Loan tape validation, explained.

Why a buyer checks a seller's loan tape before paying for it

The life of one portfolio purchase, in five steps

1. The seller sends a tape and opens a data room48 loans, $562.8 million of principal, 61 columns, and 229 documents: credit agreements, 83 amendments, security agreements, registers and two intercreditor papers.
2. The buyer cleans and maps the tape61 seller columns become the buyer's own terms. Dates and spellings are put in one format, and the totals are added up and matched to the seller's.
3. Every cell is tied to its document384 cells: balance, benchmark, margin, floor, maturity, collateral, lien position and title for each loan. This is where the 2029 maturity, the two pledged loans and the five repricings are found, or missed.
4. The breaks become a price and a scheduleLoans left out, prices cut, part of the price held back, and every break listed in the purchase agreement's schedule of exceptions.
5. Committee approves, the deal closes, and the tape comes back every monthAfter closing, each new statement from the loan administrator is checked against the same documents.

Step 3 is the subject of this guide. The funds, borrowers, people and figures in this example are fictional.

Loan tape surprises: thin data promises and a fraction-of-a-percent sample

Loan tape, as amended, break and schedule of exceptions

Fourteen terms are defined in the one-page summary.

Eight terms on one row: what a corporate loan tape claims, and which document decides each

In shortA corporate loan tape carries dozens of columns, but eight of them decide what the buyer is paying for: balance, benchmark, margin, floor, maturity, collateral, lien position and title. Each has one document that decides it. The work of validation is knowing which document that is, finding its latest signed version, and reading the clause.

Sellers' tapes vary. The fictional tape in this guide has 61 columns: identifiers, sponsor and industry codes, ratings, leverage and covenant figures, and the contract terms. The analytical columns matter for pricing, but they are the seller's own estimates. The eight contract terms are facts a document either states or does not, and those are what a buyer can check cell by cell. Eight terms on 48 loans make 384 cells.

Three checks that are often confused

What one data room holds

The example has 83 amendments for 48 loans, almost two a loan, and one loan has four. A reader who opens the credit agreement and stops there will get the terms of 2023, and the buyer is paying for the terms of 2026.

Which rules reach a loan tape review: auditing standards, SEC disclosure rules and the UCC

In shortNo federal rule tells a private fund how to check a tape before it buys loans. The rules that shape the work come from three places: the accounting profession's standard for agreed-upon procedures, the SEC's disclosure rules for rated asset-backed securities, and the UCC's rules on who has the first claim to a loan. Banks that finance portfolios add their own supervisory guidance, which changed in 2026.

The rating agencies' own criteria

When a pool is securitized and rated, the rating agency sets what review it expects, and the diligence firm certifies which criteria it followed. One 2025 certification lists five, by name and date: DBRS's "Third-Party Due-Diligence and Representations & Warranties Criteria for U.S. RMBS Transactions" (September 30, 2024), Fitch's "U.S. RMBS Rating Criteria" (October 26, 2023), KBRA's "U.S. RMBS Rating Methodology" (December 7, 2023), Moody's "Global Structured Finance Data Quality Evaluation Approach" (February 9, 2022), and S&P's "Global Methodology And Assumptions: Assessing Pools Of Residential Loans" (January 25, 2019) with its U.S. and Canada supplement of February 21, 2025.5

Moody's title names the concern they share: the quality of the data before the pool is modeled. These criteria were written for mortgage pools and are published to subscribers. A secondary purchase of corporate loans has no equivalent rulebook. The buyer and its counsel decide how much to read, and the purchase agreement decides what happens when the tape is wrong.

The latest signed document decides: reading a credit agreement and every amendment in order

In shortA credit agreement is the starting point, and each amendment rewrites part of it. To know a loan's terms today, read every executed amendment in date order and carry each change forward. Balances come from the loan administrator's register, rank from the intercreditor agreement, and title from the public filings. In the example, half of the 22 breaks were set by an amendment.

Credit agreements for mid-sized companies run to 150 or 200 pages, with the defined terms at the front. Amendments are short: often a dozen pages that say "the definition of 'Maturity Date' is amended and restated in its entirety" and give the new text.

Amendments are signed for many reasons. A borrower buys another company and needs its lenders' consent. It is about to fail a leverage test and asks for a reset. Its credit has improved and it asks for a lower margin. Each one can change a term on the tape, and each one is a separate file in the data room.

