LibrarySource: Copywriting canon, classics to modern web and SaaS copyv7 · Updated Report

Source: Copywriting canon, classics to modern web and SaaS copy

Status: source note, first research pass on 29 September 2026 by a research agent. Every title, author, year and URL was checked against a live source in that pass; items marked as unverified were not. Evidence class: practitioner canon and single reported tests unless a study is named. Transfer limit: most controlled headline evidence comes from print mail order; confirm specifics on your own audience.

In short

  • Rules it backs: take the words from the customer (principle 2); clarity first: what it is, who it is for, why it matters, why you (6); a fact in place of every superlative (12); the three tests on every claim (13). Principles holds where this note differs.

  • Writing a headline or hero: Part 3: Frameworks and Part 5: Working rules.

  • When the schools disagree (long or short, clear or curious, hype or understatement): Part 4, resolved in Conflicts.

  • What to read first: Top 8 must-reads.

Top 8 must-reads (for a founder writing SaaS and consumer web copy)

  1. Breakthrough Advertising (Schwartz, 1966). A framework for choosing which headline to write. The reader's awareness and the market's sophistication decide the message before any wording does. It is hard going, so read the chapters on awareness and sophistication first.

  2. Scientific Advertising (Hopkins, 1923). Short, free, and still the clearest argument for specificity, "salesmanship in print" and testing.

  3. Tested Advertising Methods (Caples, 1932; revised 1997). The strongest historical evidence on headlines. It is based on keyed-coupon returns and gives the ranking self-interest > news > curiosity.

  4. Copyhackers: review mining and conversion copywriting (Wiebe). A practical method for getting the customer's own words onto the page. This research step is what most founder copy skips.

  5. Wynter: B2B Message Layers (Laja). Clarity, relevance, value, differentiation, in that order. A model for the order of a B2B homepage.

  6. Julian Shapiro: Landing Pages. The most concrete breakdown of a modern landing page: header test, subheader, features as objection handling, CTA.

  7. Harry Dry: Marketing Examples and his "three rules". A fast library of before/after examples. The three tests (can you picture it, can it be proven false, could nobody else say it) are a short self-edit checklist.

  8. NN/g: Concise, Scannable, and Objective (Morkes and Nielsen, 1997). Evidence specific to the web: people scan, and copy that was concise, scannable and objective measured 124% more usable than a promotional version of the same site, in one 1997 test with 51 users (web reading evidence).

Runners-up:

  • Ogilvy on Advertising: overlaps Hopkins and Caples but is easier to read.

  • Paul Graham's "Write Simply": the case for plain style.

  • April Dunford's Obviously Awesome: positioning has to come before copy.

  • Ann Handley's Everybody Writes: writing process and editing.

Part 1: The classics

Scientific Advertising

  • Author / year / format: Claude C. Hopkins, 1923, book (public domain, about 20,000 words)

  • Summary: The origin of measured, test-driven advertising. Ogilvy said nobody should work in advertising "until he has read this book seven times."

  • Principles:

    • Replace every superlative with a figure (chapter "Being Specific").

      • "Platitudes and generalities roll off the human understanding like water from a duck."

      • "Our prices have been reduced" does nothing; "reduced 25 per cent" gets the full value.

      • "Our net profit is 3 per cent" beat "Lowest prices in America."

    • Judge every line with one question: "Would this help a salesman sell the goods?" (chapter "Just Salesmanship"). Cut any cleverness a salesperson would never say to a customer.

    • Use the headline to pick out the right reader, not to entertain: "We pick out what we wish to read by headlines." Hopkins's example: a story about the most beautiful woman in town goes unread under the headline "Egyptian Psychology."

    • Assume a hurried reader. People "skip three-fourths of the reading matter."

    • Once you have attention, tell the full story: "Bring all your good arguments to bear." Leave out one claim and you lose the readers it would have convinced (chapter "Tell Your Full Story").

    • Test small before spending big. Let returns, not opinions, pick the claims and headlines (chapter "Test Campaigns").

Confessions of an Advertising Man / Ogilvy on Advertising

  • Author / year / format: David Ogilvy, 1963 and 1983, books

  • Summary: Ogilvy brought direct-response discipline (research, facts, testing) into brand advertising. He is the most readable bridge between the mail-order school and brand work.

  • Principles:

    • Spend most of your effort on the headline: "On the average, five times as many people read the headline as read the body copy. When you have written your headline, you have spent eighty cents out of your dollar."

    • Put the reader's own keyword in the headline so the right person picks themselves out. His examples: "BLADDER WEAKNESS", "MOTHERS".

    • Headlines that promise a benefit or carry news are the safe bets.

