How to learn copywriting

You learn to gather and organise the exact words buyers use before you write a single line of copy. This sits first because every later stage depends on having raw material that is not invented. Without it, formulas produce empty shells.

• Voice of customer mining from reviews, forums, and support tickets
• Pain, desire, objection, and proof sorting
• The 'swipe file' and how to use it without copying
• Reading Amazon one-star and five-star reviews for the same product
• Interviewing five real buyers or users
• Building a research document with verbatim quotes

You can now: You can produce a one-page research document for a real product, with at least 20 verbatim customer phrases sorted into pain, desire, objection, and proof, and a stranger can see which phrase came from which source.

Before writing, you define what is actually being sold and to whom. This stage sits before any copywriting formula because a weak offer cannot be rescued by strong words. You learn to state the promise, the mechanism, and the proof in plain sentences.

• The offer as promise, mechanism, and proof
• Positioning against the obvious alternative
• The 'only' statement: what this product does that no other does
• Price anchoring and the role of the comparison set
• Writing a one-sentence offer that a stranger can repeat back
• The difference between a feature, a benefit, and an outcome

You can now: You can write a one-sentence offer for a real product that names the buyer, the outcome, the mechanism, and the proof, and a stranger can repeat it back without looking at the page.

Now you learn the formulas, but only as containers for researched material. This stage sits after research and offer because formulas are sequencing tools, not idea generators. You practise AIDA, PAS, and the 4 Ps on real research documents.

• AIDA: attention, interest, desire, action
• PAS: problem, agitation, solution
• The 4 Ps: promise, picture, proof, push
• The 'slippery slide' and why it is a property of research, not style
• Headline types: direct, indirect, news, how-to, question, command
• Subheads as a summary of the page
• The call to action as a specific next step, not a button label

You can now: You can take the same research document and write three different structures for it — AIDA, PAS, and 4 Ps — and a reader can tell which is which without being told.

You write full pieces and then cut them. This stage sits after structure because the first draft is where you discover what you actually think; the rewrite is where you make it readable. You learn to write short sentences, concrete nouns, and active verbs under time pressure.

• The first draft as a thinking tool, not a deliverable
• Cutting 30% of every draft without losing meaning
• Concrete nouns and specific numbers over adjectives
• Active voice and the 'by zombies' test
• Reading aloud to catch rhythm and filler
• The 5-second test: can a reader say what the page is about after 5 seconds
• Writing a landing page, an email, and a product description from the same research

You can now: You can take a 500-word draft and cut it to 350 words without losing a single claim, and a reader can say what the page is about after five seconds of looking at it.

You learn to measure whether copy works, not whether it sounds good. This stage sits last because testing requires finished pieces. You learn to read response data, run small split tests, and revise based on what readers actually do.

• Click-through rate, conversion rate, and what they do and do not tell you
• Running a simple A/B test on a headline
• Reading heatmaps and scroll depth
• The difference between engagement and response
• Revising copy based on one metric, not opinion
• The 'one change at a time' rule

You can now: You can run a headline A/B test on a real page, read the results, and say which headline won and why, in one sentence that cites the number.

You assemble three pieces that demonstrate research, offer, structure, and revision, and you put them in front of a real audience. This stage sits last because a portfolio of untested pieces is just a folder. You learn to present work as decisions, not decorations.

• Choosing three pieces that show range: landing page, email, product description
• Writing a case study that names the research, the offer, and the result
• Getting feedback from a stranger, not a friend
• Revising based on feedback without defending
• The portfolio as a set of decisions, not a set of pages

You can now: You can show a stranger your three pieces and they can name, for each one, the research source, the offer, and the structure, without you explaining it first.