Why Your Skin Keeps Aging Despite Creams, Serums & Injections – Rovina
The two chiefs
House chief · 7/10
advertorial, story-led presell · cold traffic to a product page (found on the store's own site, not yet seen behind an ad) · women of 50 to 65 who have spent on topicals and injectables · problem-aware, market sophistication high · cause: oxidation below the surface; answer: a fat-soluble antioxidant that lodges in the cell membrane
Lead: Problem-solution lead told as a secret: a symptom mirror in the second person, a confession, then a discovery loop around a numbered comment. Grade A-minus: the first 8% is the best writing on the page; the loop is teased twice before it pays.
Offer: C. Two for one with a reason, and no price, no dollar anchor for the product, no value build, no bonus. The guarantee carries the offer alone, and carries it well.
Close: B-plus. A two-path close, a guarantee with sensory conditions, scarcity with a stated reason, then a letter that handles the fear of being fooled again. Three buttons, all in the last fifth; the scarcity has no count and no date.
Throughline: A-minus. Self-blame, absolution, hope, a stranger's verdict, relief from the search. It sags from 30 to 40% (the verdict list and the first shopping trip) and recovers on the Japan question.
Strongest
- the descent and the hinge (entries 7 to 10): identification is earned before any claim is made
- absolution before explanation (entry 21), which makes the science welcome
- one cause, three numbered symptoms (entries 26 to 29): every reader finds her own complaint in a heading
- every exit closed as it opens: other antioxidants (entries 33, 34), diet (entry 40), shop-bought astaxanthin (entries 47 to 49), Amazon (entry 68)
- the brand as a scorecard (entry 53), arriving at 59% to pass tests set at 38% and 53%
- the closing letter (entries 73 to 77), a complete small page for the reader who scrolls to the end
Weakest
- proof: nothing is sourced. For its own market the page needs one named human study and one verifiable customer between entries 21 and 53
- buttons: the first ask comes at 81%. A button after the scorecard (entry 54) would catch the reader at her most convinced
- offer: no price and no anchor for the product; the large sum spent on the old way is the only number she can compare with
- seams: repeated blocks (entries 15, 47, 75), a heading that promises steps (entry 64), a week label out of order (entry 59), a maker's pronoun in a customer's mouth (entry 66)
- persona: the byline and the plot contradict each other (entry 6 against entry 52)
Craft chief · REVISE
Read aloud it is a friend confessing at a kitchen table for the first fifth, a brochure with a chorus through the middle, and the same friend again in the closing letter.
Strongest
- entries 7 and 8: three snapshots in the second person, then the pronoun changes on the shortest line; the reader is inside the story before she has agreed to enter it
- entries 36 to 53: criteria, then name, then tests, then brand, so every reveal answers a test she has already accepted
- entry 73: the last objection is an emotion, and it is voiced at the button by someone who felt it
- entry 56: a results timeline that opens by admitting nothing happened
Weakest
- entry 37: an expert and a laboratory with no name, no place and no institution, a stub left from the sibling page
- entry 16: the claim the whole page rests on is locked in a picture with no alt text
- entries 15, 47 and 75: recycled blocks that a careful reader recognizes as repetition
- entry 71: 96 words of explanation at the moment the page should be fastest
001 PASS·noted · the descent in entries 7 to 10 sings (fragments that widen, a one-word hinge); the refrain that closes entries 26 and 27 and two verdicts in 32 turns mechanical, and entry 71 drones at 96 words. · 002 PASS · attention in entries 3 to 5, interest 10 to 21, desire 39 to 46 and 56 to 59, action 63 to 69, then a second complete close as a letter, 73 to 78. The close is finished, twice. · 003 PASS·noted · plain words nearly throughout; the register turns clinical where the kidney template shows through (entries 29, 32, 37). · 004 REVISE · cost: attention. Entry 15 repeats the tease of entry 14; entries 33 and 47 run one shopping scene twice under one heading formula; the heading of entry 64 promises steps that never come; entry 75 retells 56 to 59. · 005 REVISE · cost: trust. The main loop is honest (opened in entry 3, paid in 23 and 42). Two promises are not kept: the deck's study (entry 5) turns out to be a thread she read (entry 10), and the headline's third product is never explained (one mention, entry 19). The page-bound offer in entry 1 has no reason and no date. · 006 PASS · picture-able from first to last (morning light, a jawline, a store window, a travel document), and the deal carries a reason (entry 64). The numbers are noted under 025. · 007 PASS · in the capture, colour marks the three product nouns of the headline, a highlight marks the deck's first sentence, bold marks the two-word verdict in entry 8; capitals are kept for labels (entries 26 to 28, the timeline). · 008 REVISE · cost: momentum in the buying half. Duplicate blocks (entry 15; 47 against 33; 75 against the timeline), one refrain used four times (entries 26, 27, 32), and the longest paragraph placed in the urgency block (entry 71). · 009 PASS·noted · scene and person, a gap planted and held across entries 10 to 17, a moral that equals the premise of the offer (entry 46), the loop called back (entry 70). For source-truth review: narrator, thread and comment are template parts shared with the kidney page. · 010 PASS · contractions, fragments, sentences that open on a conjunction; two off notes no speaker would produce (entries 36 and 74). · 011 REVISE · cost: the argument itself, for any reader with images off. The central claim exists only inside the image of entry 16, and none of the 14 images carries alt text. Headline and hero (entries 3 and 4) do work as one argument. · 012 PASS·noted · the furniture is the template's default (entries 1, 81, 83), identical on the sibling; the three button labels are written, each naming a different step (entries 69, 78, 82). · 013 PASS · entries 7 to 9 name her frustration in her own terms, and entry 73 reaches the root emotion, the fear of being fooled again. · 014 PASS · character, adversity, turn, resolution. The named emotion is vindication (entry 21); the dread of the mirror is set up in the opening and resolved in entries 57 and 75. · 015 REVISE · tier 1. 