    • Facts beat puffery. Show the benefit specifically and factually. His Rolls-Royce headline proves quality with one checkable detail: "At 60 miles an hour the loudest noise in this new Rolls-Royce comes from the electric clock."

    • Respect the reader: "The consumer isn't a moron; she is your wife." (Confessions, via Ogilvy's quotation book)

    • Research and positioning come before execution. Ogilvy: "Unless your advertising is built on a BIG IDEA it will pass like a ship in the night" (How to Create Advertising That Sells, via Ogilvy's quotation book).

Breakthrough Advertising

  • Author / year / format: Eugene M. Schwartz, 1966, book (reprinted)

  • Summary: It turns "write a good headline" into a diagnosis with two questions: what does this reader already know, and how many similar claims has this market already heard?

  • Principles:

    • Channel desire; do not try to create it: "Copy cannot create desire for a product. It can only take the hopes, dreams, fears and desires that already exist in the hearts of millions of people, and focus those already-existing desires onto a particular product."

    • Write the headline to the reader's awareness stage:

      • Most aware: little more than the product name and the offer.

      • Product aware: what is new or better.

      • Solution aware: why this approach wins.

      • Problem aware: name the problem, then the solution.

      • Unaware: no price, no product name, no function. Write "an identification headline" that echoes an emotion and calls the market together.

    • The more aware the reader, the less you need to say. The less aware, the more the copy has to build before the product can appear.

    • Work out how sophisticated the market is:

      • First in a market: state the claim simply and directly, and prove it.

      • Second: enlarge the claim.

      • When the claims are used up: introduce a new mechanism (how it works).

      • When rivals copy the mechanism: enlarge or elaborate it, so it works easier, quicker or more surely.

      • When the mechanisms are used up (the fifth stage): help the reader identify with the product.

    • When everyone makes your claim, sell the mechanism: name it, describe it, feature it.

Tested Advertising Methods

  • Author / year / format: John Caples, 1932. A later edition revised by Fred E. Hahn (Prentice Hall, copyright 1997) reprints David Ogilvy's foreword to the fourth edition. Book.

  • Summary: Headline advice backed by keyed-coupon returns, not taste. Caples wrote the ad "They Laughed When I Sat Down at the Piano" (1926).

  • Principles:

    • Rank headline types by evidence: self-interest (a reader benefit) first, news second, curiosity third. "For every curiosity headline that succeeds in getting results, a dozen will fail."

    • What you say matters more than how you say it.

    • Long headlines that say something beat short headlines that say nothing.

    • Specifics are more believable than generalities.

    • Test all the time. Caples reports one mail-order ad selling "19-1/2 times as much merchandise" as another in the same space and publication.

    • Long copy sells more than short copy (Ogilvy's foreword to the fourth edition). Caples adds that long copy is for crowding in sales arguments, not for filling space.

How to Write a Good Advertisement

  • Author / year / format: Victor O. Schwab, 1962 (Harper), book

  • Summary: A compact course built from split-run tests. It includes a list of 100 good headlines with the reasons they worked, and a chapter of 14 split-run tests.

  • Principles:

    • Follow the five fundamentals in order: get attention, show people an advantage, prove it, persuade people to grasp the advantage, ask for action.

    • Never skip "prove it." The step between the promise and the ask is where web copy is usually thinnest.

    • Borrow headline patterns from a tested catalogue before inventing new ones.

    • Test headline, appeal, copy length, layout and call to action one variable at a time.

The Robert Collier Letter Book

  • Author / year / format: Robert Collier, 1931 (McGraw-Hill), book

  • Summary: The classic book on persuasive letters, and the source of the most quoted idea in copywriting.

  • Principles:

    • Enter the conversation already taking place in the reader's mind. The popular one-line version is a paraphrase. In the 1950 edition Collier writes that every reader holds "a mental conversation with himself, the burden of which is his own interests" and that a letter should "chime in on that conversation" (full text: https://idoc.pub/documents/the-robert-collier-letter-book-6nq8egkr7pnw, read 29 October 2026).

    • Open with what the reader is already thinking about (a current worry, event or desire), then lead into your offer.

    • Understand the reader's needs and desires before writing. The letter sells a desire fulfilled, not an object.

The Boron Letters

  • Author / year / format: Gary C. Halbert, letters to his son Bond written from prison in 1984 (chapter 1 is dated June 12, 1984). Later compiled as a book by Bond and Kevin Halbert.

  • Summary: The most practical, conversational education in direct response. Its "starving crowd" lesson is the best argument that choosing the market matters more than polishing the words.

