Break up long sentences
Writing Guide >
The principle
Plain English is usually shorter than the writing it replaces, but not always. What makes a sentence hard to read is not the number of words in it but the number of ideas it carries before it lets the reader stop. Break one overloaded sentence into two or three, each holding a single idea, and it becomes easier to read even when the word count goes up.
Most finance writing does the opposite. It chains clauses together until a single sentence runs past 30 words. The writer believes the density signals rigour. The reader thinks about giving up.
Why it fails in practice
The unbroken sentence flatters the writer. It looks careful, comprehensive, precise; it borrows the cadence of a legal definition, and finance writers reach for that cadence because it sounds authoritative. But the authority sits on the page, not in the reader's head. A sentence that lists six things, names two kinds of actor and states a purpose, all before its first full stop, asks the reader to hold every part of it at once. Most cannot, so they go back to the start and try again. The writer has mistaken their own sense of completeness for the reader's understanding.
There is a second, quieter reason it persists. When finance writers are told to write more plainly, they reach for the delete key. They cut an adjective, trim a phrase, shave the word count — and wonder why the result still reads like treacle. The fault was never the words. It was that one sentence was doing the work of three. Cutting words from a badly built sentence gives you a shorter badly built sentence. The fix is structural, and structure does not always mean fewer words.
This is the part worth sitting with. Plain English is not the same as short English. The measure that matters is not words on the page but words per sentence — how much the reader must carry between one full stop and the next. A readability formula makes the point without argument: the bloated definition below lands several grades beyond a postgraduate reader, while the longer rewrite lands at school-leaving level. The extra words made it easier, not harder.
Before and after
Here is the definition of fraud from the Global Internal Audit Standards:
Fraud is any intentional act characterised by deceit, concealment, dishonesty, misappropriation of assets or information, forgery or violation of trust perpetrated by individuals or organisations to secure unjust or illegal personal or business advantage.
It is one sentence with 34 words and it scores a Flesch-Kincaid grade of 25, beyond any general reader.
Here is a rewrite of it:
Fraud is dishonest behaviour intended to gain an unfair or illegal advantage. It can involve deception, concealment, theft of assets or information, forgery, or abuse of trust. Individuals or organisations commit it for personal or business gain.
The new version has 37 words (an increase of three) but separated into three sentences. It scores a reading grade of 12. Three sentences in place of one reduces the readability by 13 grade levels.
The rewrite works because each sentence carries one idea: the first says what fraud is, the second what it involves, the third who does it and why. The reader takes them one at a time.
Making the actors act helps too, as 'Say who did it' explains: the original's fraud is 'perpetrated by individuals or organisations'; in the rewrite, individuals and organisations 'commit it'.
You can go further, where the setting allows it:
Fraud is theft through lying.
Five words, about grade 1: a child could read it. It loses the detail a formal definition needs, so it will not serve everywhere, but it shows what is possible.
What to do
Find the long sentences in your draft and count its words. Past about 20, they are probably doing the work of two or three, and the reader is paying for it.
Then look for the joints. In the original above they are visible: 'characterised by', 'perpetrated by' and 'to secure' each mark the point where a new sentence wants to begin. Cut at the joins to give each idea its own full stop.
Do not judge the result by whether it got shorter. Judge it by whether a reader gets it the first time.