31 em dashes across 28 rows; contrast flips in entries 14, 21, 46 and 61; stacked fragments of one cadence in entry 29; a throat-clearing frame in entry 25. The reader is never the enemy (entry 21), which passes rule 12. · 016 PASS · no naked internal term on a cold surface; the species name in entry 53 is glossed in its own line, and the ingredient arrives with a picture (entry 43). · 017 PASS · emotion (spend that led nowhere), curiosity gap (the cause), stakes (it goes on); the payoff is honest and the loop closes in entries 23 and 42. · 018 PASS · colour judged as colour: the numbered comment, the Osaka vignette (entry 45), the two ages (entry 59), the store window (entry 75). · 019 REVISE · cost: trust in the signature. The signed person and the story disagree: a researcher of thirty years in the byline (entry 6), the daughter as the family's researcher (entry 52), the maker's plural in a heading (entry 66). · 020 REVISE · tier 1, cloned strings. 35 rows are identical to the store's kidney page and roughly 30 more nearly so, persuasion included: the opening descent, both doubt beats, the closing letter. · 021 N/A · a single page, not a sequence that one buyer receives. · 022 PASS · applied to this pass: each tier-2 REVISE names its cost; the staccato opening, the superlatives and the bold multipliers are left standing; proof worries are notes, never verdicts. · 023 N/A · a cold story lead delays the object by design (009 rule 8); the object arrives in entry 53. · 024 PASS·noted · the product's verbs overreach once: entry 61 says damage is undone, while entry 58 honestly reports a partial result a few screens earlier. · 025 PASS·noted · for the guard. The timeline disagrees with itself (a week-six label after weeks eight to ten in entry 59, the same scene at twelve weeks in entry 75, 127 days in entry 73); three numbers recur on the sibling page under another author (4,237, 5,847, $14,000); 38,000, 95% and both multipliers name no source.
Take
- Mirror first, narrator second: three second-person snapshots before the first 'I', with the pronoun changing on the shortest line (entries 7, 8).
- Plant self-blame early and lift it in a heading before the science begins (entries 8 and 21).
- One cause, three numbered symptoms, each label pairing the cause with the sign she can see (entries 26 to 29). The same three signs appear four times: mirrored, questioned, explained, resolved in the timeline.
- Criteria, then name, then tests, then brand: specification at 38%, ingredient at 46% in a one-word heading, label tests at 53%, brand at 59% as a ticked scorecard (entries 36, 42, 48, 53).
- Close each exit the moment it opens: other pills, diet, the shop shelf, the marketplace (entries 33, 40, 47, 68).
- Name the idea in one short sentence at the midpoint (entry 46), so the reader crosses into the buying half with something to repeat.
- A results timeline that opens by admitting nothing happened, with the reason planted 14% earlier (entries 44 and 56).
- Let a third party deliver the peak, with the picture planted by the expert a quarter of the page earlier (entries 40 and 59).
- Decide in principle before hearing terms: promise, two paths, then deal and guarantee, and only then a button (entries 61 to 69).
- A closing letter built on the last emotional objection, complete enough to work for the reader who scrolled past everything (entries 73 to 77).
- Button labels that escalate: get it, check it, test it (entries 69, 78, 82).
- Planted numbers that return: the years (five times), the sum spent (three), the minutes of hesitation (two), so the story feels kept by one memory.
- Proportions by position: chrome, hook and descent 0 to 8%; thread and comment 8 to 19%; absolution, cause and verdicts 19 to 33%; wrong pills and specification 33 to 41%; desire story and ingredient 42 to 51%; label tests 51 to 56%; brand and timeline 56 to 68%; promise, paths and terms 70 to 81%; urgency and letter 81 to 95%.
Leave
- Fabrication: the invented narrator, her credentials, the physician-style avatar and the ratings count (entry 6); the number is shared with another author on the same store.
- Fabrication: the thread of 38,000 women, the numbered comment, the hero collage and every screenshot of them (entries 4, 10 to 17, 70). Composed, and unverifiable by anyone.
- Fabrication: the unnamed researcher and laboratory, the Osaka woman, the stranger at the market, and any before-and-after picture (entries 37, 45, 59, 60).
- The verdict that her creams, serums and injections did nothing (entries 3, 8, 24, 30, 32, 63). Our frame is solution-side only: The Cell Shield makes what she already takes hold.
- Breakdown that outruns rebuilding (entry 26) and collagen narrated as being torn apart (entries 18, 23, 44): one sentence from the banned enzyme attack (p2), with the banned plateau story (p1) next door.
- Reversal claims: damage undone, collagen rebuilt, years taken off (entries 45, 59, 61). The gate holds wrinkle depth, elasticity and hydration against placebo, and no more.
- Every Japan claim: shops, skin doctors, the 1980s (entries 39 to 45, 54, 63). Nothing in the gate supports a population story.
- The carrier-oil test and tick (entries 48, 53, 68): carrier superiority is banned (b8).
- The 90-day guarantee (entries 65, 69, 81, 82): ours is the 30-Day Money-Back Guarantee.
- Scarcity with no count, no date and no evidence, and the undated update (entries 70, 71, 78, 79).
- A first button at 81% and a deal with no price.
- The template seams: repeated blocks, a heading that promises steps and gives none, week labels out of order, a maker's pronoun in a customer's story (entries 15, 47, 59, 64, 66, 75).