  • Principles:

    • Find the starving crowd first. Asked which advantage he would want for a hamburger stand (better meat, a better location, the lowest price), Halbert answers: "NONE OF THE ABOVE! The best advantage of all is a STARVING CROWD."

    • Structure a sales message as attention, interest, desire, action, and ask for the action now (chapter 16).

    • Train your ear by copying great ads out by hand (chapter 17).

    • Read your copy aloud, and let drafts rest before revising (chapters 15, 18 and 22, per summaries).

    • Use simple, everyday words, short sentences and short paragraphs.

Advertising Secrets of the Written Word / The Adweek Copywriting Handbook

  • Author / year / format: Joseph Sugarman, 1998. Revised and expanded as The Adweek Copywriting Handbook, 2006 (Wiley). Books.

  • Summary: An account of how to keep someone reading, from the mail-order writer behind BluBlocker (1986).

  • Principles:

    • "The sole purpose of the first sentence in an advertisement is to get you to read the second sentence."

    • "All the elements in an advertisement are primarily designed to do one thing and one thing only: get you to read the first sentence of the copy." Headline, image and subhead all serve the first line.

    • Build a "slippery slide": every sentence pulls the reader into the next, and the first sentence is short and easy.

    • Write as much copy as the sale needs, and no more.

    • Write one-to-one, in a personal, honest voice the reader can nod along with.

The Ultimate Sales Letter (optional)

  • Author / year / format: Dan S. Kennedy, first published around 1990 (sources give 1990 or 1991; later editions 2000, 2006, 2011), book

  • Summary: Mainly useful as the source of the Problem-Agitate-Solve formula. Copyhackers writes that Kennedy called PAS the most reliable copywriting formula for sales ever invented (https://copyhackers.com/2015/10/copywriting-formula/, read 29 October 2026). The rest is aimed at sales letters and info-marketing more than SaaS.

  • Principles:

    • State the problem factually, then agitate (make its cost felt), then present the solution.

    • Write for a defined list of prospects, not the general public.

    • Treat agitation as a dial. On a SaaS page, stating the cost accurately usually beats dramatizing it.

Part 2: Modern web and SaaS copy

Copyhackers (Joanna Wiebe): review mining and conversion copywriting

  • Author / year / format: Joanna Wiebe. "How to Find Your Message Using Review Mining" (Oct 2014), "Conversion copywriting defined" (June 2023), "Copywriting Formulas" (Oct 2015, since updated). Site and courses.

  • Summary: Makes research into customer language the foundation of copy, with a repeatable method and test results.

  • Principles:

    • Mine reviews of adjacent products, in places where your prospects already talk about the problem: Amazon book reviews, TripAdvisor, review sites, forums.

    • Sort what you find into "what they love," "what they hate" and "what they worry about," and keep phrases word for word: "Their words become yours. You'll take what they say, and mirror it back to them."

    • Let the customer's language beat yours. For a rehab center, the review-mined headline "If you think you need rehab, you do." got over 400% more clicks than "Your Addiction Ends Here" and 20% more form submissions later in the funnel. This is a single test reported by Wiebe.

    • Work in three phases: research and discovery; writing, wireframing and editing; validation and experiments. Frame recommendations as hypotheses.

    • "Copywriting formulas work best when you use them with copy research." Formulas give the structure; research gives the content.

Wynter / CXL (Peep Laja): message testing and the B2B Message Layers

  • Author / year / format: Peep Laja (founder of CXL; founder and CEO of Wynter). Framework article (November 2022), site and testing product.

  • Summary: A model built from thousands of B2B message tests. It tells you the order in which a page must answer the reader's questions. Wynter also tests messages with panels of the target audience, which suits products without enough traffic for A/B tests.

  • Principles:

    • Answer four questions in order:

      • Clarity: what is it?

      • Relevance: is it for me?

      • Value: do I want it?

      • Differentiation: why you?

      • "Buyers cannot judge value until they understand the offer."

    • Lead with the category in plain language (for example "project management application"), then back it up.

    • Make relevance concrete. Name the specific pains and use cases of the target buyer, not broad benefits for everyone.

    • Differentiate by naming the alternatives and backing each claim with proof: "Do not invent difference by adding adjectives."

    • Fix the earlier layers before the later ones. Change one hypothesis per revision so the feedback stays readable.

Marketing Examples (Harry Dry)

  • Author / year / format: Harry Dry. Site (ongoing) and a podcast interview, "Learn Great Copywriting in 76 Minutes," on David Perell's How I Write (July 17, 2024).

  • Summary: The fastest way to learn from examples: short before/after case studies aimed at founders.

  • Principles:

    • Run three tests on every line: Can I visualize it? Can I falsify it? Can nobody else say this?

      • "He looks like Ryan Gosling" beats "He is handsome."

