maxick, Moore Colson, CFO Strategies) already sells this exact deliverable as a paid, outsourced service; "Field Examiner" is an active job title with 3,290+ current Glassdoor listings and a published $72k–$116k salary band (ZipRecruiter, 2026) — direct evidence lenders are paying real money (in-house or outsourced) for this function today.
• Economic sizing: $537B in US ABL commitments and $148B in annual factoring volume (SFNet, Q4 2025 study) imply, conservatively, tens of thousands of active facilities each requiring 1–4 field exams per year. Even an Inferred per-exam fee of $2,000–$6,000 (a fraction of the $5k–$25k+ CPA-firm day-rate range implied by industry commentary, itself Unverified as no firm publishes exact pricing) against a low-single-digit-percent share of the non-bank segment supports a credible multi-million-dollar ARR wedge before any factoring-side expansion.

Rubric scorecard (six gates)

Total: 25/30. Anti-commoditization: a lender handing raw exports to a generic chat model gets an unreliable one-off analysis with no accountable sign-off, no versioned per-lender rule library, no debtor-confirmation workflow, and no defensible audit trail for their own credit file — BaseTrue sells the accountable, repeatable, RELEASE-backed system, not a prompt.

Target buyer

ICP: US non-bank/independent asset-based lenders and factoring companies with $10M–$750M in committed facilities, 2–25 person credit/risk teams, running 10–200+ field exams per year, currently paying outside CPA/consulting firms on a day-rate basis with no dedicated in-house examiner headcount.

Economic buyer: Chief Credit Officer / VP Portfolio Risk / Head of Underwriting — owns the third-party exam budget line and the credit-quality outcome.

Champion: Portfolio Manager / Credit Analyst / Loan Operations Manager who schedules exams, chases outside firms for turnaround, and assembles the credit file.

Beachhead: Independent, non-bank commercial finance companies and factoring companies in the $20M–$300M portfolio range — the fastest-growing segment (SFNet: non-bank ABL outstandings +12.6% QoQ, Q4 2025) and the one least likely to have an in-house examiner team.

Jobs-to-be-Done

• When I'm underwriting a new ABL or factoring facility, get me an independent field exam back fast enough that it isn't the bottleneck to closing.
• When my loan agreement requires a quarterly or semi-annual field exam, deliver it on a predictable schedule and at a predictable price without me hiring a full-time examiner.
• When a borrowing base certificate looks off — dilution spike, aging deterioration, a new large debtor — get me a fast diagnostic before I have to decide on an advance or a hold.
• When my portfolio grows, give me exam capacity that scales without linearly adding headcount or CPA-firm day-rate spend.
• When fraud is trending up industry-wide, give me a systematic way to catch duplicate invoicing, ghost debtors, and related-party dilution before I've advanced against them.

Painful problem

Independent field exam and consulting firms bill field exams like manual audit fieldwork — day-rate or hourly, with engagements commonly running multiple business days on-site plus report-writing time, and turnaround measured in weeks rather than days (Inferred from the structure of every incumbent firm's marketing copy, none of which publishes fixed pricing or committed turnaround — a fragmentation signal in itself). That bottleneck lands at the worst possible moments: at new-loan closing, when covenant-driven periodic exams come due, and — increasingly — when a borrowing base certificate throws an early-warning signal that needs same-week diagnosis, not a four-week wait. Meanwhile 89% of factoring/ABL professionals report fraudulent activity increased over the past financial year, with duplicate invoicing and synthetic-identity/related-party schemes named as top 2025 threats requiring cross-lender data standardization (IFA Commercial Factor, 2025) — exactly the kind of pattern-matching-at-scale problem legacy manual field exams are worst positioned to catch quickly. Non-bank lenders, the fastest-growing segment of the market, feel this hardest: they rarely carry an in-house examiner team the way large banks do, so 100% of their exam capacity is purchased from the same small set of outside boutiques competing on relationships and availability, not speed or price.

