The 19-Chapter Word Master Report: Structure and Content

A finished study has two artifacts, and they answer different questions. The Excel workbook is the data document — the numbers, sheet by sheet, the record a reviewer opens to check a figure. The Word master report is the narrative master document: the structured, chaptered write-up of the benchmark in the form a reviewer, a client or a regulator reads — the engagement framed, the tested party characterised, the method selected and justified, the benchmark documented, the range computed and the conclusion stated. It is a .docx, generated on demand, and its structure is fixed: nineteen numbered chapters plus eight appendices, on every study, in every jurisdiction. The same generation run also renders a branded PDF for the client — from the same numbers, but with a narrower chapter set, so the two documents are deliberately not the same shape (Word and PDF: one record, two shapes).

How it is built: the report engine

The master report is produced by a deterministic report engine in three layers — the generator never calls AI, and no chapter’s content is free-form:

1. The knowledge base — frozen reference data: jurisdiction modules (curated content for 226 jurisdictions — the original six, India, UAE, USA, UK, Singapore and Germany, plus three curated tiers — and an explicitly-labelled OECD-baseline module for the handful without curated coverage), industry modules, the F01–F30 function set, transaction templates, transfer-pricing method definitions, characterization profiles, PLI definitions, and the report structure itself: the 19-chapter registry plus the eight named appendices.
2. The decision engine — deterministic, defensible rules: the tested party’s characterization (derived from the FAR affinity, never from user free-text), the method selection and its justification, the PLI reconciliation, the benchmarking strategy, and the formula-driven transfer-pricing adjustment where the tested party falls outside the range.
3. The generator — a python-docx builder that walks the fixed chapter registry and renders each chapter from the study’s persisted data. Narrative text comes only from the engagement’s narrative slots (the overrides supplied at generation time) or is produced deterministically — there is no third source.

The structure is data, not code: the chapters live in a registry (MASTER_REPORT_STRUCTURE), the generator iterates it, and a new section is an entry in the registry, not an edit to the builder. Every one of the nineteen chapters has a registered builder, so none of them renders as a placeholder — what varies between studies is the data behind a chapter, not whether the chapter exists.

The front matter

Before the chapters, every report carries:

• The cover — “Transfer Pricing Report”, the platform identity line, the tested party’s name, “Benchmarking Study and Arm’s-Length Analysis — FY”, and the engagement block: Prepared for, Prepared by, Date of report and Jurisdiction. Prepared for falls back to the tested party and the date to the day the document was built; Prepared by is the popup’s one required field, so in practice it carries what the author typed (the engine’s own fallback is the platform identity).
• Confidentiality and Caveats — the fixed engagement caveat: exclusive use by the addressee, no reliance by others, no legal or tax advice, conclusions based on the facts as of the report date.
• Table of Contents — a live Word TOC field over the level 1–2 headings. The generator writes the field, not a rendered list: as delivered the page carries the placeholder “Right-click and select “Update Field” to refresh the contents”, and entries and page numbers appear only when the field is updated in Word.
• The running footer — on every page: the generated-by attribution, the product line and a Word PAGE field, so the numbering follows the document when it is edited.