      • New Balance's "Worn by supermodels in London and dads in Ohio" is easy to picture.

    • Write calls to value, not calls to action: "Create your website" rather than "Sign up now."

    • Name the target customer outright. People pay attention when something is clearly for them.

    • Start ideas from substance, not adjectives: "Adjectives are okay. Just ideas don't start with them." Avoid stock landing-page words such as unlock, unleash and empower.

    • Let customers write the copy. Ask one question at signup ("what is your biggest frustration with ...?") and collect the answers (his Design Academy example).

    • Prefer short sentences, active voice and a structure people can scan. Never exaggerate: "An honest line always feels warmer."

Julian Shapiro: Landing Pages (Startup Handbook) and Writing Well Handbook

  • Author / year / format: Julian Shapiro. Online handbooks, undated and maintained.

  • Summary: The most concrete, founder-oriented breakdown of a landing page, by the founder of the growth agency Demand Curve.

  • Principles:

    • Think of conversion as "Purchase Rate = Desire - (Labor + Confusion)": raise desire, cut effort and confusion.

    • Apply the header test: "If the visitor reads only this text on your page, will they know exactly what you sell?" Prefer "Visually design and develop sites from scratch. No coding." to "Improve your workflow!"

    • Use the subheader (one or two sentences) to say how the product works and which features make the header's claim believable.

    • Write each feature as a value proposition plus the objection it raises. Find objections by asking customers "What almost stopped you from buying?"

    • Make the CTA the next step toward the header's claim: "Start learning," not "Request a meeting."

    • "Don't fear writing a long homepage. But, ensure your hero is incredible." The writing handbook adds: write a terrible first draft fast, then rewrite hard for clarity.

Everybody Writes

  • Author / year / format: Ann Handley, 2014; 2nd edition October 2022 (Wiley). Book, plus essays on her site.

  • Summary: The best process book for people who are not copywriters but have to write marketing text every week. Strong on editing.

  • Principles:

    • Accept "The Ugly First Draft" as necessary, then edit hard. Good writers "are excellent editors of their own work."

    • Edit in four passes:

      • Put everything on the page.

      • Keep only the strongest points.

      • Take the reader's view: what do they already know, and what should they do?

      • Add delight, without trading clarity for cleverness.

    • "No one will ever complain that you've made things too simple to understand." "Assume the reader knows nothing, but don't assume the reader is stupid."

    • "Buzzwords and jargon are the chemical additives of writing."

    • Sales copy should sound like the product it sells, "like the same author wrote it." Use short, parallel bullet lists of three or five.

    • "White space is oxygen." Format for people who skim and scroll, and read your copy aloud.

VeryGoodCopy

Customer Camp / Why We Buy (Katelyn Bourgoin)

  • Author / year / format: Katelyn Bourgoin. Weekly newsletter Why We Buy and courses (customercamp.co redirects to learnwhywebuy.com, read 29 October 2026).

  • Summary: Buyer psychology plus a concrete interview method, built around the moment a person starts looking for a solution.

  • Principles:

    • Interview recent buyers who fit your target best; they still remember how they bought.

    • Pull out five things:

      • the trigger event that started the search

      • their goals (the job they need done)

      • the alternatives they considered

      • what they disliked about those alternatives

      • their "selfish desires"

    • Write for trigger moments. Every purchase starts when a person goes from not knowing they have the problem to actively looking for a fix.

    • In interviews, ask about what people actually did. Do not pitch what you are building.

Sway Copy (Josh Garofalo) (optional)

  • Author / year / format: Josh Garofalo. A SaaS copywriting practice, plus interviews (for example Omniscient Digital, Jan 12, 2022).

  • Summary: Practitioner advice specific to SaaS sites (homepage, features, pricing). He has published little long-form material, so treat this as a supplement.

  • Principles:

    • Interview customers who bought in the last 6 to 12 months and still remember why they chose you over the alternatives.

    • Headlines state what you do, who it is for, and why you are different.

    • Replace short testimonials in sliders with long, specific ones placed next to the claim they support.

    • Open the pricing page with value and ROI proof before the tiers.

    • Keep features together. Spreading them across many pages makes a platform look like a single-purpose tool.

Paul Graham: essays on writing

  • Author / year / format: Paul Graham. "Writing, Briefly" (March 2005), "How to Write Usefully" (February 2020), "Write Simply" (March 2021). Essays.

  • Summary: The argument for plain style from a founder-essayist; a counterweight to hype.

  • Principles:

    • "Write a bad version 1 as fast as you can; rewrite it over and over." "Cut out everything unnecessary."

    • Write in a conversational tone with simple, Germanic words. Read it aloud to find where you stumble.