One loan, three documents, one changed term

Halvorsen Fire & Life Safety, from the fictional example. The tape carried the 2031 date from the first document.

Which document decided each break in the example

Rules for reading a loan's terms

Six tests a tape cell must pass, and how each break moves the purchase price

In shortEach term has its own test: balances within a small tolerance, margins and floors exact, dates exact after allowing for business-day rolls, collateral and rank as written, title clear in every filing. Each break then has a remedy the purchase agreement sets: exclude the loan, buy it at the true balance, cut the price for a lower margin, price a last-out share lower, hold back money for a missing document, or disclose it with no price change.

Format differences are worth showing and never worth counting. In the example, 12 of 384 cells differed only in how they were written; counted as breaks, they would have buried the 22 that matter.

From 384 cells to three loans left out

The arithmetic, on the example's numbers

Par is a loan's principal amount. Loan prices are quoted as a percentage of par, so a price of 97.25 means 97.25 cents for each dollar of principal.

Fictional example. The price rules (three-year cap on repricing cuts, 93.00 for a last-out share, exclusion of pledged and second-lien loans, holdbacks for missing documents) are this purchase agreement's own terms; every agreement sets its own.

Three loans make up most of the money. The other breaks still matter. The portfolio manager who owns the loans after closing builds cash-flow forecasts from the maturities, margins and floors; a maturity two years early or a margin 0.75% lower is wrong in every forecast until someone corrects it.

The benchmark test after LIBOR

Many floating-rate loans used US dollar LIBOR as their benchmark until its panels ended after June 30, 2023; contracts then moved to rates built on SOFR.11 For contracts the federal LIBOR Act moved, the law adds a fixed "tenor spread adjustment", 0.26161% for three-month LIBOR.24 Loans amended to Term SOFR before the cutover carry whatever their amendment says, with or without an adjustment. A tape that shows "SOFR + 5.75%" for a loan whose contract says SOFR plus an adjustment plus 5.75% understates its coupon, and only the amendment that made the switch shows it.

What a sample of loan files can tell a buyer, and what it cannot

In shortSampling works when errors are common and alike: a clean sample of 150 loans bounds the error rate below about 2%. It works badly for rare, costly breaks in a small portfolio. A buyer that reads 10 of 48 files misses both pledged loans about six times in ten. Securitization filings show accountants reading 0.2% to 0.35% of a pool, and one of them declining to say whether that was enough.

How much of a pool the accountants read

Both reports are honest about their limits. One accountant compared eight characteristics, such as the rate, the term and the first payment date, with each sampled loan contract.3 The other compared more than twenty, and found one: a contract that listed a car's primary use as "Agricultural".4

That report says the firm makes "no representations as to the adequacy of the sample size" and draws no conclusions about the whole tape.4 Both say the work did not satisfy any rating agency's due diligence criteria.3,4 For a pool of 40,000 small, standard loans, a clean random sample says a lot. For 48 negotiated loans, each different, it says much less.

The arithmetic of a clean sample

Rating-agency criteria for mortgage pools set sample sizes from a confidence level and a tolerable error rate; DBRS's 2017 update moved to "a single sample approach" at "the same confidence level and threshold error rate".25 Some 2025 mortgage reviews read every loan instead: the diligence firm in one certification reviewed "100% of the loan population".5 A corporate loan tape is short enough that reading everything is a staffing question. Forty-eight loans with 229 documents is, by our rough estimate, weeks of reading for one person. That is why buyers sample, and why breaks get through.

The nine kinds of break a seller's tape can carry, and what catches each

In shortTapes go wrong in four places: amendments the seller's system never captured, cash events the register has and the tape lacks, rank and title facts that live outside the credit agreement, and documents missing from the data room. Most breaks are honest. A few are not, and they look the same on the tape.

How often a tape disagrees with its files

No public study measures the error rate of corporate loan tapes. The nearest published evidence comes from securitizations, where diligence firms file what they found. One full review of a 2025 residential mortgage pool is a useful marker, because it read every loan.