The 19 chapters

#	Chapter	Contents
1	Executive Summary	The study at a glance: tested party, jurisdiction, nature of business, PLI, selected method; the key findings (accepted count, arm’s-length range, comparable median, tested-party margin, conclusion); the recommendation
2	Introduction	The purpose of the study (a narrative slot), the scope and basis of preparation, and the report’s own structure
3	Regulatory Framework	The jurisdiction module: the regulatory framework, domestic legislation, applicable regulations, OECD alignment, the arm’s-length standard, penalties and documentation requirements — then the rule layer: the rules in force at the year end (applicability, documentation and filing obligations, the documentation threshold, safe harbour, penalties, APA), each with its source, effective window and verification status, or an explicit statement that no rule set was consulted
4	Group Overview	The group’s structure and business (the chapter points to Appendix A), with the tested party’s position in the value chain, and a table of the group companies supplied at generation (entity, jurisdiction, relationship, business). When none were supplied the chapter says the detail is not in the study data rather than printing an empty table
5	Company Overview	Corporate information from the study parameters, the business description (a narrative slot), the organisational structure pointer, and the shareholding schedule (shareholder, relationship, holding, jurisdiction) as supplied at generation
6	Industry Overview	The industry module: description, key characteristics, value chain, value drivers, key risks, competitive landscape, regulatory environment, typical business models
7	Controlled Transactions	The transaction description (a narrative slot), the transaction template details in table form, and the transaction’s transfer-pricing significance
8	Functional Analysis	The FAR matrix as a table — category, item, and the allocation between local entity and associated enterprise — with the FAR narrative slot
9	Characterization	The decision engine’s characterization: the justification, the characterization criteria, and the basis of characterization from the matched profile
10	Selection of Transfer Pricing Method	The method justification, the method’s definition and when it is appropriate, the selected PLI with its justification, and the most-appropriate-method determination
11	Economic Analysis	The arm’s-length principle, the tested party’s margin, and the statistical indicators (mean, median, quartiles, CV) from the accepted set
12	Benchmarking Analysis	The search strategy (geographic market, industry classification, size filters, independence and related-party criteria, loss-making and outlier criteria, the range basis, the multi-year treatment, the database and workflow) — then the engagement search details (database used, regions searched, comparable years), each line appearing only when it was supplied at generation — and the final comparable set as a table of the first 25 accepted companies, any remainder counted in a line pointing to Appendix G
13	Comparability Analysis	The comparability criteria actually applied: the jurisdiction’s prescribed points of comparison where the rule set carries them, each cited, otherwise the OECD five-factor baseline (product characteristics, functional analysis, contractual terms, economic circumstances, business strategies), plus any scoped local constraints and qualifications. The closing “Adjustments for Comparability” subheading states that quantitative and qualitative adjustments were considered where differences were material and that the net-margin PLI is robust to residual differences — the engine proposes no adjustment figures
14	Financial Analysis	The tested party’s key financials as a two-column table: operating revenue, operating cost, operating profit, and the tested-party margin (the statements themselves sit in Appendix C)
15	Arm’s-Length Range	The range computation for the band actually applied — a percentile band prescribed by the jurisdiction’s rule where one governs, otherwise the interquartile range (25th to 75th percentile), or an arithmetic-mean point estimate where the rule prescribes one — with the distribution statistics (count, minimum, quartiles, median, maximum), the bounds, the stated basis and its cited authority, and the tested party’s position with the conclusion
16	Transfer Pricing Adjustment	Where the tested party falls outside the range: the decision engine’s computed adjustment, stated through its formula narrative, followed by the fixed paragraph on a primary adjustment and the need to follow domestic procedural requirements. Where it does not: the engine’s reason that no adjustment is required
17	Conclusion	The arm’s-length conclusion in full, with the jurisdiction-specific conclusion from the jurisdiction module
18	Limitations	The fixed limitations: data as provided, public financials of comparables, professional judgement per the OECD Guidelines, engagement-specific facts, and subsequent change in law or practice
19	Appendices	The eight named appendices — a heading and a deterministic pointer for each, below

The appendices (A–H)

All eight are named in the structure registry, and each renders as a subheading followed by one deterministic sentence: a pointer to where the material lives, not the material itself. Opening Appendix E does not give you a search log — the schedules that carry data (the comparable set, the FAR matrix) are tabled in their chapters, and the appendices say so.

Appendix	What the chapter prints
A	Group Structure — the group is described in Chapter 4; the organisational chart is stated to sit in the engagement file
B	Intercompany Agreement — the agreement governing the Chapter 7 transactions is referenced to the engagement file
C	Financial Statements — the tested party’s statements for the year under review support Chapter 14
D	Detailed FAR Matrix — set out in Chapter 8
E	Comparable Company Search — performed as described in Chapter 12, naming the database from the benchmarking strategy
F	Rejected Companies — the count screened out from the persisted set, documented in the working files
G	Financial Data of Comparables — the count of accepted comparables, set out in the benchmarking dataset
H	Arm’s-Length Calculation — the calculation behind Chapters 15 and 16, documented in the working files

Narrative slots vs. deterministic content

The split is what makes the report defensible and regeneration possible:

