Writing Without Bullshit

Bernoff's definition is precise: bullshit in business writing is communication that is deliberately vague, evasive or inflated, and it exists because clear statements can be held against their author.

His framing is more useful than generic complaints about jargon because it identifies a motive. Passive constructions, abstraction and hedging are not carelessness — they are working as intended, protecting the writer from accountability, and that is why exhortations to write clearly have so little effect.

Vague business prose is not careless. It is working as intended, shielding the writer from being held to anything.

Devmark’s reading — not a quotation from the book

His opening calculation: a document sent to two hundred people multiplies every wasted minute by two hundred, so the writer's convenience is bought at a large collective cost.

His argument that the writer should spend time so the reader does not is the ethical core of the book, and it reframes revision from perfectionism into consideration for people who did not choose to read this.

His central rule: treat the reader's time as more valuable than your own. Its main consequence is front-loading — put the conclusion, the request or the news first, and the reasoning after it.

His argument against the build-up structure taught in school is that a business reader decides whether to continue within a few lines, so withholding the point until the end guarantees it will not be read by the people who most needed it.

His catalogue of corrections: eliminate passive voice where it hides an actor, cut weasel words, delete jargon that has a plain equivalent, replace vague quantifiers with numbers, and remove qualifiers that signal uncertainty the writer does not actually have.

His treatment of numbers is the sharpest — that a claim of significant improvement is worthless and a claim of eleven per cent is not, and that writers reach for the first precisely because the second can be checked.

His planning method, applied before writing: identify the Readers, the Objective, the Action you want, and the iMpression you want to leave.

Its value is that it forces the writer to name a desired action, and a great deal of corporate writing fails the test immediately because no action was ever intended. Documents written with no purpose beyond being sent are the largest single category of waste he identifies.

Chapters on email, reports, blog posts, social posts, presentations and press releases, each with the specific failure modes of the form.

The email chapter is the most immediately valuable — subject lines that state the request, the action in the first two lines, and the recognition that a long email is a meeting nobody agreed to attend.

His sections on the organisational reality: editing other people's work, surviving review cycles, dealing with legal and compliance input, and writing in groups.

His honesty that clear writing is frequently degraded by review — because each reviewer adds protection for their own position — is the part that most rings true to anyone who has watched a good draft pass through four approvals.

Each reviewer adds protection for their own position, which is how a clear draft becomes an unreadable one.

Devmark’s reading — not a quotation from the book

It is a book of tactics with a repetitive structure, and its examples date quickly because they are drawn from contemporary corporate and marketing material.

It also assumes the writer has the standing to write directly, which many people in organisations do not. Its advice is easier to follow the more senior you are, and the book acknowledges this only briefly.

Where Zinsser and Strunk address prose, this addresses documents — things with a purpose, an audience who did not choose them, and an organisational context that shapes what can be said.

That makes it the more directly applicable of the three for most working readers, and the least interesting to read.

Front-loading, which is the highest-return change available in business writing and requires no skill, only the decision to state the point before justifying it.

And the ROAM check, specifically the action question, because a document that cannot name the action it wants usually should not have been written at all.