- The word mechanism in customer copy (entry 32), 31 em dashes, and the stack of contrast flips.
- A claim locked inside an image, and 14 images with no alt text (entry 16).
- The narrator dropping her routine for water and a softgel (entries 45, 55): ours replaces nothing.
- Blame placed on dermatologists as a class, and the doctor used for pressure (entries 5, 73): on our pages doctors are allies.
Line by line · 83 entries
Every line is described in our words; the page's own text stays on the page (open it beside this table). Shaded rows carry the most pull.
| # | Beat | What the line does | Lever · reader | Why here | Craft | On our page |
|---|---|---|---|---|---|---|
| 1 0.0% | page-furniture p · 9 w | A full-width bar above everything announces a time-limited offer that is valid only through this page. | exclusivityurgency problem-aware | It is the first thing painted, so the idea of a page-bound deal rides along for the whole read and is old news when the terms arrive at 76%. Nothing has to be believed yet; it only has to be seen. It returns as the last row, a bookend. | All capitals, two halves split by a vertical bar; no number, no date, no reason. Template furniture: the identical bar tops the store's kidney page. It also works against the editorial costume that the next two rows put on. | page-furniture A top bar only if it is true. On our editorial chassis the strip above the masthead is the advertising disclosure; the launch-week code line belongs to the PDP and checkout. |
| 2 0.3% | page-furniture fine · 8 w | A three-step breadcrumb files the piece under a health and wellness section, in a category about the health of cells. | borrowed-authority problem-aware | It sits where a news site prints its section path, between the bar and the headline, so the eye registers an article before it reads the claim. The category word also seeds the cell frame that the cause section leans on from 20%. | Small grey type, easy to skip, which is the point: it works on the reader who does not study it. The last crumb is the brand's vocabulary, not a reader's. | page-furniture Our chassis carries its own masthead and crumbs; editorial chrome never contradicts the disclosure strip (L-07). |
| 3 0.5% | hook-headline h2 · 17 w | The headline promises the true cause behind skin that goes on aging, and lists three things the reader has paid for (a topical, a serum, an injectable) as having achieved nothing. | curiosity-gapmirroringvindication problem-aware | It has to select the reader and open the loop in a single line: the cause is paid at 21%, the answer at 46%. It comes before any narrator because the narrator's standing depends on sharing this exact failure. | Seventeen words, one breath. In the capture the three product nouns are set in red, so the stress falls on her purchases and not on the promise. The closing absolute overstates, and a reader whose retinol did something will reject it. The third noun is never paid: the body explains why two topicals and a moisturizer fall short, and injectables get one passing mention in entry 19. | rewrite-claim The device transfers (one reason, her own shelf named). The verdict that her products did nothing does not. Ours is solution-side: what she already takes for her skin finally holds (CS-A08). The H1 is recognition in her words (CL-06); The Cell Shield never leads it. m1p6 |
| 4 1.2% | image img | Hero image, seen in our capture: a collage of five dark-mode social comment cards, each with a woman's name, avatar, time stamp and reaction counts, all discussing a bottle that is never named. | social-proofconsensuscuriosity-gap problem-aware | Set between headline and deck, so the first evidence she meets is peers, ahead of science and ahead of the narrator. It pre-sells the thread story that starts at 8%: she has seen the crowd before she is told about it. What they are all taking becomes a second open loop beside the headline's. | Fabrication: a designed collage, not a screenshot of a thread, and nobody could verify a card. The voices are well made: lower case, under-punctuated, five jobs for five cards (a surprised result on the arms after a daughter's nagging, a question about taking it with food, a hedge plus a husband's remark, a friend tagged with a note that its reviews go back years, a tried-everything verdict). One commenter shares the narrator's first name. | drop A composed comment collage does not transfer. The slot (peer voices above the fold) opens only when we hold real, attributable customer words; until then the hero is the softgel or the scene. |
| 5 2.5% | deck p · 26 w | The deck makes the failure plural, claims the narrator examined more than 38,000 women whose skin aged ahead of schedule, and says skin doctors and beauty counters all overlooked what she found. | earned-authorityspecificitycommon-enemyidentification problem-aware | After a headline and a crowd, the reader needs a who. The deck brings in 'we' and 'I' in the same breath, so the narrator is at once a fellow failure and an investigator. It precedes the byline so that the byline reads as confirmation of a claim already made. | Title case throughout makes it a second headline; in the capture its first sentence is highlighted. The verb of study is the dishonest word: the body shows she read a social thread, and the sample is a member count. The sibling kidney page runs the same deck with a different number and different specialists. | rewrite-claim We hold no study of 38,000 women and we never blame her dermatologist. Our deck may point to the human trials, to the body of published work, and to how few women have met astaxanthin. Missing claim: any population count of our own. b1r14p7 |
| 6 3.2% | authority-stamp 3× p · 19 w | A star row with 4,237 ratings, then the byline: Margaret Ellison, described as a skin researcher and cosmetic formulator of thirty years in San Diego, dated early September 2026. | earned-authoritysocial-proofspecificity problem-aware | It answers the question of who she is to say so one line after the deck raises it, and before the first-person body begins, because a confession only works from someone the reader has already granted standing. | Fabrication: the persona is a template part. The ratings figure is identical on the store's kidney page under a different author, and an article cannot have star ratings. In the capture the avatar is a woman in a white coat with a stethoscope beside a green tick: a formulator dressed as a physician. The credential also fights the plot, in which a researcher of thirty years learns of free radicals from a stranger's comment. | swap-proof The byline slot takes the brand's own advisor persona, a composite disclosed in the footer (CL-14), with a true date and no ratings count until a real one exists. r19 |
