Skip to content
FRAMEWORKSTHE BOOK'S SPINE, DEFINED

The Eight Frameworks.

One per chapter of How Bankers Think — the question each answers, the parts it has, and where this site works it for free.

8 FRAMEWORKS · 3 QUESTIONS · 1 TEMPERAMENT · FREE · NO SIGNUP

These are the instruments the book is built from, written here to be quoted whole. Each entry gives the question the framework answers, a definition that stands on its own, the parts in the chapter's order, and — where the chapter gives one — the rule for when the parts disagree. The glossary carries the short form of each; this page carries the working form.

None of it is a curriculum. The eight are not steps, and the book does not ask to be read in order. They are eight ways of asking one question in eight kinds of moment, and the point of naming them is that a named instrument can be picked up deliberately, before the years that would otherwise make it instinct.

// THE ARCHITECTURE

“How do I get my money back, with interest, across cycles?”

The one question underneath every banking decision. Beneath it sit three, and every framework below is one of the three, worn as a habit.

  1. QUESTION 01 · CREDIT
    Will I get paid back?
    Cash flows, assets, business model, resilience across cycles.
  2. QUESTION 02 · PRICING AND STRUCTURE
    Am I being paid enough for the risk?
    Whether the spread compensates for the downside the structure cannot price.
  3. QUESTION 03 · EXIT
    Can I get out if I'm wrong?
    Covenants, secondary markets, alternative lenders, structural exits.

The reading map, in the book's own clustering: Chapters 3 and 4 are the technical core and live inside Question One; Chapters 5 and 6 widen into the industry and the portfolio, where Question Three enters; Chapters 7 and 8 bring all three together and turn them back on the banker. Chapter 2 sits just outside the three — its subject is the interview that selects who gets to ask them.

// THE EIGHT, IN CHAPTER ORDER

01CHAPTER 1Why Bankers Think Differently

The Three Questions

THREE QUESTIONS · ONE TEMPERAMENT

ANSWERS  What is every credit decision actually deciding?

The Three Questions are the whole of commercial banking judgment stated before any technique is reached for: will I get paid back, am I being paid enough for the risk, and can I get out if I am wrong. They are the working form of the one question underneath every banking decision — how do I get my money back, with interest, across cycles. Every other framework in the book is one of the three applied to a particular kind of moment: a working capital request, a set of financial statements, an industry, a deal to decline. Senior bankers do not run them in sequence; after enough years the three fuse into a single read, and that fused read is what the book means by temperament.

THE PARTS
  1. 01Will I get paid back? — Credit. Cash flows, assets, business model, resilience across cycles.
  2. 02Am I being paid enough for the risk? — Pricing and structure. Whether the spread compensates for the downside the structure cannot price.
  3. 03Can I get out if I'm wrong? — Exit. Covenants, secondary markets, alternative lenders, structural exits.

SERVESQ1Q2Q3

WORKED INHow banks approve business loansHow banks work

SHORT FORMin the glossary

02CHAPTER 2The Hidden Curriculum

The Four Signals

FOUR FACES OF ONE DISPOSITION

ANSWERS  What is a senior banker listening for beneath an interview answer?

The Four Signals are what a senior banker listens for underneath the printed interview questions: the early traces of the Three Questions before they have become habit. Credit instinct is a mind that moves first to durability and only then to growth. Structural thinking organises a description by category rather than chronology. Intellectual honesty names the edge of what it knows without bluffing, and coachability revises the view when new information arrives. They are not four separate checks; they are the disposition that, given time, produces the three questions — which is why an interviewer who could not name them still recognises a candidate who has something there.

THE PARTS
  1. 01Credit instinct — Moves first to the downside, not the growth — the early trace of will I get paid back.
  2. 02Structural thinking — Organises by category and framework, not chronology — the early trace of can I get out if I'm wrong.
  3. 03Intellectual honesty — Names precisely what it does and does not know — the first half of am I being paid enough: do I know enough to price this risk?
  4. 04Coachability — Revises the view when new information arrives — the second half: can I revise my pricing as the facts change?

SERVESQ1Q2Q3

WORKED INCommercial banking interview questionsInterview quizCredit analyst interview

SHORT FORMin the glossary

03CHAPTER 3Reading Real Demand vs. Plugging Holes

The Three Diagnostics

THREE TESTS · RUN IN SEQUENCE

ANSWERS  Is this working capital request real demand, or a plugged hole?

The Three Diagnostics are three tests run in sequence on any working capital request, to separate a borrower who needs more capital because they are growing from one who needs it because they are deteriorating. Each test points scepticism at one of the three places the truth tends to live: the sales claim, the cash flow statement, and the shape of the facility being asked for. They cut both ways — when all three say growth, the discipline is to believe them and approve quickly, which is why the files published on this site include an approval alongside the declines.

