Copyeditors catch internal contradictions. They will notice if page 40 says 1974 and page 210 says 1976. What they will not do is check either date against reality — that is a separate discipline, and assuming otherwise is one of the more expensive misunderstandings in nonfiction publishing.
If your book makes verifiable claims, someone should verify them. Here is a method that works.
Build a claims list
Go through the manuscript and extract every checkable assertion into a spreadsheet: statistics, dates, quotations, attributions, technical descriptions, anything stated as fact.
This is tedious and it is the whole job. A full-length nonfiction book typically yields between 200 and 600 claims. Seeing them in a list rather than embedded in prose changes how you look at them.
Grade by consequence
Not every claim carries the same risk. Sort into three tiers.
High: anything a critic would attack, anything about a named person or organisation, anything a reader might act on, anything central to your argument.
Medium: supporting statistics, historical detail, technical description.
Low: background colour that would embarrass you but not undermine you.
Check high-tier claims to primary sources without exception. Medium-tier to a reliable secondary source. Low-tier as time permits.
Go to the primary source
The most common failure in nonfiction is citing a source that cites a source. A statistic gets quoted in an article, quoted from that article in a book, and quoted from the book in yours — and somewhere in the chain a qualifier was dropped, a date shifted, or a population changed.
For every high-tier claim, find the original study, the original document, the original transcript. This will occasionally destroy a favourite paragraph. Better now than in a review.
Quotations are where books fail
Check every quotation against the original, word by word, including punctuation. Then check the context — is the speaker saying what your framing implies they were saying?
Misquotation is the most damaging error in nonfiction because it looks like dishonesty rather than carelessness, and readers do not distinguish.
Record as you go
For every checked claim, note what you checked it against and when. If you are challenged after publication, this record is the difference between a considered response and a scramble.
It also makes updating a second edition dramatically less painful.
Consider a second pair of eyes
You will read what you meant to write. For books where accuracy is the whole value — health, law, finance, history — a professional check on the high-tier list is worth its cost. It is a fraction of what a public correction costs.
Priya Raghunathan
Guest contributor
Part of the editorial team at Winfrey Publishing. Articles here are written by the people who do the work, not by a content agency.
Want this applied to your own manuscript?
A manuscript assessment gives you a written report from an editor in your genre — including when the honest answer is that you should keep revising.