A residential mortgage pool, cited as evidence that tapes and files disagree often, never as the rate for corporate loans. The diligence firm said the debt-ratio differences came mostly from income calculations and that the ratios "generally remained within guidelines".5

Title and rank breaks: rarer than balance breaks, and costlier

Why tapes go stale faster in 2026: repricings, extensions and interest paid in kind

In shortThe more often loans are amended, the faster a tape drifts from its contracts. Valuation firm Lincoln International counted amendment activity up 13% in the last quarter of 2025, and the share of loans paying interest in kind up from 7% in 2021 to 11% in 2025. Each amendment and each capitalized interest payment is a cell that a tape built last quarter can no longer be trusted on.

Paying in kind (PIK) means the borrower adds interest to the loan's principal instead of paying it in cash. Lincoln, which values loans for lenders and covers about 1,800 private companies in its index, found 11% of the loans it values paid PIK interest in 2025, up from 7% in 2021, and calls the share that switched to PIK after closing, 6.4% in the fourth quarter of 2025, a "shadow default rate".26 The Financial Stability Board's 2026 report put PIK in about 12% of loans, with options to switch to PIK in about half of those.14

For a tape, PIK has a simple effect: the balance grows every period, and a balance typed at the last cut-off is already low. In the example, one loan's tape balance was $96,220 below the register because interest had been capitalized; the buyer pays more for it, $93,574, and the price is right.

Amendments hide in the default numbers too

Lincoln's senior debt index has warned that "a higher default rate may be disguised by significant amendment activity to avoid potential defaults".27 For a buyer, that means the loans most likely to have been amended recently are also the ones whose credit has moved, and the ones a tape is most likely to have wrong. Covenant resets in exchange for an earlier maturity, like the Halvorsen loan in the example, are exactly this trade.

Pledged twice: what Tricolor, First Brands and three other cases teach a loan buyer

In shortIn the 2025 failures, lenders and investors relied on loan and receivable data the borrowers controlled. Prosecutors and the SEC allege collateral was pledged to more than one lender and loan data altered to make bad loans eligible. A second pledge shows up in the lien filings, their releases and the loan administrator's records, and nowhere on the tape.

Why the first filing wins

A buyer of a loan is protected the moment the sale is made: under UCC §9-309, the interest of a buyer of a payment intangible or a promissory note is among those "perfected when they attach", with no filing needed.20 That protection does not beat a lender that got there first. Under §9-322(a)(1), conflicting perfected security interests "rank according to priority in time of filing or perfection".1

A seller's lender whose financing statement (the public UCC-1 filing that records its claim) covered "all loans now or later owned" in 2025 ranks ahead of a buyer in 2026 unless it released the loan, and a release can be partial: §9-512 lets the secured party file an amendment that deletes some collateral.34 So the question for each loan is whether it appears on the schedule the release kept back.

The search has to cover the right names in the right states. A registered organization such as a Delaware LLC is located in its state of organization, so filings against a Delaware financing vehicle are found in Delaware.21 Funds often hold loans through such vehicles, and a search against the fund's own name alone misses them.

What the industry says went wrong

The Secured Finance Network's 2026 task force surveyed 29 member organizations; 22 named "less stringent monitoring and controls" as a primary vulnerability.35 An asset manager's review of First Brands and Tricolor recommends that lenders "Perform daily tie-outs between collections and positions at the unique ID level".36 A law firm's review puts it plainly: structures "are only as reliable as the controls supporting collateral reporting".37 The Financial Stability Board found that the failed companies left lenders "with insufficient or poor information about their clients prior to bankruptcy".14

More tapes change hands each year: private credit, secondary trades and bank lending to funds since 2019

In shortPrivate credit has grown to between $1.5 trillion and $2 trillion worldwide, about $1 trillion of it in the United States. Funds now sell portfolios to each other more often: credit secondary volume reached $20.4 billion in the first half of 2026, more than all of 2025. Banks lend to these funds against their loans, and bank loans to nondepository financial institutions passed $2 trillion in 2026. Every one of those trades and facilities starts from a tape.

Evercore reports that buyers in 2026 placed "greater emphasis on granular borrower-level and asset-level underwriting".38 Credit secondaries were 8.1% of all secondary volume in the first half of 2026, up from 4.0% in 2020.38 A GP-led trade sells loans out of a fund the manager runs into a new one, often with the same manager on both sides, which is a reason for the incoming investors to want the tape checked by someone.

Part of that lending is the back-leverage that funds use to buy loans: a bank lends against a pool of the fund's loans, often as a "borrowing base facility" where the fund "can draw dynamically against a borrowing base determined by eligible loans", or as an asset-based facility with "loan-to-value ratios up to 65%".14 Every eligibility test in those facilities reads tape fields: balance, rank, maturity, whether the loan is pledged elsewhere. A tape error that inflates eligible collateral inflates the advance.