    • "Writing simply keeps you honest. If you say nothing simply, it will be obvious to everyone, including you."

    • "The easier something is to read, the more deeply readers will engage with it."

    • Useful writing "tells people something true and important that they didn't already know, and tells them as unequivocally as possible." Importance, novelty, correctness and strength multiply together.

37signals / Basecamp

  • Author / year / format: 37signals (Jason Fried, David Heinemeier Hansson). Getting Real (2006, free online), Signal v. Noise posts, interviews.

  • Summary: The model for opinionated, plain SaaS copy written by the founder. In a December 2023 interview, Fried said he still writes "all the key pages" (landing, pricing, features) himself.

  • Principles:

    • "Copywriting is interface design. Great interfaces are written." Button labels, empty states and help text are all copy.

    • "Don't use acronyms or words that most people don't understand. Don't use internal lingo."

    • Clarity first: "CLARITY is of utmost importance on this page." Do not give up clarity for cleverness.

    • Start with words: "everything always starts with writing." If a feature cannot be explained clearly in writing, it "probably isn't ready."

    • "Write what I'd want to read."

Positioning before copy: Obviously Awesome

  • Author / year / format: April Dunford, 2019; updated and expanded second edition 2026 (five components, five steps), book

  • Summary: Copy most often fails because the positioning is unclear. Dunford gives a method for deciding what the page should say before anyone writes it.

  • Principles:

    • Define five components, each building on the one before: competitive alternatives, unique attributes, value, target customers, market category.

    • Choose the market category on purpose. It sets the reader's expectations and is the "clarity" layer of the page.

    • Work out the value from what is unique compared with the real alternatives, including spreadsheets or doing nothing.

Evidence on how people read the web

The full evidence, with sample sizes and grades, is in the web reading note; this section summarizes it.

  • NN/g, "Concise, SCANNABLE, and Objective: How to Write for the Web" (John Morkes and Jakob Nielsen, 1997), https://www.nngroup.com/articles/concise-scannable-and-objective-how-to-write-for-the-web/ . Companion article "How Users Read on the Web": https://www.nngroup.com/articles/how-users-read-on-the-web/

    • 79% of test users always scanned new pages; 16% read word by word ("How Users Read on the Web").

    • Compared with a promotional version, usability was 58% higher for concise copy, 47% for scannable, 27% for objective, and 124% with all three combined.

    • Users "detest" marketese and prefer facts.

    • Caveat: a 1997 study with small samples, though later NN/g work points the same way.

  • Unbounce Conversion Benchmark Report (2024), https://unbounce.com/conversion-benchmark-report/

    • Landing pages written at a 5th to 7th grade reading level converted at 11.1%, against 7.1% at 8th to 9th grade and 5.3% at a professional level.

    • Higher word counts went with lower conversion (a -18.6% correlation), and difficult words of three or more syllables more so (-24.3%). Financial services was an exception.

    • Caveat: these are correlations across many pages, not controlled tests.

Exemplars, with analysis sources

  • Apple.

    • Sources: Ken Segall, Insanely Simple (2012), https://en.wikipedia.org/wiki/Ken_Segall . Segall was Jobs's agency creative director for 12 years, worked on "Think different," and came up with the "i" naming of iMac, iPod, iPhone and iPad. Harry Dry, "How Steve Jobs clarified Apple's message," https://marketingexamples.com/copywriting/think-different , contrasts the spec-heavy Lisa launch with the two words "Think different."

    • Lessons: boil the message down to one idea. Consistent naming is part of the copy. Identity and story work after the brand is known (Schwartz's fifth stage of market sophistication), not for an unknown product.

  • Stripe.

    • Sources: RC Victorino, "How Stripe Built a Writing Culture" (Slab, Sept 2, 2020), https://slab.com/blog/stripe-writing-culture/ , covers short paragraphs, subheads, bullets, sample documents and review of high-impact documents. Homepage, https://stripe.com (read 29 October 2026): the headline leads with the category ("Financial infrastructure to grow your revenue" in the English page title). A customer logo strip sits directly under the hero. Proof by numbers follows (135+ currencies and payment methods, 99.999% historical uptime), and the enterprise section adds that 50% of the Fortune 100 have used Stripe.

    • Lessons: category plus outcome in the headline (Wynter's clarity layer), numbers as proof instead of adjectives (Hopkins), logos as social proof (Shapiro).

    • Flag: no rigorous independent teardown of Stripe's marketing copy was found. This analysis is the research agent's own, applying the frameworks above.

  • Linear.

    • Homepage, https://linear.app (checked Sept 2026): "The product development system for teams and agents" / "Purpose-built for planning and building products. Designed for the AI era."