THE PARTS
  1. 01Is the growth real? — Volume or price · receivables moving in line with sales · the industry confirming the story.
  2. 02Is the cash conversion intact? — Operating cash flow against net income over years · the cash conversion cycle · the underrated trend in payables days.
  3. 03Does the structure match the need? — Revolving against term · a facility that self-corrects with the trade cycle · what the borrower asked for, and why.

RULING  Not a checklist — a way of pointing scepticism at the three places where the truth tends to live.

SERVESQ1Q2Q3

WORKED INWorking capital calculatorWorking capital case studyIssue 01 · Pacific PremiumIssue 04 · HighviewCash conversion cycle, worked

SHORT FORMin the glossary

04CHAPTER 4Reading the Numbers Behind the Numbers

Quality of Earnings Triangulation

THREE ANGLES · RUN IN PARALLEL

ANSWERS  Is the profit this company reports the kind of profit a bank can lend against?

Quality of Earnings Triangulation tests reported profit from three independent angles at once — whether it converts to cash, whether the balance sheet is moving at the speed the income statement implies, and what the notes say and choose not to say — and reads the answer from whether the three agree. The redundancy is the point. A company can dress up one angle; dressing up two is hard; dressing up all three while keeping the auditor satisfied is close to impossible. The name borrows a private-equity due-diligence phrase and repurposes it for a lender's question.

THE PARTS
  1. 01Profit-to-cash conversion — Operating cash flow over net income across at least five years, read for level, trend and volatility. The trend is the signal; the level is the context.
  2. 02Balance-sheet velocity — Whether receivables, inventory and payables are accelerating or slowing — margin pressure hides inside slowing velocity before it reaches the income statement.
  3. 03Disclosure language — What the notes, accounting policies and commentary say, what they have stopped saying, and how the language drifts over time.

RULING  When the three angles disagree, bet on the cash flow and the notes.

SERVESQ1Q2Q3

WORKED INIssue 01 · Pacific PremiumCash flow analysis for business lendingCredit analysis for beginners

SHORT FORMin the glossary

05CHAPTER 5Reading the Industry, Not Just the Borrower

The Lendability Matrix

A 2×2 · A THIRD DIMENSION · A MODIFIER

ANSWERS  What kind of debt can this industry support — and where does that read break?

The Lendability Matrix organises industries on two dimensions — asset intensity and cash flow stability — into four quadrants, each of which supports a different kind of lending. It does not tell a banker whether to lend; it tells them what shape of debt the industry can carry: term lending against assets, asset-based lending through a cash-flow cycle, cash-flow lending against contracted income, or lending with discipline where neither income nor collateral catches you. The matrix is the entry point. Cycle position and concentration are the refinements that stop it misleading, and the situations where it breaks are where senior judgment lives.

THE PARTS
  1. 01High asset · stable cash flow — Term lending heaven — the quadrant commercial banks were built to lend into.
  2. 02High asset · volatile cash flow — Asset-based lending territory — lend against the asset and ride the cash-flow cycle.
  3. 03Low asset · stable cash flow — Cash-flow lending territory — lend against the contracted income, not the balance sheet.
  4. 04Low asset · volatile cash flow — The banker's caution zone — neither income nor collateral catches you. Lend with discipline, or not.
  5. 05Then: cycle position — The third dimension the matrix does not capture — how stable any quadrant placement actually is right now.
  6. 06Then: concentration — The within-quadrant modifier — customer, supplier and geographic concentration adjust lendability inside a quadrant.

SERVESQ1Q2Q3

WORKED INIssue 03 · Kauri DairyIssue 05 · Meridian LogisticsManufacturing, read by a bankCorporate lending and credit risk

SHORT FORMin the glossary

06CHAPTER 6Spotting Trouble Before It Spots You

The Three Modes of Watching

THREE KINDS OF ATTENTION · RUN CONTINUOUSLY

ANSWERS  How does early recognition of trouble become a reliable output rather than occasional luck?

The Three Modes of Watching are three forms of attention held continuously across a portfolio, so that trouble registers while there are still options. Pattern watching holds a baseline for each borrower clearly enough that small departures register before any single one is large. Coupling watching reads the relationships between variables, so that a decoupling shows even while each variable is still inside its normal range. Silence watching attends to what has stopped arriving — the report that is late, the call not returned, the dinner not rescheduled — and treats absence as information. Each mode catches what the other two miss, and the banker who sees trouble at twelve months has choices the banker who sees it at three does not.

THE PARTS
  1. 01Pattern watching — Single variables, across time. Requires memory and recognition — the rhythm, and the deviation from it.
  2. 02Coupling watching — The relationships between variables, at any time. Requires structural thinking — revenue and receivables decoupling, a CFO's tone not matching the operational story.
  3. 03Silence watching — What is not happening at all. Requires deliberate attention — covenant certificates in a changed format, the calls that stopped being made.

RULING  Each mode catches what the others miss. Run together, they produce the attuned attention.

SERVESQ1Q2Q3

WORKED INThe bank annual reviewHow banks manage credit riskLoan covenants, explained

SHORT FORMin the glossary

07CHAPTER 7The Deal You Don't Do

The Walk-Away Framework

FOUR SCENARIOS · FOUR SELF-TESTS

ANSWERS  How does a banker reach a no they can defend, rather than declining on discomfort or proceeding on momentum?

