Audience redesign

Rewrite the same explanation for three different real audiences you actually know: grounding register in people, not abstract labels.

When to use this: While you practise
Learning impact: Reliable gain
Learning evidence: Credible with caveats
Where you are with the topic: Applying in new contexts
How often to use it: Worth repeating often

When to use it

You can explain an idea once, in one voice, and need to retarget it for people you actually know.
Abstract “make it formal” is not constraining your word choice enough.
You are about to message a sibling, a supervisor, and a teammate who hates jargon, with the same underlying point.

What to ask AI

Type something like: “Idea: how caching speeds up a site. Audiences: (1) my sister who hates jargon, (2) my CS professor, (3) my product-manager teammate. I will write three versions. Critique each for audience fit and hidden assumptions. Do not rewrite for me.”

Name three real people and what they know or care about. Fake “general public” audiences do not constrain you.

Keep the hard part on your side: you write all three; the AI only critiques fit.

How it helps

Abstract formal/informal labels are weaker than “Sam will zone out if I lead with equations.”

Audience redesign grounds register in people you know, so word choice has real stakes.

Rewriting the same explanation three times also reveals what you were assuming, the knowledge you never stated.

This is production for communication, not a comprehension exercise that merely compares styles.

How to run it

Name three real people and what each knows or cares about.
Write three versions of the same explanation yourself.
Ask for critique of fit to each person, not generic “clarity.”
Optional: send one version (if appropriate) and note what still failed.
Extract reusable moves per audience type for next time.

Example exchange

You

I want to explain why we need sleep for memory to three people: my 14-year-old brother, my biology tutor, and my rowing coach. I'll paste three versions, critique fit and hidden assumptions, but don't rewrite them.

Paste the three versions whenever you're ready.

You

Brother: Sleep helps your brain “consolidate” what you learned. Tutor: Slow-wave sleep supports hippocampal-neocortical transfer after encoding. Coach: Sleep is recovery so you're sharp for 5 a.m. practice.

Brother version still opens with “consolidate,” which you didn't unpack. Tutor version never says what evidence you'd cite if asked. Coach version treats recovery as readiness for early practice, say whether you mean muscular, cognitive, or both. What would each person ask next?

Copyable prompt

Idea: [ ].
Audiences: (1) [real person + what they know], (2) …, (3) …
I will write three versions. Critique each for audience fit and hidden
assumptions. Do not rewrite for me.

The Tell

Here is how you know this method has flipped: you cannot say what each named person would ask next.

Three lengths of the same paragraph are not three audiences.

If the AI rewrote the versions for you, you practised receiving prose, not aiming it.

Principle evidence

Strength of the underlying learning idea, not a claim about AI products.

The underlying learning idea is rated moderate. Rewriting the same explanation for real known audiences is generative practice for rhetoric and transfer of explanation skill. Audience adaptation is a standard writing and communication aim; the evidence is moderate as a teaching principle rather than a single landmark trial.

AI delivery evidence

Whether an AI tutor delivers this method well is a separate question.

AI judgement of audience fit is speculative and often generic. Named real audiences help, but you still decide whether the critique matches those people. Never let the model write the three versions for you.