The outcome we sell

A credentialed-examiner-released, lender-credit-file-ready Field Exam / Borrowing Base Verification Report for a specific loan and exam cycle: eligible/ineligible collateral schedule tested against that lender's own loan-agreement definitions, dilution and turnover trend analysis, a fraud-pattern exception memo (duplicate invoices, ghost debtors, related-party concentration, EIN/address reuse across borrowers), and recommended reserve/ineligible adjustments — delivered in days, not weeks, at a fixed per-exam price. BaseTrue never makes the lending decision, never issues an audit opinion under GAAS/SSAE attestation standards, and never contacts a borrower's debtors to demand payment. Success metric: exam delivered within the committed SLA and accepted into the lender's credit file without material rework; secondary metric: dollar value of ineligible collateral or fraud exposure identified and prevented from being advanced against.

First one-feature MVP wedge

Evidence summary

• Verified: US ABL commitments $537B / outstanding $210B; US factoring annual volume $148B / outstanding $20B (SFNet 2025 market-sizing study, released Feb 2026); non-bank ABL outstandings +12.6% QoQ and factoring volume +16.6% YoY (SFNet Q4/year-end 2025 indexes); 89% of factoring/ABL professionals report fraud increased over the past year, duplicate invoicing and synthetic identity named top 2025 threats (IFA Commercial Factor magazine); 3,290+ active "field examiner" job listings (Glassdoor) and $72k–$116k salary band (ZipRecruiter); at least nine named firms (Young & Associates, Rosenberg & Fecci, LCG Advisors, ABLC, Tanner, Bank Advisors Ltd, Freed Maxick, Moore Colson, CFO Strategies) actively marketing paid outsourced field-exam services; LAMA.ai markets a "Borrowing Base Certificate" feature inside a lender-operated origination platform, confirming buyer-side pain but as a co-pilot the lender runs, not a done-for-you service.
• Inferred: Typical per-exam CPA/consulting-firm pricing lands in the $5,000–$25,000+ range with multi-week turnaround (no firm publishes exact fixed pricing; inferred from engagement-scope descriptions and industry commentary); willingness to pay $1,800–$9,000 for a faster, fixed-price AI-native alternative when the covenant deadline or closing timeline is tight; no scaled national AI-native DFY competitor yet serving this specific niche.
• Unverified: Exact national count of active ABL/factoring facilities requiring annual field exams; exact average field-exam fee charged by incumbent firms; conversion rate from free BBC Red-Flag Scan to paid exam.

Note on conflicting market data: IBISWorld's narrowly-defined "Invoice Factoring" industry classification (firms whose primary NAICS-coded revenue is factoring fees) shows a small, declining $3.0B revenue base (247 businesses, -8.2% CAGR in business count) — this is a caution flag, addressed directly in the risk register. The trade-association SFNet data (which captures ABL and factoring activity across banks, non-banks, and specialty finance companies broadly, the standard industry benchmark) shows a far larger and growing $537B/$148B market. BaseTrue's ICP and TAM are anchored on SFNet's broader, volume-based figures and the mid-market non-bank segment specifically, not on IBISWorld's narrow factoring-fee-revenue category.

Claim table

Source-claim matrix

Market and demand evidence

SFNet's 2025 market-sizing study (the secured-finance trade association's benchmark data set, released February 2026) puts total US secured finance outstanding at $12.2 trillion, with $537 billion in ABL commitments ($210 billion outstanding) and $148 billion in annual factoring volume ($20 billion outstanding). The same organization's year-end 2025 index shows non-bank ABL outstandings surging 12.6% quarter-over-quarter while bank ABL outstandings declined 5.1% on seasonal paydowns — the non-bank segment is both growing fastest and structurally least likely to carry an in-house field-exam team the way a large bank does. Demand for the underlying function shows up concretely as an active, well-compensated job category (3,290+ "Field Examiner" listings on Glassdoor; $72k–$116k salary band per ZipRecruiter) and as a small but entrenched cottage industry of named firms selling exactly this deliverable today. Fraud pressure is rising in parallel — 89% of surveyed factoring/ABL professionals report increased fraudulent activity, with duplicate invoicing and synthetic-identity schemes flagged as the top 2025 threats requiring better, faster detection than manual spreadsheet review provides.