    • Beliefs live on a separate narrative page, https://linear.app/readme . Working principles are at https://linear.app/method (for example "Write issues not user stories").

    • Analysis: Daniel Schwarz, "Linear design: The SaaS design trend that's boring and bettering UI" (LogRocket, June 7, 2025), https://blog.logrocket.com/ux-design/linear-design/ , on direct, minimal copy paired with restrained visuals.

    • Lessons: put the category and the audience in the headline, with a one-line subhead. Put the manifesto on its own page so the homepage stays easy to scan. Short, declarative principles read as confidence.

Part 3: Frameworks

FrameworkSource (checked)What it isWhen to use itCaution
AIDAAttributed to E. St. Elmo Lewis (1898). The acronym was coined by C. P. Russell in Printers' Ink (1921). A historical paper argues the Lewis attribution rests on weak evidence: https://ojs.library.carleton.ca/index.php/pcharm/article/view/4340/3310Attention, interest, desire, actionLong pages, launch emails, sales letters (Halbert ch. 16)Wiebe notes most copy forgets Desire. Schwab's five fundamentals add the missing "prove it" step
PASDan Kennedy, who by Copyhackers' account called PAS the most reliable copywriting formula for sales ever invented: https://copyhackers.com/2015/10/copywriting-formula/Problem, agitate, solveReaders who know they have the problem; a single section or emailAgitation slides easily into promotional language, which NN/g's 1997 test found less usable than objective copy. Use the customer's words for the problem and keep the agitation factual
Before-After-BridgeWidely taught; no originator was found in the sources checkedCurrent state, desired state, how the product gets you thereHero and subhead, feature blocks, case studies and testimonialsThe "after" must be specific and checkable, or it is just a superlative
4 U'sMichael Masterson and Bill Bonner, AWAI, June 18, 2001: https://www.awai.com/2001/06/a-review-of-the-4-us/Urgent, unique, ultra-specific, useful. Originally a check for bullets, now used to grade headlinesGrading candidate headlines and bulletsUrgency is rarely honest on a SaaS homepage. "Unique" is the same test as Dry's "nobody else can say this"
Awareness and sophisticationSchwartz (above)Choose the message by what the reader knows and what the market has already heardEvery page, per traffic sourceNeeds research; guessing the stage defeats the point
Message LayersLaja / Wynter (above)Clarity, relevance, value, differentiationB2B homepages and product pagesA clarity problem hides everything else, so test in that order
One reader, one ideaGreat Leads (Michael Masterson and John Forde; AWAI release Jan 17, 2012): https://www.awai.com/2012/01/just-released-great-leads/ . It gives six lead types from direct to indirect: offer, promise, problem-solution, big secret, proclamation, story; summaries stress a single "big idea." Also Ogilvy's Big Idea, Shleyner's persona lesson, Dry's "write for one reader" and Sugarman's one-to-one voiceEach page argues one core idea to one real personEvery pageTwo-sided products need a page per side, not one page for both readers
Specificity over superlativesHopkins ("Being Specific"), Caples ("Specifics are more believable than generalities"), the 4 U's, Dry (falsify), Wynter ("Do not invent difference by adding adjectives"), NN/g (objective copy +27%)Numbers, names, mechanisms, checkable claimsEverywhereSpecific claims must be true. A false number is worse than a vague one

Headline formulas with evidence behind them (strongest first):

  1. Self-interest (reader benefit) headlines, then news, then curiosity. Caples, from keyed-coupon returns: "for every curiosity headline that succeeds ... a dozen will fail."

  2. Long headlines that say something beat short ones that say nothing. Caples, as summarized by Ogilvy.

  3. The reader's own keyword or category in the headline. Ogilvy; this matches Wynter's "lead with the category."

  4. Headlines built from customer language. Copyhackers' review-mined headline got 400% more clicks in one reported test.

  5. Headlines matched to the reader's awareness. Schwartz; a framework from experience rather than a controlled study, but the most complete one.

  6. Pattern catalogues: Schwab's "100 good headlines" and Caples's chapter of 35 headline formulas (fifth edition, 1997; the fourth edition has 29).

Almost all controlled headline evidence comes from print mail-order. It carries over to the web as principles; confirm the specifics by testing on your own traffic.


Part 4: Where the schools disagree, and when each applies

These are the source positions. The organization's rulings on these tensions are in Conflicts; where the two differ, Conflicts holds.

  1. Long copy vs short copy.

    • For long: Hopkins ("Tell your full story"), Ogilvy and Caples ("long copy sells more"), Sugarman ("as much copy as you need").