The Walk-Away Framework converts the instinct to decline into a defensible, communicable decision. Its first half is a typology — four scenarios a banker actually meets, each calling for a different conversation: a counter-proposal, a deferral, a clean exit, or a pass. Its second half is four self-tests run internally before committing, in a sequence that gets harder as it goes. It exists because a career in credit is shaped more by what is declined than by what is approved, and because the hardest declines are not the weak files but the ones where a deadline, a long relationship or pipeline pressure is doing the arguing.

THE PARTS
  1. 01Right deal, wrong structure — Counter with the structure the bank can do. The most common scenario, and the one that most often ends in a deal rather than a decline.
  2. 02Right business, wrong moment — Defer, and mean it. The warning signs are outside the borrower — the cycle, the regulator, the bank's own concentration.
  3. 03Right credit, wrong relationship — Decline and end the file cleanly. Behaviour during the courtship is a fair preview of behaviour under stress.
  4. 04Right everything, wrong bank — Pass. The deal is good, but not for this bank's hold limit, sector depth or expertise — and ask why it reached your desk.
  5. 05Self-test 1 · the credit-committee defence — If this goes wrong eighteen months from now, can I defend today's decision in front of the committee?
  6. 06Self-test 2 · courtship as preview — Is this client's behaviour now a fair preview of how they will behave in a stress event?
  7. 07Self-test 3 · the right lender — Am I the right lender for this deal — or just the available one?
  8. 08Self-test 4 · the right banker — Am I, at this point in my career, the right banker for this client?

SERVESQ1Q2Q3

WORKED INWhen the bank says noRelationship manager interviewCommercial banking interview questions

SHORT FORMin the glossary

08CHAPTER 8Reading the Room: The Banker as Translator

The Three Translations

THREE HANDOVERS · EACH WITH A FAILURE MODE

ANSWERS  Where do a sound business and a lending process most reliably fail to understand each other?

The Three Translations name the three joints between a business and the process that decides whether it gets funded, where reading most often breaks down. Cycle versus calendar: the business runs on its own rhythm, the bank reviews it on quarter-ends, and a snapshot at the wrong point in the cycle reads as distress. Visible versus actual: documents capture what a business can produce on paper, not the owner judgment, supplier relationships and key staff that carry it through a hard year. Slice versus flow: the bank prices one transaction while the client is living a multi-year relationship of which this request is the next instalment. The earlier frameworks operate on information; these operate on handovers, and they name the work the others have been instruments of.

THE PARTS
  1. 01Cycle versus calendar — The most common. Untranslated, a business read mid-cycle looks like a business in trouble.
  2. 02Visible versus actual — The deepest. Untranslated, both sides misdiagnose each other — the paper is mistaken for the business.
  3. 03Slice versus flow — The most underrated. Untranslated, the deal is priced and structured blind to the relationship it belongs to.

SERVESQ1Q2Q3

WORKED INWhat a commercial banker doesHow banks read youThe bank annual review

SHORT FORMin the glossary

THE FULL TREATMENT

Each entry above is the working form. The chapter behind it is where the framework is developed — where it came from, where it breaks, what a senior banker does when it does, and the cases that turn it from checklist into instinct.

How Bankers Think — eight chapters →

// QUESTIONS PEOPLE ASK

What are the eight frameworks in How Bankers Think?
One per chapter: the Three Questions (Chapter 1), the Four Signals (2), the Three Diagnostics (3), Quality of Earnings Triangulation (4), the Lendability Matrix (5), the Three Modes of Watching (6), the Walk-Away Framework (7) and the Three Translations (8). They are organised under one question — how do I get my money back, with interest, across cycles — and the three questions beneath it. Chapters 3 and 4 are the technical core, 5 and 6 the contextual expansion, and 7 and 8 the integrative pair that turns the instruments back on the banker.
What is the foundational question in How Bankers Think?
How do I get my money back, with interest, across cycles. Every framework in the book is a structured way of answering it in a particular setting. Beneath it sit the Three Questions — will I get paid back, am I being paid enough for the risk, can I get out if I'm wrong — and the book's argument is that senior judgment is what those three become once they have fused into a single read.
Which framework applies to a working capital request?
The Three Diagnostics, run in sequence: is the growth real, is the cash conversion intact, does the structure match the need. If the doubt is about the financial statements themselves rather than the request, Quality of Earnings Triangulation is the instrument — the two are built to be used together, the first on the ask and the second on the numbers behind it. This site's working capital calculator runs the diagnostics on your own figures, free.
Do I need the book to use these frameworks?
No. The definitions on this page are complete enough to run, and the drills, calculator and guides on this site work them for free. What the book adds is the development — where each framework came from, where it breaks, what a senior banker does when it does, and the cases that make it instinct rather than checklist. This page is the reference; the chapters are the reading.