Active buyer conversations

• IFA Commercial Factor (trade magazine of the International Factoring Association) publishing industry-wide calls for standardized, collaborative fraud-defense data practices — a direct signal the current manual/fragmented approach is recognized as inadequate by practitioners.
• SFNet's quarterly ABL and Confidence Index releases and participant surveys, which capture ongoing sentiment and operational metrics (DSO, credit quality, commitment growth) directly from lender risk teams.
• Incumbent field-exam firms' own marketing content (Rosenberg & Fecci, Moore Colson, LCG Advisors) explaining "what a field exam is" and "the good, the bad, and the ugly" — education content aimed at buyers who are actively evaluating or re-evaluating exam providers.
• Active hiring for in-house "Field Examiner" and "Commercial Loan Field Examiner" roles (Indeed, SimplyHired, Glassdoor, ZipRecruiter) — direct labor-market evidence that lenders are currently solving this problem either by hiring or by paying outside firms, both of which are more expensive and slower than an AI-native done-for-you alternative.

Competitive landscape

• CPA/consulting field-exam boutiques (Young & Associates, Rosenberg & Fecci, LCG Advisors, ABLC, Tanner, Bank Advisors Ltd, Freed Maxick, Moore Colson, CFO Strategies) — the direct incumbents; manual, day-rate/hourly billed, multi-week turnaround, examiner-availability-constrained.
• In-house examiner teams — the default at large banks; largely unavailable to the non-bank/independent lender segment BaseTrue targets, which is exactly why they buy externally.
• Lending-origination SaaS with borrowing-base features (e.g., LAMA.ai) — sells software the lender's own team operates to build/track a borrowing base certificate; a co-pilot, not a done-for-you verified exam report, and does not replace the need for an independent third-party exam.
• Generic AI chat tools — can summarize a spreadsheet but cannot hold a versioned per-lender ineligible-rule library, run debtor confirmation workflows, or provide an accountable, signed RELEASE a credit committee can rely on.

Competitor and budget validation

Budget already exists and is being spent today as: (1) CPA/consulting-firm field-exam fees per engagement; (2) in-house examiner salaries ($72k–$116k) at lenders large enough to staff the function; (3) portfolio-manager/credit-analyst hours spent scheduling, chasing, and reviewing outside-firm output; (4) fraud losses from undetected duplicate invoicing and related-party dilution that a faster, more systematic exam would catch earlier. BaseTrue redirects a slice of (1)–(3) and reduces (4) — the win condition is materially faster turnaround and a lower, fixed per-exam price than the day-rate alternative, not the absence of competitors.

Pricing evidence and proposed pricing

Never hourly. Price anchors to the avoided cost and delay of the CPA/consulting-firm day-rate alternative (Inferred $5,000–$25,000+ per exam, multi-week turnaround) and to the dollar exposure a single undetected ineligible-collateral or fraud finding represents. The existing job market for in-house examiners ($72k–$116k/year, i.e., roughly $35–$56/hour loaded) further calibrates that lenders already value this function highly enough to staff or outsource it continuously.

Regulatory and compliance considerations

The core regime is contractual (loan-agreement covenants defining the borrowing base and requiring periodic independent field exams) reinforced by bank regulatory expectations around collateral monitoring and leveraged-lending risk management for regulated lenders. A field exam report is not an audit or attestation engagement under GAAS/SSAE standards, not a legal opinion, and not investment or credit advice — BaseTrue's report is explicitly scoped and labeled as an independent collateral-verification work product, matching the exact deliverable category incumbent CPA/consulting firms already sell. Debtor confirmation sampling must be scripted strictly as an audit-style confirmation ("please confirm this invoice amount and that you are the obligor"), never as a payment demand, to avoid any appearance of debt-collection activity. Sensitive borrower financial data, debtor PII, and bank statement data require SOC2-aligned handling controls, encrypted storage, and contractual data-protection commitments with each lender client.