    • For short: NN/g (concise copy +58%), Unbounce (more words, lower conversion), and the minimalism of Apple and Linear.

    • Resolution: length follows the reader's awareness and the size of the decision. Schwartz: the more aware the reader, the less you need to say. Ogilvy said long copy sells more "particularly when you are asking the reader to spend a lot of money" (Ogilvy on Advertising, 1983, confirmed through reviews; see Verification notes). Shapiro reconciles both for the web: a short, complete hero, then a long page for readers who keep scrolling. Every section still has to pass the concise test.

  2. Clarity vs curiosity and cleverness.

    • For clarity: Caples ranks curiosity third, 37signals and Wynter put clarity first, and Handley warns against trading clarity for cleverness.

    • For intrigue: Halbert, Dry ("avoid contained titles") and Schwartz's identification headlines.

    • Resolution: homepages, product pages and pricing pages need clarity, because their visitors already know solutions exist or know your product. Curiosity belongs in ads, emails, social posts and content for readers who have not yet started looking.

  3. Hype vs understatement.

    • For hype: direct-response agitation (Kennedy, Halbert).

    • For understatement: Hopkins on superlatives ("They lead readers to discount all the statements that you make"), NN/g's penalty for marketese, Dry's "An honest line always feels warmer," and Handley's warning against sounding like a TV shopping channel.

    • Resolution: for software buyers and markets that have heard every claim, understated, factual copy wins. Agitate only in the reader's own words, and only as strongly as the real cost justifies.

  4. Facts vs story and identity.

    • Facts: Hopkins, Ogilvy and Caples sell with facts.

    • Story: Jobs's "Think different," Linear's README and Shleyner's show-don't-tell sell with story.

    • Resolution: Schwartz explains both. Known brands in saturated markets can sell identity (his fifth stage of market sophistication, identification). An unknown product must first make a clear, specific claim. Stories work inside pages as proof (customer scenes, case studies), not as a replacement for saying what the product is.

  5. Formulas vs research.

    • Formulas: collections such as PAS, AIDA, BAB and the 4 U's.

    • Research first: Wiebe, Bourgoin, Halbert's starving crowd, and Schwartz's "copy cannot create desire."

    • Resolution: formulas give the order; the content comes from customer research. Wiebe says this directly.

  6. Personas vs one real person.

    • Conventional marketing personas, against Shleyner ("A persona is a Frankenstein") and Bourgoin's real buyer stories.

    • Resolution: use segments to decide whom to target, but write each page to one real customer you have interviewed and whose words you have.

  7. A/B testing vs judgment.

    • Hopkins and Caples assume you can measure responses at volume. Early-stage SaaS rarely has the traffic.

    • Resolution: test clarity and relevance with target-audience panels (Wynter's approach) and customer interviews. A/B-test only high-traffic elements (hero, CTA, pricing), one hypothesis at a time.


Part 5: Working rules for website copy (with sources)

These rules sum up this note's sources. The canonical rules are in Principles, which holds where the two differ.

  1. Position before you write. Decide the category, the real alternatives, what is unique and who it is for before you draft a word. (Dunford; Wynter's clarity layer; Ogilvy's positioning and Big Idea)

  2. Research first, then borrow the customer's words. Mine reviews of adjacent products, interview recent buyers, add a one-question signup survey, and keep phrases word for word. (Wiebe's review mining; Bourgoin's Trigger technique; Dry's "Let your customers write your copy"; Garofalo)

  3. Find the starving crowd. Write to a desire that already exists. If nobody is looking for the outcome, better copy will not fix it. (Halbert; Schwartz's "copy cannot create desire")

  4. Work out each page's reader, per traffic source.

    • Pricing and signup pages speak to the most aware: offer and terms.

    • Homepages reached from search speak to people who know solutions exist or know your product: what it is, why it is better.

    • Cold ads and content speak to people who only know the problem, or not even that: lead with the problem or the identity, not the product name.

    • (Schwartz)

  5. Check how worn out your claim is. If every competitor makes it ("AI-powered," "all-in-one"), lead with the mechanism or with numbers, not a louder adjective. (Schwartz; Hopkins's "3 per cent" example)

  6. The hero passes the "read only this" test. Headline plus subhead say what it is, who it is for, and why it is different. The subhead explains how it works in one or two sentences. (Shapiro; Wynter; Garofalo; 37signals)

  7. Clarity before cleverness. Use benefit or news headlines on product pages; save curiosity for ads and emails. (Caples's three classes; 37signals; Handley's fourth pass)

  8. Replace every superlative with a fact: a number, a name, a mechanism, a customer result. (Hopkins's "Being Specific"; Caples; the 4 U's "ultra-specific"; Wynter's "Do not invent difference by adding adjectives")