Asset-based finance, lending against pools of loans, leases and receivables, is larger again: KKR's head of private credit called it a "$6 trillion" market that forecasts say could exceed "$9 trillion by 2029".41 That figure is KKR's own characterization.

Representations, the schedule of exceptions, and who pays for a wrong cell after closing

In shortA purchase agreement turns the tape into promises. The seller warrants that the tape is true except as the schedule of exceptions says, and the buyer's remedies for a breach (repurchase, indemnity, price adjustment) depend on how carefully both sides wrote the schedule. Breaks found before signing become exclusions, price changes, holdbacks or disclosures. Breaks found after closing become claims against a seller that may no longer exist.

Loan sale documents differ in how much the seller promises. A public example is a 2015 loan portfolio purchase agreement for 20 term loans, about $67.8 million of commitments, filed with the SEC.2 The seller there represented that it "is the sole legal and beneficial owner of and has good title to each Loan", free and clear of liens other than permitted ones.

The agreement used the industry's standard terms for par trades, under which the seller would use "commercially reasonable efforts" to provide the credit agreement if asked, and the trade was "not subject to receipt by Buyer of Credit Documents prior to the Trade Date". The price was set on "the outstanding amounts of the relevant Loan at settlement". The seller promised title; checking every other cell was left to the buyer.

Securitizations set out the same promises in public. Regulation AB asks the issuer to "Summarize any representations and warranties made concerning the pool assets" and the remedies for breach, and to describe how asset terms can be modified.19

Four remedies a buyer can write in before closing

The schedule of exceptions is where the work becomes enforceable. Each item names the loan, the field, the tape's value, the document's value and the clause. Counsel for both sides can check every item against the page it cites in minutes. A vague schedule ("certain loans may have been amended") protects the seller; a cited one protects the buyer.

Software that reads amendments: where SR 26-2, third-party guidance and fund duties apply, and the decisions people keep

In shortSoftware can read credit agreements and amendments, build each loan's terms as amended, compare them with the tape and draft the schedule. For banks, the 2026 model guidance leaves fixed rules and arithmetic outside its definition of a model and generative AI outside its scope, so the bank's own governance and third-party rules apply. Funds answer to their investors and, if registered, to SEC examiners. Either way the test is the same: every value must carry its page, and the tool's misses must be measured.

SR 26-2, issued April 17, 2026 by the Federal Reserve with OCC Bulletin 2026-13 and replacing SR 11-7, defines a model as "a complex quantitative method, system, or approach that applies statistical, economic, or financial theories", and excludes "simple arithmetic calculations" and "deterministic rule-based processes". A footnote puts generative AI, and AI that acts on its own, outside its scope and leaves them to each bank's own risk management.13,22

Comparing a tape value with a clause, and pricing a break by a formula in the purchase agreement, are rules and arithmetic. Reading a 22-page amendment to find the clause is the part where AI does the work and must be checked.

The 2023 interagency third-party guidance keeps the responsibility with the bank that uses a provider, and a proposed rewrite was open for comment until November 16, 2026.42,15 NIST's AI risk framework gives a structure many firms use: GOVERN, MAP, MEASURE, MANAGE.43 For a private fund, no rule covers the tool itself; the fund's duty to its investors, and the SEC's focus on private credit and valuation, do.23

How you would know a tool works: plant breaks and count

A tool that ties tape cells to documents can be measured the way a reviewer would be. Take a tape you have already closed, whose breaks you know. Change cells you choose: a margin from a superseded amendment, a balance before a prepayment, a lien position, a loan you add to a release schedule. Then count what the tool catches, by kind, and count the false alarms on cells you left alone.

Pricing a break, chasing the seller and signing the findings

• Deciding whether a break is priced, excluded, held back or disclosed, within the purchase agreement.
• Asking the seller for missing amendments and releases, and judging the answers.
• Confirming facts outside the documents: that a release was filed, that a register is current, that a borrower exists and pays.
• Signing the findings, and approving the purchase. The responsibility stays with the firm that signs.42

Inside a diligence team's week before closing: six roles, one deadline and the hours spent reading amendments

In shortA diligence lead runs the review, analysts read a sample of files, outside counsel searches liens and writes the schedule, loan operations checks registers, and the head of underwriting takes it to committee. Most hours go to reading amendments and chasing documents that are missing. The work is squeezed between the seller's last tape and a closing date already set.

