How to Read 1,000 Pages of Filings in a Week: A 4-Stage Triage System

How to Read 1,000 Pages of Filings in a Week

A 1,000-page regulatory filing is not a document to read — it is a corpus to triage. The right output is a one-page brief backed by a structured record of what you read, what you skipped, and why. The four-stage system below is the one that takes you from a pile of PDFs to that brief without reading every page.

The point is to know what you did not read

The point of a triage system is not to read everything. The point is to read enough that the brief is defensible, and to be honest about the parts you did not read. A brief that names its gaps is more useful than a brief that pretends to have read every page.

Tip

Stage 1: Skim (1-2 hours)

The first stage is a fast read of the document's skeleton. The goal is to build a map, not to extract findings.

The moves:

• Read the table of contents. Build your own version in your notes. The official TOC is a hint; the real structure is what the author thinks the document is for.
• Read the executive summary in full. This is the only section most readers will see. It is also the author's framing, which is data.
• Skim the headings of every section. Mark the sections that look relevant to your question, the sections that look boilerplate, and the sections that look surprising. The surprising ones are where the brief's findings usually live.
• Note the page count of each section. You will use this in stage 3 to plan your sample.

At the end of stage 1, you have a one-page map of the document in your own words. You know what the author thinks the document is. You know where you will look in stage 3.

Stage 2: Structure (2-3 hours)

The second stage turns the map into a structured record. The goal is to commit, on the page, to what the document contains.

The moves:

• Build a table of contents in your notes. Section, page range, one-line description in your own words. The description is doing the work; it forces you to commit to what each section is for.
• Identify the table that the brief will use. Most long filings have a small number of tables that carry the key numbers — risk factors, financial summary, related-party transactions, segment breakdown. Note the page number of each.
• Identify the prose passages the brief will quote. Most long filings have a small number of sentences that capture the document's argument. Mark them now; you will return to them in stage 4.
• Identify the boilerplate. Risk factors, forward-looking statements, accounting policies, definitions. These are often 30-40% of the document. Knowing they are boilerplate lets you skip them in stage 3 without missing findings.

At the end of stage 2, you have a one-page schema for the document. You know which tables and which sentences the brief will use, and which sections are boilerplate.

Stage 3: Sample (1-2 days)

The third stage is where you actually read. The goal is to read the document well enough to write the brief, not to read it cover to cover.

The moves:

• Read the executive summary in full. Yes, again. The first read was for the structure; this read is for the substance.
• Read the sections you marked in stage 1 in full. The relevant sections, the surprising sections, and the sections specifically named in your question.
• Sample the rest. Two honest sampling moves:
  • Random sample with a seed. Number the pages, generate random numbers with a fixed seed, read the first 100 pages your generator returns. The seed matters — it lets a critic reproduce your sample.
  • Maximum-variation sample. Pick pages that disagree with each other, then add pages that fill the gap between them. Good for exploratory work; weak for final counts.
• Read the tables in full. The tables are denser than the prose and contain the structured facts the brief will use.

A useful rule: a structured sample of 10-15% of the document, plus a full read of the marked sections, will catch the 80% of findings that drive the brief. The 20% you miss is in the methods section of the brief.

Sample honestly

What you should not do is sample by "the most interesting ones" or "the ones at the top of the document." Both quietly bake selection bias into your conclusions. Name the sampling method in the brief. A reader who knows how you chose the pages can argue with your choice; a reader who does not know cannot.

Warning

Stage 4: Write (1 day)

The fourth stage is the brief. The goal is a one-page artifact a stranger can audit.

The moves:

• Write the precise claim first. One sentence. Not "the company is doing well," but "the company reported $X in revenue for FY2024, a Y% increase over FY2023."
• Write the supports and contradicts lists. Each item is a record — section, page, exact wording. The exact wording matters; you will re-check it against the source.
• Write the weights paragraph. A short paragraph naming the source's weight in your argument. A regulator's filing is a primary, audited, public-interest source. The company's investor letter is a primary, unaudited, in-scope source. Two of the first outrank a dozen of the second.
• Write the corpus biases paragraph. A short paragraph naming what you did and did not read. "We sampled 12% of the 1,000 pages, randomly with seed 42, plus a full read of the executive summary, risk factors, and section 7 (related-party transactions). The 88% we did not read may contain material findings we did not surface."
• Write the "what would change my mind" paragraph. The specific kind of evidence that, if found, would change the conclusion. The act of writing it forces you to imagine a world in which your conclusion is wrong.

At the end of stage 4, you have a one-page brief. The brief is auditable. A future reader can re-run the same four stages from the same document and arrive at the same conclusion, or argue with the conclusion on the basis of the same evidence.

A few habits that scale

• Read at the same time every day. Triage is a routine. The 4-stage system works best as a daily practice, not a one-time push.
• Keep a running log of surprises. When you find a section that does not match the structure you built in stage 2, write it down. The surprises are usually where the findings live.
• Quote, then paraphrase, then summarize. The order matters. Quoting is a direct lift; paraphrasing is your words against the source; summarizing is your model of the source. The brief should mostly quote, with paraphrasing for context, and summarizing only when the underlying source is too long.
• Do not skip the write stage until the end. The writing is where most of the real thinking gets done. If you are avoiding it, the project is avoiding you.

The move that catches the most misses

A single read of a long filing is rarely enough. A first read for structure, a second read for substance, and a third read in the write stage is the minimum. The third read is where you catch the sentences you misread in the first two, and the claims that did not survive the writing.

The most expensive research mistakes happen when the write stage is rushed. A claim that looked solid in stage 3 may not survive the precision of stage 4. The write stage is the place to find that out.