Licensing boundary

Required disclaimers: "This Field Exam Report is an independent collateral-verification work product prepared for internal credit-file use. It is not an audit or attestation engagement under GAAS/SSAE, not a legal opinion, and not investment, credit, or lending advice. It does not guarantee collectability of collateral or repayment of any loan. The lender retains sole responsibility for all credit and lending decisions." UPL/tax/medical/insurance/credit/debt-collection/immigration/privacy risk analysis: No legal advice or representation is given (not law practice); no tax positions are taken (not tax practice); no medical or insurance claims are adjudicated; the report informs but never makes the lender's credit decision (not credit counseling or advice); debtor confirmation calls are scripted as audit-style verification only, never a payment demand (not debt collection, no FDCPA exposure); no immigration matters are touched; borrower and debtor financial/PII data is handled under SOC2-aligned controls with data-minimization and contractual confidentiality commitments (privacy risk mitigated, not eliminated).

AI-native advantage

AI compresses the multi-day, spreadsheet-by-spreadsheet reconciliation that currently consumes most of a field exam's billable hours: reading dozens of incompatible ERP export formats, matching AR aging line items to the sales journal and cash receipts journal, computing dilution and turnover trends, and flagging duplicate-invoice or related-party patterns across thousands of line items. A frontier model doing multi-document extraction and anomaly detection turns a multi-day manual task into a same-day AI-assisted first pass, leaving the credentialed examiner to spend their scarce time on the handful of genuine exceptions and the debtor confirmation calls that actually require judgment. Every exam outcome (which flags were real, which were false positives) feeds back into the fraud-pattern model and the per-lender rule library, so exam quality and speed both improve with volume — a genuine, compounding AI-native advantage rather than a thin UI wrapper on a CPA firm's checklist.

Internal AI engine architecture (10 layers)

1. Intake — secure portal upload (AR aging, GL trial balance, sales journal, cash receipts journal, bank statements, borrowing base certificate history, debtor concentration list, loan agreement).
2. Normalization — parse and standardize across incompatible ERP/accounting export formats (QuickBooks, NetSuite, Sage, custom CSVs, scanned PDFs).
3. Retrieval/knowledge — versioned per-lender loan-agreement borrowing-base definitions; ineligible/concentration/dilution rule cards; historical exam library for the same borrower.
4. AI workbench — AR-to-GL-to-bank-statement reconciliation; dilution/turnover trend computation; fraud-pattern detection (duplicate invoice numbers, sequential/altered invoice dates, new-debtor concentration spikes, EIN/address reuse across borrowers).
5. Deterministic rules — cross-aging thresholds, concentration caps, contra/related-party exclusions, foreign-obligor and bill-and-hold restrictions per the specific loan agreement.
6. Human chokepoint — credentialed field examiner reviews AI-flagged exceptions, runs debtor confirmation sampling, and RELEASEs (or rejects/requests more data).
7. QA — secondary reviewer spot-checks a sample of released exams; red-team scan for missed ineligible categories.
8. Delivery — Field Exam Report PDF + exception schedule spreadsheet + portal delivery to lender credit file.
9. Learning loop — outcomes tagged (confirmed fraud, false positive, ineligible upheld/overturned) feed back into fraud-pattern models and per-lender rule libraries.
10. Model-portability — provider-agnostic extraction/LLM adapters; deterministic rules engine kept separate from model weights so the model backend can be swapped without re-authoring rules.