• Deterministic, from the persisted study data: every chapter’s numbers — the statistics in Chapter 11, the range in Chapter 15, the comparable set in Chapter 12, the adjustment in Chapter 16 — plus the regulatory, industry, transaction and method content matched from the knowledge base and the decision engine.
• Narrative slots, supplied at generation: exactly four — the purpose of the study (Chapter 2), the company overview (Chapter 5), the transaction description (Chapter 7, where it leads the chapter and the matched transaction template still follows) and the FAR narrative (Chapter 8). Where a slot is left empty, the generator falls back to a deterministic sentence — the chapter is never blank.
• A fifth slot nobody types: the context also carries adjustment_formula, and Chapter 16 reads it — but the popup does not collect it. The decision engine fills it at build time from the selected PLI, the computed adjustment and the margin type, so the adjustment narrative cannot be steered away from the arithmetic.

One record, two renderings: the same statistics the results dashboard shows are the ones in Chapters 11 and 15, and the same dispositions the grid holds are the ones behind Chapter 12’s table and Appendix F’s rejected count. Before the document is built, the generation applies the accept/reject overrides recorded in review to a build-only copy of the results — which is what keeps this report and the Excel workbook telling the same story — and the persisted analysis data is left as it was, so the review workspace still shows the live lists. When two artifacts disagree, the study’s persisted data is the authority and the older artifact is the one to regenerate.

Word and PDF: one record, two shapes

The same on-demand call also produces a branded PDF of the study, built from the same report context — and it is not the same document in another format. Expect six chapters, not nineteen:

	Word master report	Branded PDF
Structure	19 numbered chapters + Appendices A–H	6 chapters: Introduction; Corporate Background; Industry Overview; Functions, Assets and Risk Analysis (“FAR Analysis”); Economic Analysis; Annexures
Schedules at the end	Appendices A–H, each a named pointer	Annexures A–G: Transfer Pricing Regulations; Documents and Information Relied Upon; Accept–Reject Matrix; PLI Computation of the Tested Party; PLI Computation of Comparable Companies; Business Description of Comparable Companies; Index of Principal Documents
Front matter	Cover, confidentiality page, TOC field to refresh in Word	Cover, running header, contents with real page numbers (the document is built twice so the numbers and the page count are final), abbreviations
Numbers and range methodology	Computed in the report engine	Imported from the Word builder, not re-derived — the two cannot disagree on a figure or on the range
Purpose	The editable narrative master document	The locked, client-facing rendering

So: a reader who needs the regulatory framework, the characterization reasoning, the method-selection argument or the limitations as numbered chapters needs the Word file. The PDF is the client-facing cut of the same analysis. Its identity is always Quartyl’s — the brand tokens are frozen, and white-label branding does not reach either generator.

Generating the document

Both files are generated on demand, never inside the analysis pipeline — the partner or firm admin triggers them from the study with the engagement details, and the engine rebuilds the report from the persisted record. The Word leg is the primary deliverable: it is built and stored first, and the PDF is a best-effort second leg, so a PDF failure leaves the Word report intact with no PDF on the study. The full flow — the popup, the job, and what the run does to the study’s state — is in Generating the Final Document.

FAQ

My jurisdiction has no curated module in the knowledge base — what renders? Every chapter built from the study’s own data renders in full. The regulatory-framework chapter then speaks from the OECD Transfer Pricing Guidelines baseline and opens with a printed notice that the statements are not verified against that jurisdiction’s domestic legislation — local prescribed ranges, safe harbours, thresholds and penalties must be confirmed before the report is relied on. It never substitutes another country’s legislation. The industry chapter states plainly that no industry content is available for the engagement, and the transaction chapter does the same when no template matches; the rest of the report is unaffected.

The contents page is empty when I open the file — a bug? Expected. The generator writes a live TOC field, and Word fills it on update (right-click → Update Field, or F9 — the same applies to the page numbers in the footer). The PDF by contrast ships with the contents already typeset and numbered.

Can I hand-edit the generated .docx? You can open it, but the document of record is the generated one — regenerate with corrected engagement details or narrative overrides rather than editing the artifact, so the record and the file stay the same thing.

Does the PDF carry all nineteen chapters? No — six, plus Annexures A–G. See Word and PDF: one record, two shapes.

How does this relate to the Local File? The master report is the benchmark’s narrative write-up; the documentation tiers it feeds into — Local File, Master File, CbCR — are covered in the TP documentation guide.