  9. Run the three tests on each claim. Can the reader picture it? Could it be proven false? Could a competitor sign it? If a competitor could, rewrite it. (Dry; Ogilvy's Rolls-Royce clock)

  10. One page, one reader, one idea. Write to one real customer, addressing them as "you." Two-sided products get a separate page for each side. (Great Leads; Shleyner; Dry; Sugarman; Fried's "write what I'd want to read")

  11. Pair each benefit with the feature that makes it believable. Ask "what is the benefit of the benefit?" for the headline, then prove it with the concrete feature below. (Shleyner; Shapiro on subheaders and features; Schwab's "prove it")

  12. Every section answers an objection. Collect objections by asking customers "What almost stopped you from buying?" and handle the biggest near the top. (Shapiro; Hopkins's "Tell your full story")

  13. Put proof next to the claim. Long, specific testimonials, numbers and logos go beside the claim they support, not in a carousel at the bottom. (Garofalo; Stripe's homepage; Shapiro on social proof; Ogilvy on testimonials)

  14. The CTA continues the promise. Name the value the click delivers ("Start learning," "Create your website"), not the mechanics ("Submit," "Request a meeting"). (Shapiro; Dry's "call to value"; Getting Real on button labels)

  15. Length follows awareness and stakes. Write a short, complete hero, then as much as a hesitant buyer needs, with every section earning its place. (Hopkins; Ogilvy; Sugarman; Schwartz; NN/g on concise copy; Unbounce's word-count finding; Shapiro's "Don't fear writing a long homepage")

  16. Write for people who scan. Use headings that carry the argument on their own, short paragraphs, parallel bullets and white space. (NN/g: 79% scan; Handley's "White space is oxygen"; Dry's "Write scannable copy")

  17. Use plain words. Aim for grade 7 to 9 for general readers and no higher than about grade 12 for specialists (Principles 14). No jargon, no internal lingo, no stock landing-page words. (Unbounce 2024, observational; Graham's "Write Simply"; Handley on buzzwords; Getting Real; Dry)

  18. Keep the tone objective. Understatement earns trust; marketese costs it. (NN/g: objective copy +27%; Hopkins on superlatives; Dry's "Don't exaggerate"; Ogilvy's "The consumer isn't a moron")

  19. Build a slippery slide. Keep the first line short, make each sentence earn the next, and let headline, image and subhead exist to get the first line read. (Sugarman; Dry's tips 14 and 16)

  20. Draft ugly, edit in passes, read aloud. First everything, then only the strongest points, then the reader's view, then delight. (Handley; Graham's "Writing, Briefly"; Halbert; Shapiro's writing handbook)

  21. Treat product UI text as copy. Write buttons, empty states, errors and onboarding in the same voice as the site, and draft features in words before designing them. (Getting Real, "Copywriting is Interface Design"; Fried's "everything always starts with writing")

  22. Use formulas for structure, not content. PAS for a section aimed at readers who know the problem, BAB for the hero and case studies, AIDA or Schwab's five steps for long pages, the 4 U's to grade headlines. (Kennedy via Copyhackers; AWAI; Schwab; Wiebe's "formulas work best ... with copy research")

  23. Earn identity before you sell it. Unknown products sell specifics. Put the manifesto and story on their own page once the homepage has made the claim clear. (Schwartz's fifth stage of market sophistication; Apple via Segall and Dry; Linear's separate README)

  24. Test messages, one hypothesis at a time. When traffic is low, use panel message tests or interviews for clarity and relevance. When it is not, A/B-test the hero, CTA and pricing. Keep losing tests cheap. (Hopkins's "Test Campaigns"; Caples; Wynter)

  25. Train your ear by copying great work. Copy strong ads and pages out by hand, and imitate writers you admire. (Halbert ch. 17; Graham's "imitate writers you like")


Verification notes

  • Not checked against the original text: the originator of Before-After-Bridge, and a single first-edition year for Kennedy's book (sources say 1990 or 1991).

  • Confirmed only from secondary sources:

    • Ogilvy's long-copy line for expensive purchases, via reviews of Ogilvy on Advertising.

    • Halbert's chapter numbers for the editing tips.

    • The "one big idea" emphasis in Great Leads.

  • Left out because no checkable source was found in this pass: the popular studies claiming numbers in headlines get more clicks. They measure clicks on content headlines, not landing-page conversion, in any case.

  • Web search ran out partway through the pass; after that items were checked by fetching known URLs and archive.org full texts directly. The gap this left is independent teardowns of Stripe's and Linear's copy.

Oddlynew Copy and Messaging PrinciplesSign in

Referenced by

and more