| 7 4.0% | symptom-mirror 3× p · 48 w | Three verbless snapshots addressed to the reader: a face that ages each year against the promises of what she owns; a dull morning tone and new slackness at the jaw; the be-patient speech her bottles keep giving her. | mirroringidentificationspecificity problem-aware | The body opens on 'you', not 'I'. The narrator has just been credentialed; now the page proves it knows the reader's mornings before asking her to care about the narrator's. The same signs (tone, jaw, products that ask for time) are the ones the cause section explains at 24 to 28%, so the mirror is also a planting. | Fragments without a main verb read like captions under photographs. Sensory anchors: the light, the jawline, the shelf. Treating a serum as someone who keeps asking for patience is the freshest image in the opening. The sibling page has the same three slots filled with lab values. | rewrite-claim Keep the three-snapshot mirror, in her own words. Ours mirrors the shelf, the steps and the spend without the verdict that any of it failed. p6 |
| 8 5.1% | failed-solutions 4× p · 30 w | The voice switches to first person: a two-word verdict that nothing helps, her confusion, the sense of having done it all correctly without anything lasting, and a suspicion that she was simply born unlucky. | identificationmirroringopen-loop problem-aware | It follows three 'you' lines with no transition. The pronoun changes on the shortest line, so the reader is moved from her own mirror into the narrator's story without noticing the join. Self-blame has to be planted here to be removed at 19%. | In the capture the two-word verdict is bold. Line lengths step outward: 2, 5, 11, 12 words. The verb chosen for lasting is the one our own angle is built on. | rewrite-claim A narrator may confess fatigue, never a verdict that the routine does not work. Ours: she did everything right, and one layer was never covered. p6p3 |
| 9 6.3% | stakes 4× p · 67 w | She describes the isolation: others see a wrinkle, she sees the one thing her own profession could not solve on herself; the tiredness of standing at a mirror waiting for the next change; eleven years of it. | identificationfear-of-declinestory-transportspecificity problem-aware | It sits at the bottom of the descent, just before the hinge. The reader has to feel how long and how lonely it was for the next word to land. The eleven-year figure is planted here and returns four times down the page. | A mid-sentence ellipsis imitates speech. The professional irony (studied it, could not fix it) is the only place the story uses her credential. Eleven is an odd, believable number. | keep-structure The cost beat in her own words from the voice-of-customer file: what skincare fatigue does to a morning. No decline is dramatized. p6 |
| 10 7.7% | turning-point 2× p · 18 w | A one-word hinge, then the discovery: a crowd of 38,000 women stuck at the identical point after trying everything. | open-loopconsensussocial-proofstory-transport problem-aware | At 8%, straight after the lowest point. The deck promised this number; here it is paid for the first time and its meaning shifts, from subjects examined to women found. | A single word with an ellipsis stands as its own paragraph, a drum roll. The honest verb for what she did arrives here, some fifteen rows after the dishonest one in entry 5. | swap-proof Our crowd is real and sourced: the women who already take collagen and have never met astaxanthin. No invented thread, no member count. p7i1 |
| 11 8.4% | image img | An image slot (370 by 207) under the line about the crowd; by position it most likely shows the group or thread with its member count. | social-proof problem-aware | Placed after the number and before the scene of reading the thread, so the reader sees the place before she is walked through it. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. If it shows the thread, it is as unverifiable as the hero collage. | drop A screenshot of an invented community does not transfer. If a picture stands here it is a real, sourced one. |
| 12 9.5% | turning-point 2× p · 24 w | She goes into the thread braced for the three stock tips and finds instead thousands of women describing it in her own vocabulary. | mirroringconsensuspattern-interruptstory-transport problem-aware | The stock tips are named first so the reader can tick them off as things she has been told as well. Only then does the surprise land. It prepares the claim, two rows on, that identical words mean an identical cause. | The tips come as three clipped sentences of one, two and three words, the rhythm of someone counting on her fingers. The reader has just read the narrator's own wording in entries 7 to 9, so she can test the claim against the page itself. | keep-structure Her own words, quoted from real and named public sources in the voice-of-customer file, stand where the invented thread stood. p6 |
| 13 10.4% | image img | A second image slot of the same size; by position it most likely shows a run of comments from that thread. | social-proofconsensus problem-aware | It splits the claim about shared words in two, so the eye rests on evidence between the assertion and its interpretation. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. If it matches the hero collage, it is composed. | drop Invented comments do not transfer. |
| 14 11.5% | turning-point 4× p · 42 w | She stresses that routines and budgets differed while the wording was identical, says that had a cause, and teases a single numbered comment, #5,847, that accounted for every failure. | consensusreason-whyopen-loopspecificity problem-aware | The reader has accepted the crowd; now the crowd has to mean something. Ruling out routine and budget in advance removes the two explanations she would reach for. The numbered comment at the end is the hook that carries her across the first subhead. | A correction in two fragments insists on exactness, the first of the page's many contrast flips. The comment number is the most memorable specific on the page. Fabrication: the same number appears on the store's kidney page, on a different platform. | rewrite-claim The pattern-then-cause step transfers. Ours: most antioxidants she has tried guard only part of the cell, and that is the common thread. No numbered comment. p3p7 |