AI-vs-human operations pipeline

Dynasty translation layer

• Buyer: Non-bank ABL lender/factoring company CCO or VP Portfolio Risk paying to protect collateral integrity and close exams on time.
• Service: DFY Field Exam / Borrowing Base Verification Report; automated extraction + credentialed examiner RELEASE.
• Workflow: Intake → normalize → deterministic rules → AI fraud/dilution flags → examiner exception review & debtor confirmation → RELEASE → deliver to credit file → outcome-tag → learn.
• Tooling: Day-1: secure Drive/Dropbox intake, Airtable pipeline, spreadsheet rules calculator, Claude/GPT extraction, Stripe billing; later: dedicated intake portal with direct ERP export.
• Sales: "We deliver your field exam in days, at a fixed price, before your covenant deadline — not weeks, not day-rate billing."
• Delivery: Manual specialist-led pipeline first; automate extraction/reconciliation/rule-testing next.
• Expansion: On-site inventory addendum via partner network; portfolio-wide multi-loan programs; continuous BBC Assurance monitoring retainers; factoring-specific fraud verification-as-a-service add-on.

Anti-duplication analysis

Read manifest.json in full (594 prior runs, every slug/market/buyer/workflow/outcome_sold field) and keyword/semantic-scanned it for "asset-based lending," "factoring," "borrowing base," "field exam," and adjacent terms. Result: zero matches on any of these terms except one unrelated, tangential mention of "factoring companies" as a phase-2 channel buyer inside carrier-detention-accessorial-recovery-desk (a trucking accessorial-collections business, entirely different buyer, workflow, and outcome). The manifest's financial-services entries (aml-alert-sar-investigation-engine, workers-comp-msa-section111-engine, subrogation-recovery-engine, ucc-lien-perfection-monitoring-engine, appraisal-review-collateral-qc-engine, mid-market-aged-unapplied-cash-exception-desk) touch adjacent commercial-finance/collateral themes but serve different buyers (banks' AML teams, WC insurers, subrogation claimants, UCC filers, appraisal reviewers, AP teams) with different workflows and different outcomes sold. None address ABL/factoring field exams or borrowing base verification. BaseTrue's buyer + workflow + outcome combination is confirmed open.

Anti-commoditization analysis

If frontier models eventually let a lender's own analyst run extraction and dilution math themselves, BaseTrue still wins on: a continuously-updated per-lender ineligible/concentration/dilution rule library built from real loan agreements; a cross-client fraud-pattern dataset (duplicate invoice numbers, EIN/address reuse) that only grows more valuable with more exams processed; a licensed local inspector partner network for the on-site inventory addendum; and — most importantly — an accountable, credentialed examiner RELEASE that a lender's own credit committee and bank regulators can rely on, which a self-service chat session cannot provide. The product is the accountable verification system and its compounding fraud-pattern data, not a one-off extraction prompt.

Service delivery workflow

1. Lender triggers via portal request (new-loan, periodic covenant exam, or early-warning BBC anomaly).
2. Secure upload of AR aging, GL, bank statements, sales journal, cash receipts journal, BBC history, loan agreement.
3. AI extraction, normalization, and deterministic ineligible/dilution/fraud-pattern first pass within hours.
4. If data gaps: structured request back to lender/borrower for missing exports.
5. Examiner reviews AI-flagged exceptions, runs debtor confirmation sampling on a risk-weighted subset.
6. Examiner RELEASE within committed SLA (standard: 5 business days; rush SKU: 2 business days for closing-critical exams).
7. Report delivered to lender credit file; outcome tagged for the learning loop.

Operations as product

Per-lender SOPs capturing each loan agreement's exact borrowing-base definitions; standardized intake checklists per ERP/accounting system; automated completeness checks before an exam enters examiner review; exception queues segmented by severity (fraud flag vs. routine ineligible vs. missing data); reviewer assignment by lender relationship and complexity tier; confidence scoring on every AI-flagged exception; full audit trails from raw upload to RELEASEd report; version-controlled per-lender rule libraries; gold-standard exam examples for training and calibration; red-team checks simulating known fraud patterns; standardized report/exception-schedule templates; root-cause postmortems on any exam requiring lender-requested rework.