Diligence lead, underwriters, loan operations and counsel on one purchase

How a tape moves from the seller's email to the committee

Each new tape version means rereading the amendments

A tape usually changes while it is being checked. The example's buyer is working from version 3, received on the Friday evening before a Wednesday committee; versions 1 and 2 came earlier. Each version means re-reading what changed. Teams handle the squeeze by narrowing the sample, by leaning on counsel's lien search, or by accepting the seller's representations and planning to claim later.

A sampled workbook, agreed-upon procedures or full-tape software

Tape version 3: 48 loans checked between a Friday evening email and a verified price

In shortThis is the example from the top of the page, inside the product. Penwyth Capital Lending Fund I, winding down, sends version 3 of its tape at 6:05 p.m. on Friday, October 9, 2026. By 7:12 p.m. six agents have read all 229 documents, tied 362 of 384 cells to their clauses and drafted the schedule of exceptions. On Monday Elena Marchetti, director of loan diligence at Brackenfold Credit Partners, works through the 22 breaks, signs the findings, and Daniel Okafor, head of underwriting, approves the purchase at the verified price.

1

The screen is the seller's tape, sorted by money at stake

Elena opens the tape at 8:12 a.m. on Monday. Every row is a loan, every column one of the eight terms, and every cell carries a pin. A blue pin means the cell ties to its clause; a red glow means a document contradicts it, with the tape value struck through and the document's value underneath; amber means the deciding amendment was never delivered. She sorts by exposure, so the biggest loans come first.

The panel on the right is the agents' run on Friday evening, step by step, and the price the tape implies against the price the documents support: $547.3 million against $512.9 million.

2

The largest loan matures two years early

The top row is Halvorsen Fire & Life Safety, $18,412,650. Its maturity cell glows red: June 15, 2031 on the tape, June 15, 2029 in the second amendment. Elena clicks it. The clause opens beside the grid: page 3 of the second amendment, signed April 9, 2025, where §2.1(b) restates the Maturity Date, with the new date highlighted.

Under it, the term's history: the 2023 credit agreement set 2031, the first amendment left it alone, the second brought it forward. The field checker's note explains the trade: an earlier maturity in return for a covenant reset.

3

Two loans the seller cannot sell free and clear

Kingsferry Logistics and Wexley Home Health say "Free to transfer". Both title cells glow. In 2025 Penwyth financed its fund with a loan from Quenby Credit Funding II, which filed a UCC-1 financing statement in Delaware covering all loans its financing vehicle owned. Quenby's partial release of January 12, 2026 freed every loan except the 31 on its Schedule B. Kingsferry is line 14; Wexley is line 27.

The lien and pledge checker searched filings against the seller and its vehicle to October 7, read the release and its schedule, and checked the pledgee field on all 48 registers. The two loans are excluded until Quenby releases them.

4

Every break becomes a cited item in the purchase agreement

The schedule writer has drafted Schedule 4.1(k) of the loan portfolio purchase agreement: Part A, three loans excluded; Part B, ten price adjustments; Part C, two holdbacks; Part D, seven breaks disclosed with no price change. Each item names the loan, the field, the tape's value, the document's value and the clause, with a pin back to the page.

Elena reads each item against its page, then sends the schedule to Penwyth's counsel. The agents never send anything to the seller; a person does.

5

A verified price, two signatures, a closing date

The price screen walks from the tape price, $547,309,823, to the verified price, $512,928,540: three loans out, four balances to the agent's register, five repricings, one last-out share at 93.00. Two holdbacks bring the amount paid at closing to $492,938,802.

Elena signs that the tape was checked in full: 384 terms on 48 loans, 362 pinned to their clause and 22 exceptions set out in the schedule. Daniel Okafor approves the purchase. The deal is cleared to close on Friday, October 16.

The solution's design targets are every term on every loan tied to its clause, no double-pledged loan bought, and findings signed the same day the last document arrives. They are targets, to be measured on your own tapes.

What else the product shows: repricings, missing amendments, re-checks, the data room and the fund's rules

In shortOne tape is the story. These screens show the rest: a margin cut by a repricing, a term resting on an amendment nobody delivered, the re-check after the seller sends more documents, one loan's terms through every amendment, the holdbacks and disclosures, the signature, the data room, every deal in diligence and the fund's own rules.