| 15 13.2% | turning-point h2 · 2× p · 34 w | The first subhead announces something baffling; the next two lines repeat, almost unchanged, the tease about the numbered comment that would account for each of her failures. | open-loopcuriosity-gap problem-aware | The page's first subhead arrives at 13%, exactly where the loop is about to pay. A skimmer landing on it gets the tease without the run-up. | The repetition is a seam, not a device: the line before the subhead and the lines after it are near duplicates, which shows the section was assembled from blocks. It costs a careful reader the feeling of having read this already. | keep-structure A subhead at the point of payoff, written once. Our builder refuses repeated sentences. |
| 16 14.4% | image img | The image slot that holds the comment itself. The lines after it show what it states: three things free radicals do to skin at once, namely oxidize it, tear down collagen, and dry out the barrier. | loop-closednew-mechanismsocial-proof problem-aware | The reader has waited since 12.7% for this. Showing it as a screenshot instead of typing it makes the claim arrive as found evidence and spares the narrator from asserting it in her own voice. | Fabrication: an anonymous comment is the page's entire source for its cause. The key claim exists only inside a picture with no alt text; with images off, the next line reacts to nothing. Not seen in our capture; its content is read off entries 17 and 18. | swap-proof The found artifact becomes a real one: the published trial's record, captured with its PMID verified (L-05), and the claim also set in text. b2b1 |
| 17 15.5% | turning-point 2× p · 23 w | She notes that the comment came with no reasoning and no sales motive, as if its author were weary of repeating it, and that she looked at it for twenty minutes. | objection-preemptstory-transportspecificity problem-aware | Directly after the artifact and before any interpretation: the reader is told how to feel about the source (no agenda) before she is told what it means. | Four fragments, then a simile that gives an anonymous commenter a temperament in a handful of words. The twenty minutes return in entry 73, where she hesitates over the button for the same span: a deliberate echo. | swap-proof Our disinterested source is the journal that published the trial, named. b1x3 |
| 18 16.2% | villain-reveal 3× p · 16 w | Three short questions restate the comment in her own voice: oxidation of the skin, collagen being torn, the moisture barrier drying out, all of it simultaneously. | new-mechanismcuriosity-gapidentification problem-aware | At 16% the cause is named in words for the first time. Phrased as incredulous questions, it lets the reader doubt alongside the narrator, so the research scene that follows answers her doubt too. | Question fragments of 5, 3 and 8 words. One cause with three effects is set up here and drives the numbered section at 24%. The tearing verb used for collagen appears three times on the page. | rewrite-claim One sentence from the banned attack framing (p2). Ours names what The Cell Shield guards, never what is destroying her collagen: ordinary antioxidants leave the inner membrane exposed; astaxanthin spans it. p3m2r1 |
| 19 17.1% | failed-solutions 3× p · 37 w | She tallies it: eleven years of adjustments, fourteen thousand dollars on topicals and two filler sessions, laser and surgery now under consideration; then asks whether neither her routine nor her discipline was ever at fault. | specificityanchoringloss-aversionvindication problem-aware | It stands between the claim and the research. The reader must feel what the wrong diagnosis cost before the right one is explained, and the sum sets the anchor that makes a bottle of softgels look small at 76%. | A dollar figure with its breakdown and an escalation ladder (serum, filler, laser, knife). The closing question joins two negatives with a conjunction and stumbles. The same sum appears on the sibling kidney page: a template number. It returns twice here, in entries 45 and 73. | rewrite-claim The tally is ours to use as her investment, the thing The Cell Shield protects, never as money wasted. One softgel a day replaces nothing in it. p6b9 |
| 20 18.4% | turning-point 2× p · 13 w | A time-stamped night scene: in the small hours she locates the science and says it altered everything. | open-loopstory-transportspecificity problem-aware | So far the cause rests on an anonymous comment. Before the explanation begins, the page needs the word research on the table. Placed at the end of a section, it is also the cliffhanger into the next subhead. | The hour is the only concrete thing in it. The second line is a stock phrase that could sit on any page, and does sit on the sibling. No study is named here or anywhere else on the page. | swap-proof At this slot we name the research: the human trials and the journals that published them. b1b2x3 |
| 21 18.9% | villain-reveal h2 · 16 w | The subhead absolves her: the cause was her body's biology, together with her own ignorance of how it operates. | vindicationrelief problem-aware | Absolution arrives before explanation, and as a heading, so she reads the science already forgiven. The third clause keeps back a small share of responsibility (not knowing), which is what makes acting on the knowledge the remedy. | Three sentences inside one heading; the last is loose in grammar and sounds spoken. It admits a limit, which reads as candor. Identical on the sibling page. | keep-structure Vindication is compatible with our frame and stronger in it: nothing she did was wrong and nothing she bought was wrong; one layer was never covered. p3p7 |
| 22 19.5% | image img | An image slot directly under the absolving subhead; by position it most likely diagrams skin under attack from below the surface. | new-mechanism problem-aware | A diagram ahead of sixty words of biology lets the reader take the paragraph as a caption. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | swap-mechanism Our one illustration of The Cell Shield: the molecule lying across the full thickness of the cell membrane. m2 |
| 23 20.7% | villain-reveal p · 60 w | The longest paragraph yet, spoken to the reader: skin is cast as a structure under attack from within by free radicals from five everyday sources; they harm its support proteins, lipid barrier and tone, producing the opening's three signs. | new-mechanismcommon-enemyreason-whymirroring problem-aware | First full explanation, at 21%, after the absolution. The three signs were mirrored at 4%, questioned at 16%, and now receive a cause, so the explanation is experienced as recognition. | The pronoun goes back to 'you' for the science, so the diagnosis is the reader's own. Five sources and three effects in one sentence: dense but parallel. Dashes fence the list of sources. | swap-mechanism Our cause for skin is oxidative stress from modern life and UV, and the gap ordinary antioxidants leave at the inner membrane. We describe what The Cell Shield guards; we do not narrate her collagen being destroyed. Never positioned as a replacement for sunscreen. r1p3m4 |