No-holes quality engine

• Hard stop if a loan agreement's borrowing-base definitions cannot be confidently parsed — routed to examiner for manual rule entry before any automated testing proceeds.
• Hard stop if AR aging totals do not tie to the GL trial balance within tolerance — flagged as a data-integrity exception, not silently reconciled.
• Every AI-flagged fraud pattern requires examiner review before appearing in the final report as a finding (no unreviewed fraud allegations reach a lender).
• Debtor confirmation sample size scales with the size and concentration of the receivables being tested.
• Two-person rule (examiner + senior reviewer) on any exam where total collateral tested exceeds $10M or where a fraud finding is included.
• No exam RELEASEs with an open, unresolved data-integrity exception.

What the human expert actually does

Minimum viable offer

"5-Day Desktop Field Exam" for one active ABL or factoring loan: AR aging + GL + bank-statement reconciliation, ineligible/dilution testing against the client's own loan agreement, fraud-pattern scan, examiner RELEASE. Founding price $2,200 flat. Money-back if the committed 5-business-day SLA is missed (not if the exam surfaces adverse findings). Free BBC Red-Flag Scan as lead magnet.

Fulfillment process (first 3 customers)

1. Manual Airtable pipeline + secure Drive folder per loan/exam cycle.
2. Founder/examiner runs AI extraction (Claude/GPT with document parsing) and pastes normalized output into a spreadsheet rules calculator.
3. Rules calculator applies that lender's specific ineligible/concentration/dilution definitions.
4. Examiner manually reviews flagged exceptions, places debtor confirmation calls/emails, drafts findings.
5. Report assembled from template; delivered via secure link. No custom software until after 5 exams with timed COGS data.

Tools and systems

Day 1: secure file-sharing (Drive/Dropbox with access controls), Airtable pipeline, Stripe billing, Calendly for intake calls, Slack/email for lender communication, document-extraction LLM APIs, spreadsheet rules calculator, PDF report assembly. Later: dedicated secure intake portal with direct ERP/accounting export ingestion, automated debtor-confirmation workflow, and a per-lender rule-library management interface.

Human-in-the-loop quality control

No report ships without a credentialed examiner's RELEASE. Auto-flag codes force examiner attention on any unresolved data-integrity exception or AI-detected fraud pattern before a report can be marked complete. Weekly calibration sessions on anonymized gold-standard exams across the examiner team. Lenders cannot compel RELEASE of an exam we've flagged as incomplete or data-insufficient — protects report reliability and BaseTrue's credibility with credit committees.

Nonlinear scaling and unit economics

COGS stack: model inference/extraction, secure storage, examiner minutes, secondary QA sample, debtor-confirmation communication costs, payment processing, rework. Path to 50%+ gross margin is credible by day 90 as extraction and rule-testing automation absorb the majority of what is currently manual reconciliation labor. Revenue per FTE target year-1: $300k–$500k with a mix of per-exam and BBC Assurance retainer revenue.

Distribution proof table

Sales and outreach plan

Lead with diagnosis, not demo: free BBC Red-Flag Scan → 20-minute review call → single Desktop Field Exam offer if a covenant deadline or closing is near → BBC Assurance retainer offer after 2 successful exams. Outbound personalization uses public deal announcements, SFNet directory data, and firmographic signals only — never scraped confidential borrower data.

Founder-led content plan

Teach: how non-bank ABL/factoring field exams actually work; the SFNet 12.6% non-bank growth story and what it means for exam capacity; duplicate-invoice and related-party dilution fraud patterns with anonymized teardown examples; borrowing base ineligible-category checklists; dilution and turnover trend interpretation; "what a $5,000 day-rate exam actually costs you in delay" cost-of-delay math. No generic AI hype — every post ties to a real loan-agreement mechanic or fraud pattern.

