Rewriting a piece for a different reader changes the argument, not just the words

Vocabulary is the easiest thing to change and the least important

Hand over a technical document and ask for a version for a general audience, and what usually comes back is the same document with the jargon replaced and a sentence of explanation added after each term. It is more readable. It is aimed at nobody in particular.

The reason is that changing vocabulary is a local operation and changing audience isn’t. Two readers do not differ mainly in the words they know; they differ in what they already believe, what they are deciding, how much time they will give it and what would make them stop reading.

None of that was in the request. So the piece keeps the order of an internal document, keeps its assumption that the reader has already accepted why the subject matters, and simply says it all more gently.

State the decision the new reader is making

The most useful thing to supply isn’t a description of the audience but the question they are trying to answer. Whether to fund it. Whether to switch. Whether this affects them at all. Whether it is safe. A piece written towards a decision reorganises itself, because the order follows what the reader needs first.

That is why the same material becomes a different document rather than a translated one. A technical report leads with method because its readers judge conclusions by method. A version for someone deciding whether to act leads with the conclusion, its confidence, and what it would cost to be wrong.

Say explicitly what the new reader already knows and what they do not care about. The second half of that is doing more work than people expect: most adaptations are too long because nothing was authorised to be cut.

Some things must not survive the simplification

When a claim is simplified, the qualifications go first. Under laboratory conditions, in this jurisdiction, for adults without existing conditions, in the sample we observed — all of it reads as clutter to a shortening pass, and all of it was the reason the original claim was defensible.

So name the parts that are load-bearing before asking for the rewrite, and check for them afterwards. This isn’t a formality. A piece that has quietly become an overstatement is worse than one that is difficult, because the difficult version at least fails visibly.

The same applies to attribution. Where the original said one researcher suggests, an accessible version tends to arrive at studies show, and that is a change of meaning masquerading as a change of register.

Length and format follow the reader, not the source

Adaptation questions that are frequently left unasked: does this reader want a page or a paragraph, will they read it on a phone in a corridor, are they going to forward it, does it need to work when only the first two sentences are read.

Each answer changes the structure rather than the wording. A document that will be forwarded needs to make sense without you present. One that will be skimmed needs its claims in the headings. One read on a small screen needs shorter paragraphs and no wide tables, which is a layout decision, not a writing one.

Deciding these before drafting saves a round. Deciding them after means editing prose that was organised for a different purpose, and that edit is nearly always longer than the rewrite would have been.

Two audiences, two pieces

The tempting compromise is a single version that works for everyone, with the detail moved to the end. Sometimes that is genuinely right and it is worth saying so. Often it produces a document that opens too vaguely for the specialists and gets abandoned by everyone else before the useful part.

When the decisions being made are genuinely different, two separate pieces are the honest answer, and cheap drafting is precisely what makes that practical. The material is shared; the arguments are not.

What both versions must share is the facts. It is worth checking one against the other specifically for numbers, dates and claims that have drifted apart, because two documents saying different things about the same subject is a worse outcome than either version being slightly wrong on its own.

And where the two versions will be read by people who talk to each other, which is most workplaces, the accessible one should not say anything the detailed one would contradict. That is a stricter requirement than it sounds, since simplification usually works by dropping the conditions under which a claim holds.