| 24 22.0% | failed-solutions 2× p · 27 w | Back in first person: her effort had all gone into the top layer as the harm went on beneath, and nothing she owned could stop it, since it began deeper than any of them goes. | reason-whycontrastvindication problem-aware | As soon as the villain is named, the page must explain why good products did not stop it. Depth is the answer, and depth becomes the test the ingredient passes at 48%. | The opposition of top and below is the page's master metaphor, introduced here in two plain sentences and returned to throughout. The pronoun goes back to 'I', so the reader's shelf is judged through the narrator's. | rewrite-claim This is the failed-routine verdict and it does not transfer. Ours: her routine does its work at its own layer, and The Cell Shield covers the layer under it. It replaces nothing and protects everything in it. b9m2p3 |
| 25 22.8% | dismissal-by-authority p · 43 w | Eleven years of retinol are summed up as polish on the top layer that never touched what was tearing at collagen below; her dermatologist praised the routine, rightly, and the aging went on. Verdict in two words. | vindicationcontrastreason-why problem-aware | The reader's likeliest reply to the depth theory is that her doctor would have told her. The page answers before she asks: the doctor saw a good routine and was right about it. This is an approval, not a shrug; the authority is wrong only about the target. | One sentence opens on a throat-clearing frame and a dash. The dermatologist gets one quoted line and no blame, and the narrator concedes the routine did look good, which keeps her likeable. The deck blamed the profession as a class; in person the doctor is treated gently. | rewrite-claim Doctors are allies on our pages and her routine is not judged. Ours: the routine her dermatologist approves does its work at its layer; The Cell Shield covers the membrane under it. b9m2 |
| 26 23.8% | mechanism h2 · 2× p · 60 w | A subhead promises three simultaneous effects. The first is labelled in capitals with its visible sign, then explained: free radicals take down the firming proteins quicker than they are replaced, the framework gives way, and retinol only works on top. | new-mechanismreason-whyspecificitycontrast problem-aware | The villain has been named in a paragraph; now each symptom gets its own numbered proof so the reader finds her worst one in a heading. Sagging goes first because it is the most feared and the least answered by anything in a jar. | The label format is cause, dash, visible sign, in capitals: scannable. A building metaphor carries the explanation. The item closes on a two-sentence refrain (the product on top, the cause below) that returns at the end of the next item and twice in the verdict list: a chorus that is mechanical by its fourth use. | rewrite-claim This is where the page leans hardest on a banned framing: breakdown outrunning rebuild (p2). Ours is the solution side of the same biology: astaxanthin quiets the enzymes that eat collagen, 90% and more in the human work, so the collagen she builds and buys holds. m1x1b2 |
| 27 25.9% | mechanism 2× p · 66 w | Second item: the same oxidation removes the lipid layer that seals water in, so moisture escapes and skin turns dry and crepe-like however much cream goes on; cream works from outside, the cause from within. | new-mechanismreason-whymirroringcontrast problem-aware | Second because dryness is the symptom she fights daily; after structure comes feel. It reuses the opening's own complaint as its label, so the reader meets her words again as a diagnosis. | At 57 words one of the longest paragraphs, carried by a clear chain of cause and effect. It ends on the refrain again. | rewrite-claim We may say hydration improved against placebo in the human trials. We may not say her moisturizer is futile. b1 |
| 28 27.5% | mechanism 2× p · 67 w | Third item: oxidation is said to drain the cell itself; a protected cell gives off a warm, even light, a stressed one turns flat and grey, which makeup cannot imitate because the dullness sits within the cell. | new-mechanismreason-whycontrasthope problem-aware | Last of the three because it is the softest claim and the one the story will show first (weeks three to four). It also plants the idea of radiance that the timeline and the guarantee pay out. | The least physical of the three explanations: light from a cell is an image, not a process. A dash-heavy paragraph. Its final sentence ties all three items together, which makes the recap that follows partly redundant. | rewrite-claim The Astaxanthin Glow is ours to use since round 2, and the capillary claim gives it a reason. No talk of cells being drained. b5m5 |
| 29 29.2% | mechanism p · 32 w | A recap in four fragments: three signs, a single oxidative cause, all worsening together, none answering to separate fixes because no fix touched the source. | reason-whycontrastloop-closed problem-aware | It closes the three-item section and states the rule (one source) that the next section applies product by product. | Stacked fragments of identical cadence. Its nouns are laboratory words left over from the kidney template, where they are blood values; the register slips from mirror to clinic. | keep-structure One softgel in place of juggling; one shield under everything she already uses. r23m2 |
| 30 30.0% | failed-solutions h2 · 6 w | A subhead announces the list of what failed, with reasons to follow. | open-loopreason-why problem-aware | After the rule comes its application. The heading tells her the next block is about her own shelf. | Five words and a dash. Identical on the sibling page. | rewrite-claim Our heading for this block is what each thing does, never what failed. |
| 31 30.4% | image img | An image slot under that subhead; by position it most likely shows the products about to be judged. | mirroring problem-aware | Seen before the three verdicts, a picture of familiar jars makes each verdict personal. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | keep-structure A real photograph of a routine on a shelf, the softgel beside it and not in place of it. b9 |