First 30 days of content

1. 10 posts: what a borrowing base certificate actually tests; the 12.6% non-bank ABL growth story; duplicate invoice fraud anatomy; related-party dilution red flags; cross-aging vs. concentration caps explained; new-loan exam bottleneck cost-of-delay calculator; debtor confirmation sampling best practice; "field exam" vs. "audit" — what a lender actually gets; five things that get an exam kicked back for rework; BBC Assurance monitoring ROI walkthrough (hypothetical, labeled).
2. 3 diagnostic teardowns: anonymized duplicate-invoice pattern catch; anonymized related-party concentration spike; anonymized cross-aging miscalculation in a self-prepared BBC.
3. 2 lead magnets: Borrowing Base Ineligible-Category Checklist (PDF); Field Exam Cost-of-Delay Calculator (spreadsheet).
4. 1 live review webinar: "Non-bank ABL growth and the field-exam capacity gap — 2026 outlook."
5. 1 outbound diagnosis template: loan-agnostic exam-readiness memo scoring a lender's current BBC data completeness before any engagement.

Lead magnet and waitlist plan

Magnet: Free BBC Red-Flag Scan + Borrowing Base Ineligible-Category Checklist. Upload one borrowing base certificate and AR aging; receive a scored preliminary flag list before paying for a full exam. Waitlist CTA for BBC Assurance continuous monitoring. Scan intake captures portfolio size, current exam provider, and last-exam turnaround (pain signal). Sales-ready = an exam due within 30 days or portfolio >$20M with no in-house examiner.

Warm GTM plan

Convert checklist/calculator downloaders; ask commercial-finance broker contacts for warm intros to newly-funded non-bank lenders; offer founding pricing to the first 10 lender clients; host an SFNet-adjacent lunch-and-learn on the fraud-pattern-detection angle.

Targeted outbound plan

40 personalized notes/week to CCOs and VP Risk at non-bank ABL/factoring firms with $20M–$300M portfolios: reference the SFNet 12.6% non-bank growth stat plus a specific exam-turnaround pain point, offer a free scan. No spray demos. Follow-up = Loom walkthrough of a sample (synthetic-data) exam report.

Answer-engine / search visibility plan

Pages targeting: "ABL field exam turnaround time," "borrowing base certificate verification service," "asset-based lending field exam cost," "invoice factoring fraud detection AI," "non-bank ABL field examiner." Structured FAQ content citing SFNet and IFA data directly; updated as new SFNet quarterly indexes are released.

Pilot design and early-demand-trap mitigation

• Pilot cap: 8 lenders / 20 exams before expansion.
• Incentive: founding exam price; not unlimited custom rule-building for non-standard loan agreements.
• Measure: exam-delivery SLA hit rate, examiner minutes/exam, Scan→Exam %, rework rate, fraud-finding accuracy (confirmed vs. false positive).
• Refuse custom scope creep into full GAAS audit engagements or legal-opinion work.
• Do not hire additional examiners to paper over broken intake/rules SOPs — pause new logos at caps instead.

Early-access feedback flywheel

Every exam outcome tagged (confirmed fraud, false positive, ineligible upheld/overturned, rework requested). Rework triggers a new QA rule within 72 hours. Confirmed fraud patterns feed the shared cross-client fraud-pattern model. Lender-reported friction on intake or report format changes the checklist/template. Custom-rule requests are gated by a founder+examiner change-control review before being added to any per-lender rule library.

Build-before-scale checkpoints

• After 5 pilots: harden intake checklist, data-integrity tie-out checks, and examiner exception-review SOP.
• After 10 pilots: harden per-lender rule library structure, exception queues, reviewer assignment, and report templates.
• After 20 pilots: pause new logos until COGS, rework rate, escalation rate, and cycle time are measured and within target ranges.

7-day / 30-day / 90-day launch plans

7-day: Rule-library template v1 (three common ABL/factoring borrowing-base structures); exam report template; Airtable pipeline; 20 outbound notes; publish ineligible-category checklist; 3 free scans delivered.

30-day: 6 paid Desktop Field Exams delivered; first broker referral; webinar hosted; fraud-pattern detection rules v1; measure examiner minutes/exam.

90-day: 20 exams completed; 4 BBC Assurance retainers active; day-90 COGS ≤$560/exam; rework rate <12%; decide on dedicated intake-portal build vs. staying manual.