Select any screen to open it larger. The funds, borrowers, agents, people and figures are fictional.

Who does what on a tape: six agents with fixed limits, one signer and one approver

In shortEach agent has one job and limits it does not cross. None of them edits the seller's tape, lets an unsigned draft decide a term, clears a loan with another secured party on record, approves a purchase, moves money or sends anything to the seller. The diligence lead signs the findings, the head of underwriting approves the purchase, and every value on every screen cites a document, a date, a section and a page.

Tape reader, document reader and the four agents after them

The diligence lead signs the findings; underwriting approves the price

Signatures record the name, the decision, what it covers and the time. Agent steps and human decisions are kept apart on the audit trail, every tape version is kept, and nothing reaches the seller until a person sends it.

Brackenfold Credit Partners' diligence and underwriting heads

What the seller sends, what the buyer adds, and what comes back: a pinned tape, a verified price, a cited schedule

In shortThe solution reads what sellers already send and what lien searches already return, applies the fund's own purchase or facility terms, and leaves behind every cell pinned or flagged, breaks sorted by exposure, double-pledge flags, a verified price with each adjustment, and a schedule of exceptions ready for counsel. The fund's systems of record stay where they are.

Data room, agent portals and UCC search feeding the tape check

For each, check whether it has an API, a file feed, or only screens, and who owns it.

Your purchase agreement and pricing model set the rules

Every tape cell carries its page; the latest signed document decides

Where a sampled tape review loses money: reviewer days, mispriced loans and claims after closing

In shortStart from your own deals. Nobody publishes how often corporate loan tapes are wrong or how many hours a review takes, so the figures that matter are yours. A few published numbers frame them: what an analyst hour costs, how often a fully reviewed tape disagrees with its files, and how much collateral can turn out to be pledged twice.

Reviewer days and breaks a sample missed, from your last three deals

The pledged loan you bought, and the seller that wound down

The return on full-tape checks, worked from your deal count, tape size and analyst hours

In shortPut in your own numbers. The calculator starts from a published analyst wage and asks you for the rest: how many tapes you review, the hours they take now and after, the size of a typical tape, how much price a full check finds that a sample misses, and what the tool and your testing cost a year. Every default marked illustrative is an assumption to replace.

Paying for loans as signed, clean title and fewer reviewer days

Reviewer hours and price errors found, tape by tape

A fund that reviews twenty $250 million tapes a year clears a $250,000 yearly cost if a full check finds even 0.005% of par more than its sample, before counting time. A fund that buys two tapes a year of loans it already knows well gains less, and should say so before buying anything.

How ready is your diligence desk for full-tape checks? Eight scored questions and one to reflect on

In shortPick the answer closest to your team today. Eight questions score 0 to 2; the last is for reflection. The meter adds up your score as you go, and each question says why it matters and what a strong answer sounds like.

Diligence desk results: time one tape or replay a closed one

Sixteen tests of a tape review vendor: amendments, title searches, cited evidence and a planted-break trial

In shortAsk every vendor the same questions, us included. The answers show what the tool reads, how it decides which document wins, how it handles title, what evidence you keep, whose rules apply, and how you would measure it on your own tapes.

We found no published independent test of any tape review vendor's accuracy, as of October 2026. A planted-break test on a tape you have closed is the fairest comparison, and the same test measures your current process.

An eight-week pilot: replay a closed tape with planted breaks, then run beside a live deal

In shortMeasure first, then replay a tape whose breaks you already know, with more you plant, then run alongside your team on a live deal and decide with three numbers. None of these steps commits you to a vendor.

1. Time your next tape, by task: reading, searching, chasing, pricing, writing up. That is your baseline.
2. Pick a closed tape with its data room, and list what you found then and what turned up after closing.
3. Plant breaks in a copy: a superseded margin, a pre-prepayment balance, a lien position, a loan added to a release schedule.
4. Run alongside a live deal with your team signing the findings as usual.
5. Decide with three numbers: breaks found by kind, false alarms on clean cells, and days from last document to signed findings.

Breaks found by kind, and tape cells pinned to the right page

Loan tape review and tape cracking: answers to the questions diligence teams search

In shortThe questions buyers, lenders and diligence teams search for most, each answered on its own.