| 32 31.6% | failed-solutions 3× p · 88 w | Three verdicts of one shape, each granting what the product does on top, then what continues below: retinol resurfaces an unprotected structure; vitamin C serum degrades in its bottle and stays superficial; moisturizer adds water a damaged barrier releases. | reason-whycontrastcommon-enemy problem-aware | The rule is applied to named products so that she does the same with her own shelf. It precedes the narrator's supplement attempt, so that taking something by mouth is the only door left open. | Each verdict concedes a real effect first, which sounds fair. The second and third close on a near-identical refrain that uses the word mechanism for the cause: abstract, and repeated within two paragraphs. The wall-and-foundation picture in the first verdict is the keeper. | rewrite-claim What each does, never an attack: topicals work at the surface, collagen builds, The Cell Shield keeps what she built. The word mechanism never appears in our copy (CL-02). m1b9p3 |
| 33 33.8% | failed-solutions h2 · 3× p · 36 w | Subhead: she went shopping for the fix and it was not for sale. Next morning she orders every antioxidant pill on offer, names three kinds, and gives the bill: a hundred dollars. | story-transportspecificityobjection-preempt problem-aware | The logic so far leads straight to any antioxidant pill, and a reader could leave here to buy vitamin E. The page spends the next 3% closing that exit before it introduces the right kind. | A hundred dollars beside fourteen thousand makes the attempt casual and the failure cheap. The subhead's claim is contradicted by the page itself 25% later, when the thing turns out to be for sale. | keep-structure Close the generic-antioxidant exit with the gate's own facts: vitamin C works outside the cell, vitamin E on the membrane surface, a multivitamin is underdosed. p3r3r5 |
| 34 35.6% | turning-point 4× p · 25 w | She admits the pills did little, doubts her whole theory, goes back to the science, and announces that this time she found the thing that works. | candoropen-loopidentification problem-aware | A doubt beat at 36%, after the theory and before the specification. The reader who thinks antioxidants are hype hears the narrator say so first. | A slip into present tense reads like a diary line. The last sentence is a stock bridge. Three of the four lines are word for word on the sibling page, the fourth nearly. | keep-structure One honest doubt before the specification, in the narrator's own words. |
| 35 36.8% | image img | An image slot after the announcement; by position it most likely shows the research she says she found. | borrowed-authority problem-aware | It stands between the promise of an answer and the specification, where a reader expects to be shown the paper. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | swap-proof The proof object belongs here: the real record of the human trial, captured, PMID verified (L-05). b1b2 |
| 36 38.0% | mechanism 2× p · 72 w | The specification comes before the name: an antioxidant that dissolves in fat, at a clinical dose, so it can lodge in the membranes of skin cells and remain; water-soluble or token-dose products pass through, hence the useless shop pills. | new-mechanismreason-whycontrastcuriosity-gap problem-aware | Criteria first. At 38% she learns what to look for, so the ingredient at 46% and the brand at 59% each arrive as the answer to a test she has already accepted. | The dose is called clinical and exact with no number attached; the number is held back until 54%. One adjective for a token pill is unidiomatic. Two paragraphs of pure explanation with no person in them. | swap-mechanism The Cell Shield: the molecule spans the whole membrane, where vitamins stay in the bloodstream or on the surface. 12 mg against the 4 to 6 mg brands. m2r5p3b7 |
| 37 39.8% | proof-expert 2× p · 25 w | She says she needed the genuine article, that a researcher helped, and that she found a laboratory that standardizes correctly and reviewed what it does. | borrowed-authorityopen-loop solution-aware | The criteria are set, and someone other than the narrator now has to vouch. These two lines are the hinge between her reasoning and the expert's story. | The weakest lines on the page: no name, no institution, no place, and one sentence without a subject. Fabrication by omission: an expert nobody could look up. A stub left from the kidney template, where the narrator drives to a laboratory in Ohio and watches the work. | swap-proof Our sources have names: natural Haematococcus pluvialis, closed-bioreactor cultivation, a third-party certificate for every batch behind a QR code. c3c4c5 |
| 38 40.7% | image img | A taller image slot (370 by 277); by position it most likely shows the laboratory or the researcher. | borrowed-authority solution-aware | It comes just before the expert speaks, so her words arrive with a picture attached. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | swap-proof A real photograph from our own supply chain, or the certificate itself. c3c5 |
| 39 42.2% | proof-expert 3× p · 53 w | The researcher reframes aging with a question about why women in Japan in their seventies and eighties so often look far younger; the narrator offers the stock answer, the diet of fish, tea and seaweed. | curiosity-gapborrowed-authorityopen-loopstory-transport solution-aware | Cause and criteria are settled; what the page needs now is desire. A population that visibly beat the problem supplies it, and the question restarts curiosity at 42%, the stretch where explanation loses readers. | Reported dialogue, not quoted. The wrong answer is framed as the common one, which folds the reader in. A national generalization stands in for evidence, from a speaker with no name. | rewrite-claim The gate holds nothing about Japan or any population. The second-mystery device can run on the women in the human trials. Missing claim: any population that already takes it. b1p7 |
| 40 43.6% | proof-expert p · 58 w | The researcher grants that diet explains the average, then narrows to the outliers the narrator really has in mind: a woman of eighty taken for a sibling of her granddaughter, the cases that defy explanation. | contrastcuriosity-gapspecificityopen-loop solution-aware | It closes the diet exit the way the pills exit was closed at 35%: concede, then narrow. It must come before the ingredient, so the ingredient explains the outliers and not the average. | The master metaphor is reused for an argument: an answer can be superficial too. The granddaughter picture foreshadows the narrator's own payoff at 67%. The second person here is the researcher addressing the narrator, and reads as the page addressing the reader. | rewrite-claim Concede, then narrow, with our content: antioxidants in food and in a multivitamin are real, and one of them is built to span the membrane. m2m3p3 |