Metrics and KPIs

Scans, Scan→Exam conversion, SLA hit rate, examiner minutes/exam, rework rate, quality failure rate, escalation rate, confirmed-fraud accuracy rate, gross margin, BBC Assurance retention, referral %, NPS, data-integrity exception rate.

Risks and mitigations

Confidential borrower/debtor data handling risk → SOC2-aligned controls, encryption, contractual DPAs. Perceived audit-opinion overreach → explicit non-audit disclaimers on every report. Adverse selection (lenders bringing only already-troubled loans) → transparent scope; decline if data is too incomplete to test. Fraud false-positive damaging borrower relationships → dual sign-off before any fraud finding ships. Seasonality/covenant-cycle lumpiness → BBC Assurance monthly retainers smooth revenue. Incumbent CPA firms add AI tooling → compete on speed, fixed pricing, and the compounding fraud-pattern dataset, not on being first to use AI.

Exhaustive risk register

What could kill this

• Scan→Exam conversion <8% after 60 scans.
• Quality failure rate (lender-flagged material error) ≥5% after 20 delivered exams.
• Examiner minutes/exam fail to compress below 180 minutes after hardening, with no visible automation path.
• Lenders' internal compliance/legal teams refuse to accept a non-GAAS third-party report into their credit file at scale.
• Cannot acquire lender clients under a sustainable CAC after 90 days of outbound/content effort.

Go/no-go reasoning

GO. Clear, well-defined buyer with an active job market and an established outsourced-spend pattern for the exact deliverable; real, growing, fraud-exposed $537B/$148B market segment (SFNet 2025); zero overlap across the 594-run manifest; narrow, physical-labor-free desktop MVP with a clean addendum path for the one physically-dependent piece; credible 50%+ gross margin path as extraction/rule-testing automation matures; licensing boundary manageable via explicit non-audit, non-legal, non-collection scope disclaimers matching how incumbent field-exam firms already position this exact service.

Final recommendation

Launch BaseTrue Clear as a service-first, examiner-released Field Exam & Borrowing Base Verification Desk for non-bank ABL lenders and factoring companies. Start with desktop-only exams (no on-site inventory) for the $20M–$300M non-bank segment, founding pricing, an 8-lender pilot cap, a Free BBC Red-Flag Scan as the lead magnet, and disciplined SLA/rework/fraud-accuracy metrics before scaling into on-site addenda and BBC Assurance retainers.

Source list

1. SFNet — 2025 Market-Sizing Study: Secured Finance Surges Past $12 Trillion
2. SFNet — Year-End 2025 ABL and Factoring Performance
3. SFNet — Q1 2025 ABL and Confidence Indexes
4. SFNet — Asset-Based Lending & Factoring Surveys
5. IFA Commercial Factor — Data-Led Fraud Defense in US Invoice Factoring
6. eCapital — How Factoring Companies Identify Fraudulent Invoices
7. IBISWorld — Invoice Factoring in the US Industry
8. GM Insights — Asset-Based Lending Market
9. Rosenberg & Fecci Consulting — ABL Field Examinations
10. Rosenberg & Fecci — Purpose of a Lender's Field Examination
11. Young & Associates — ABL Field Exam Services
12. Young & Associates — Importance of Field Examinations in ABL
13. LCG Advisors — What Does a Field Exam Involve in Asset-Based Lending?
14. Asset Based Lending Consultants — Field Examinations
15. Tanner — Field Exam Services
16. Bank Advisors Ltd — ABL Field Exams & Asset Quality Review
17. Freed Maxick — Asset-Based Lending Field Examinations
18. Moore Colson — Field Exams: The Good, the Bad, and the Ugly
19. CFO Strategies — Field Examination Services
20. Private Equity Bro — ABL Borrowing Base, Reserves & Field Exam Essentials
21. Glassdoor — Field Examiner Jobs
22. ZipRecruiter — Field Examiner Jobs & Salary
23. LAMA.ai — Borrowing Base Certificate Platform Feature