Auditors expect a memo that states the issue in terms of the specific transaction, lays out the facts with a source for each, quotes the governing ASC paragraphs, and applies that guidance to the facts element by element: the application section, not the guidance section, is the part reviewed hardest. Judgmental positions need the credible alternative treatment stated fairly and rejected with reasons, a conclusion that answers the issue as framed, and exhibits that foot and tie back to the analysis. A memo that pastes guidance without applying it, or whose citations do not support the sentences they follow, gets kicked back.
A technical accounting memo has one job: let a skeptical reader reach your conclusion using only what is inside the four corners of the document and its exhibits. The engagement team reads it that way, and on hard issues so does their national office. The reviewer is not grading prose. They are testing whether the analysis stands on its own: whether each fact is sourced, whether the guidance cited actually governs the question, and whether the conclusion follows from applying one to the other. A memo that merely asserts a conclusion, or merely quotes the standard, gives the auditor nothing to evaluate, so it turns into a request list.
For public companies the memo does a second job: it is control evidence. It shows that a competent review of a nonroutine transaction happened at the time, which is why undated, unversioned files draw comments even when the accounting is right.
| Memo section | What the reviewer checks |
|---|---|
| Issue statement | An answerable question about this transaction, not a topic label |
| Fact pattern | Every fact the conclusion turns on has a source |
| Authoritative guidance | The operative ASC paragraphs, quoted, not summarized |
| Interpretive guidance | Used where the Codification is ambiguous, and labeled as interpretive |
| Application | Guidance mapped to facts, element by element, with numbers |
| Alternatives | The credible other answer stated fairly and rejected with reasons |
| Conclusion | Answers the issue as framed and states the resulting accounting |
| Exhibits | Executed documents, schedules that foot, sensitivity where estimates matter |
"Accounting for the Company's debt" is a topic, not an issue. An issue statement the auditor can work with reads: "Whether the June 30, 2026 amendment of the Company's 20.0 million dollar term loan is a modification or an extinguishment under ASC 470-50, and how the lender and third-party fees are accounted for under each outcome." It names the transaction, the date, the governing subtopic, and the decision to be made. Everything after it either serves that question or does not belong in the memo.
Each fact the conclusion turns on should carry a pinpoint source: an agreement section, a schedule that ties to the ledger, dated board minutes. "Per discussion with the CFO" is a placeholder, not a source, for a contractual term. Separate facts from assumptions, and support the assumptions.
| Fact | Source |
|---|---|
| 20.0M term loan, 7.5% fixed, maturing June 2028 | Credit agreement dated June 15, 2023, sections 2.01 and 2.06 |
| Amendment signed June 30, 2026: rate to 9.25%, maturity to June 2030 | First amendment, sections 1(a) and 1(b) |
| 300,000 amendment fee paid to the lender at signing | First amendment, section 4; wire confirmation dated June 30, 2026 |
| Unamortized issuance costs of 150,000 at June 30, 2026 | Amortization schedule at Exhibit C, tied to the general ledger |
| 85,000 of legal fees paid to outside counsel | Invoices 4417 and 4432 |
When the auditor later pulls the amendment, every cited section should say what the memo says it says. Facts that turn out to be from a draft rather than the executed document undermine the whole file.
Quote the operative paragraphs verbatim, with paragraph-level citations: ASC 470-50-40-10, not "ASC 470." Paraphrase drifts, and the reviewer will check your words against the Codification's. The guidance section is not a literature review; include the scope paragraphs when scope is genuinely at issue and the recognition or measurement paragraphs that drive the analysis, and stop there.
The hierarchy matters because ASC 105-10-05-1 makes the Codification the source of authoritative US GAAP for nongovernmental entities (along with SEC rules for registrants). Everything else, including the Big 4 manuals, is nonauthoritative.
Firm interpretive guidance earns its place where the Codification is silent or ambiguous, but it must be identified as interpretive and never blended into the authoritative quotes. Where the major firms read a judgment area differently, say so and explain which reading you adopted and why. The auditor's national office knows where the firms diverge; a memo that presents one firm's view as settled GAAP reads as either unaware or selective.
This is where memos are won or lost. Application means mapping each element of the quoted guidance to a specific, sourced fact, with numbers where the test is quantitative. If the guidance sets out three criteria, the application addresses three criteria against named facts. A useful self-test: could a reader reconstruct the rule from your application section alone? If the application is shorter than the guidance section, or consists of restating the guidance with "the Company believes" in front of it, it is not application.
On judgmental questions, the memo should state the credible alternative fairly, at its strongest, and explain why it was rejected on these facts. Silence on the alternative reads one of two ways to a reviewer, and neither is good: the preparer did not see it, or saw it and avoided it. This is also where a balanced memo earns credibility for its conclusion; a memo that only marshals support reads as advocacy.
The conclusion should mirror the issue statement. If the issue asked two questions (modification or extinguishment, and fee treatment), the conclusion answers both, and then states the accounting that follows: the entry, the go-forward effective rate, and the disclosure consequence. A conclusion that answers a subtly different question than the one framed is scope drift, and reviewers notice.
Attach the executed agreement, not the draft. Every schedule should foot, and its totals should be the numbers the memo uses. Where the conclusion rests on an estimate, include sensitivity: if a 50-basis-point change in the discount rate flips the answer, the auditor will run that math whether or not you did, and the memo is stronger for showing it first. Finally, the file itself carries a date, a version, the preparer, the reviewer, and a note of the guidance in effect as of that date, including any issued-but-not-yet-adopted ASU that touches the topic.
