Manuscript Review for Business Books

Business bestsellers are compact. The median runs to 80,600 words, and the middle half of books spans just 72,000 to 82,000, the narrowest length range of any genre we studied. They read at a Flesch Reading Ease of 55, a tenth-grade level, harder than self-help or spirituality.

The difficulty comes from vocabulary. Against the median bestseller, business books lose about 5.5 Flesch points to longer words and almost nothing to sentence length. That is the genre’s jargon tax, and the exceptions prove it: Rich Dad Poor Dad, built on dialogue and plain words (“An asset puts money in my pocket. A liability takes money out of my pocket”), scores 73.5, while Deep Work scores 47.

They name one idea and make it memorable. Five of the nine had their concepts rated “very easy to follow”, and every one of the five rests on a memorable device. Thiel builds each chapter on “a single, memorable binary”: zero to one against one to n, monopoly against competition. Sinek “builds everything on one repeatable formula”, the Golden Circle. Collins pairs each idea with a picture: the right people on the bus, the hedgehog, the flywheel.

They make a claim in the first line. Business is the genre most likely to open its first chapter on a bold claim (four of nine books, against one in four across the study). “Good is the enemy of great.” “Every moment in business happens only once.” A claim works as an opening when the chapter then earns it.

They speak to the reader as a practitioner. Six of the nine fall in the Coach quadrant of our voice map: they address the reader directly and build from evidence. Getting Things Done uses 45 “you” words per 1,000, among the highest in our study.

They give the reader a next step. All nine books leave clear action plans, from Allen’s next-action question to The Millionaire Next Door’s expected-net-worth calculation.

Business books are tight. They drew far fewer transition notes than the average book (27% of chapters against 41%; only memoir drew fewer) and fewer tightening notes too (38% against 49%). Their weaknesses sit in how they use cases and evidence.

Case parades. Business drew “use fewer, deeper cases” notes in 47% of chapters, about twice the average. Every one of the nine book-level storytelling notes was about how the book uses cases: “Build one continuous narrative arc instead of serial case-study showcases” (The Lean Startup), “Replace anonymous composites with fully developed case stories” (Getting Things Done), “Let the strongest cases carry the argument” (The Millionaire Next Door). One developed case, followed through setbacks to its result, persuades more than six summarised ones.

Only the winners. Survivorship bias, drawing lessons only from companies or people that succeeded, was raised in eight of the nine books. Two reviews asked for the author’s own failures: Zero to One was told to “use PayPal’s own failures and struggles as a recurring narrative thread”, and Deep Work to “show the cost of the method, not only the wins”.

Untested frameworks. All nine book-level analytical notes asked the author to test, support or qualify the central claim: “Test the central dichotomies against counterexamples” (Zero to One), “Address survivorship bias and the circular logic of WHY” (Start With Why), “Separate correlation from causation” (The Millionaire Next Door), “Confront the evidence problem and the method’s limits” (The Lean Startup).

To see every genre’s signature notes side by side, open Exhibit 17 of the full study.

A Manuscript AI report answers eight editorial questions about the whole book and any questions of your own, and for every chapter gives a summary, recommendations, word count, reading time, Flesch reading ease and grade level. In a business manuscript, these are the parts that matter most.

These ranges describe where the business titles we measured sit, not a formula for success. A manuscript outside a range is not wrong, but it should be outside on purpose.

This sample is illustrative. We wrote it for this page, on Barnum’s public-domain lecture, to show the kind of notes the study’s engine wrote on business books; it is not engine output.

Every business title behind this page, with its measures. Click a row to mark it in Exhibit 1, or open one in the full study to follow it through all 17 exhibits.

Every business title in this playbook has a public sample report: the full analysis as a Manuscript AI report gives it, with the editorial questions answered and every chapter summarised, measured and reviewed.