Find out whether your writing is actually readable

Readability scores are blunt instruments that are useful anyway, as long as you know what they measure. Flesch and its relatives count syllables and sentence lengths. They cannot tell whether your writing is any good, whether the argument holds, or whether the reader needed the detail you cut. What they do reliably is find the sentences that got away from you, and those are usually the sentences that lose people.

The number that matters for most business writing is around grade nine to eleven, which is not an insult to the reader. It is a recognition that the reader is busy, is probably on a phone, and is deciding within a paragraph whether to keep going. Specialist material sits higher and should, but it should sit higher because the ideas are hard, not because the sentences are long.

Passive voice is worth treating as a flag rather than an error. Sometimes the passive is exactly right, when the actor is unknown or genuinely beside the point. More often it is doing something less honourable: hiding who did the thing. The difference between the process was delayed and we delayed the process is not a stylistic preference, and in a report to a client it is the whole meaning.

Long sentences are the highest yield thing to fix, because they usually contain two sentences wearing one coat. The test is to read it aloud, and where you take a breath and the sense holds, there was a full stop waiting. Doing that once through a document does more for it than any amount of word substitution.

The failure mode to avoid is writing for the score. Chopping every sentence to twelve words produces prose that scores beautifully and reads like a form. Vary the lengths, keep the long one when it is carrying a complex idea gracefully, and use the highlighting to find the ones that are long by accident rather than by design.

For anything published, this pairs with the less measurable checks: is the first line worth reading, is the promise in the heading kept, and would somebody who knows the subject find it obvious or find it useful. A score cannot answer those, and they are the ones that decide whether the piece worked.