Continuing the term loan fact pattern above, ASC 470-50-40-6 through 40-8 require an exchange or modification with substantially different terms to be accounted for as an extinguishment, and ASC 470-50-40-10 treats terms as substantially different when the present value of the cash flows under the new terms is at least 10 percent different from the present value of the remaining cash flows under the original terms. The calculation mechanics live in ASC 470-50-40-12: the new cash flows include amounts paid to the creditor as fees, net of any amounts received from the creditor, the discount rate is the effective interest rate of the original debt, and prepayable instruments are tested under both exercise and nonexercise scenarios.
| 10 percent test (Exhibit B) | Amount |
|---|---|
| PV of remaining cash flows, original terms, at the original effective rate | 19,850,000 |
| PV of cash flows under amended terms, including the 300,000 lender fee | 21,930,000 |
| Difference | 2,080,000 |
| Difference as a percentage of the original PV | 10.5% |
| Result | At least 10 percent: extinguishment |
At 10.5 percent the result is close to the line, so the exhibit should show the exercise and nonexercise scenarios ASC 470-50-40-12(c) requires when either instrument is prepayable, and note that the scenario generating the smaller change governs the test, so a prepayment scenario below 10 percent would change the answer to modification accounting, not merely weaken it. A memo that lands at 10.5 percent with no sensitivity invites the recalculation that finds 9.8 percent.
The entry, with the loss built from the write-off of unamortized costs and the lender fee:
June 30, 2026: extinguishment of the original term loan
Dr. Term loan payable (original) 20,000,000
Dr. Loss on extinguishment of debt 450,000
Cr. Term loan payable (amended) 20,000,000
Cr. Unamortized debt issuance costs 150,000
Cr. Cash (lender amendment fee) 300,000
Loss = 150,000 write-off of unamortized issuance costs
+ 300,000 lender fee, assuming the amended note's
fair value equals its principal.
The 85,000 of third-party legal fees is capitalized as issuance costs of the new instrument and amortized over its term. Had the test come in under 10 percent, the treatment inverts under ASC 470-50-40-17 through 40-18: the lender fee would roll into the yield of the modified debt through a new effective rate, and the third-party costs would be expensed as incurred. Showing that fork, briefly, is the alternatives section doing its work.
| What the reviewer sees | What it signals |
|---|---|
| Long guidance section, thin application | The memo asserts rather than analyzes |
| A citation that does not support the sentence it follows | Every other citation now has to be checked |
| No alternative view on a judgment call | Advocacy, or a blind spot |
| Schedule totals that differ from the memo's numbers | The factual base is unstable |
| A draft agreement in the exhibits | The analysis may not match the executed deal |
| No date, version, preparer, or reviewer | No evidence the review happened at the time |
The citation defect is the most corrosive. One reference that, when pulled, does not say what the memo claims converts the reviewer from reader to forensic checker, and the whole file slows down.
AI is genuinely useful in two places: retrieval, where it surfaces the operative paragraphs and the interpretive discussion faster than keyword search, and first drafts, where a structured starting point beats a blank page. The known failure mode is the confident citation: a paragraph number that exists but does not say what the sentence claims, which is exactly the defect that gets memos kicked back. The discipline is mechanical. Verify every citation against the source text before it enters the memo, keep the judgment sections (alternatives, conclusion) yours, and hold the AI draft to the same kickback list above. An unverified AI draft is not a memo; it is a lead sheet.
Where GAAP IQ fits: MemoIQ drafts a memo issue by issue, separating the quoted guidance, the applied cited analysis, and the conclusion, and pauses on judgment calls rather than deciding them for you. Before anything is shown, a verifier model checks every citation against its source text and strips what it cannot support, and Word exports carry numbered references with quoted sources so the memo can be checked outside the app. The free plan includes one memo to try, no card required.
As long as the judgments require, and no longer. Length should scale with the number of genuine judgment calls, not with the size of the transaction: a bright-line quantitative test may support a conclusion in a few pages, while a multiple-element arrangement can justify twenty. The reliable signal is proportion, not page count: if the quoted guidance outweighs the application of that guidance to the facts, the memo is too long in the wrong place.
Quote the operative paragraphs verbatim with paragraph-level citations, and use your own words for transitions and analysis. Paraphrase drifts, and the reviewer will compare your wording against the Codification's. Under ASC 105-10-05-1 the Codification is the source of authoritative US GAAP, so the authoritative text itself, not a summary of it, is what the conclusion must rest on.
Yes, and on genuinely ambiguous questions you generally should, but label it as interpretive and never present it as authoritative GAAP. It supplements the Codification; it cannot substitute for it. Where the major firms read a judgment area differently, the stronger memo acknowledges the divergence and explains which reading the company adopted and why on these facts.
At the time of the transaction, before the close it affects. Contemporaneous documentation carries more evidential weight than a memo reconstructed during fieldwork, and for public companies it is the evidence that the review control operated. Update it through dated versions that preserve the original, never silent edits, and evaluate facts arising after period end under ASC 855 rather than folding them back into the original analysis.
ResearchIQ answers it from the Big 4 handbooks plus FASB and SEC sources, with every claim cited to the page — then drafts the memo.
Try it free — no card requiredThis guide is an educational research starting point, not professional advice. Conclusions depend on specific facts and circumstances — consult your advisers, and verify every citation against the authoritative text.