| 41 44.9% | mechanism 2× p · 39 w | The answer comes in two lines: those women take it and do not merely eat it; a single concentrated antioxidant each morning, the only one with the potency and the fat solubility to accumulate in skin. | open-loopnew-mechanismexclusivityease solution-aware | Last beat before the name. It restates the two criteria from 38% (power, fat solubility), so the name lands as their solution. | Singularity is hammered three ways in two lines. The second sentence is long and built around a dash, and it climbs. | swap-mechanism The Cell Shield, one softgel a day. b9m2m3 |
| 42 46.0% | mechanism h2 · p · 52 w | A one-word subhead names the ingredient. The paragraph makes it ordinary: on sale all over Japan, advised by skin doctors there ahead of retinol, taken by women since the 1980s, as routine as dental care. | loop-closedconsensussocial-proofborrowed-authority solution-aware | At 46%, close to the midpoint: late enough that cause, criteria and desire are all in place, early enough to leave half the page for tests, story and terms. The normalizing follows the name at once, because a new word raises the question of why she has never heard of it. | The one-word heading is the strongest typographic moment on the page. Every claim about Japanese shops and doctors is unsourced. The comparison to a daily hygiene habit files it as routine, not treatment. | rewrite-claim The name lands the same way, in a heading of its own. Our normalizer is true: most of this market has never met astaxanthin, and it has more than 60 clinical studies behind it. No Japan claims. p7r14r15 |
| 43 47.5% | mechanism p · 53 w | Origin and power: the pigment behind pink salmon and flamingos, made by ocean microalgae; rated 6,000 times vitamin C and 550 times vitamin E; and fat-soluble, which the page offers as its full explanation for working where topicals cannot. | specificitycontrastreason-whynew-mechanism solution-aware | After the name and its social standing she needs to see it and size it. The property, which is the real argument, comes last in the paragraph so that it is read with the momentum of the numbers. | The multipliers are category standard and are given without what they measure. Spacing in the rows shows two phrases set apart inline, the first multiplier and the solubility. The closing clause claims totality. | swap-proof The gate holds both multipliers with their unit: singlet-oxygen quenching. Natural, from microalgae. m3c4 |
| 44 48.6% | mechanism p · 48 w | How it works, in one picture: cell membranes are fat, so the molecule lodges in them and does not rinse away, accumulates over weeks, and stops free radicals short of collagen and the barrier. | new-mechanismreason-whyfuture-pacing solution-aware | It follows the numbers so that power gets a location. The weeks are planted here, 14% ahead of the timeline, so the slow first days will read as predicted. | A sentinel image, close to our own noun: the category converges on guarding. Present tense, no hedges. The tearing verb returns. | swap-mechanism The Cell Shield in its own words: the molecule spans the full thickness of the membrane and quiets the enzymes that eat collagen. m2m1x1 |
| 45 49.6% | future-pace p · 38 w | A vignette sets a woman in Osaka, water on her face and a single softgel, looking a quarter-century younger than her documents say, against 'us' and the fourteen-thousand-dollar pursuit from outside. | contrasteaseanchoringspecificity solution-aware | It caps the explanation with a person, ahead of the narrator's own summary. The large sum returns at the midpoint, so the anchor is fresh when the buying tests begin. | A named city and a travel document as the measure of age: concrete and witty. The plural pronoun includes the reader in the spending. Fabrication: an invented woman and a number nobody could check. | rewrite-claim No years-younger claim and no invented woman. Ours is the ease alone: one softgel a day that replaces nothing on her shelf. b9 |
| 46 50.5% | villain-reveal p · 35 w | Her summary: eleven years explained; the fault lay in reach and never in routine; her products all acted on top, and this was the only one that went in. | vindicationcontrastloop-closedreason-why solution-aware | End of the teaching half, at 50%. Everything after it is buying. She crosses the midpoint with one sentence to remember. | The page's signature contrast flip, used here at its best because it names the idea. Three short declaratives in a row. | rewrite-claim Ours keeps the name-the-idea move on the solution side: her routine reaches its layer; The Cell Shield reaches the membrane. m2b9 |
| 47 51.3% | differentiation h2 · 2× p · 52 w | A second shopping trip under the same subhead formula as the first: she drives to a vitamin shop, takes the first bottle of astaxanthin with pride, then reads its label and announces three quiet ways sellers shortchange buyers. | story-transportcommon-enemyopen-loop solution-aware | Straight after the ingredient is sold, because that is when a reader opens a new tab to price it. The page gets there first. | The subhead recycles the formula of entry 33 almost word for word; it can pass as parallelism, but it is a reused block. Pride before the fall is a human beat. | keep-structure Our aisle is documented: about 64% of unverified astaxanthin products fail third-party potency testing, and the category leader stopped publishing its certificates. c1c2 |
| 48 53.0% | differentiation 3× p · 116 w | Three numbered faults: the astaxanthin on American shelves is mostly synthetic, made from petroleum; the bottle held 4 milligrams against an effective 12; and a dry capsule lacks the oil the molecule needs for absorption. | common-enemyspecificitycontrastloss-aversion solution-aware | The tests precede the brand by 6%, so the brand's checklist at 59% reads as a scorecard already agreed. | Each item opens on a verdict of three or four words, gives one number, then a consequence. Two crude comparisons (an engine fluid, a bathroom fixture) are the memorable parts. The 95% figure has no source. | rewrite-claim Two tests transfer on gate claims: natural versus petrochemical, and 12 mg versus the 4 to 6 mg brands. The third rests on carrier-oil superiority, which is banned (b8); ours is the potency test and the published certificate. r16b7c1c3c4 |
| 49 55.6% | differentiation 2× p · 28 w | She reports testing the market: bottles from three shops plus Amazon, each failing one or more tests and most failing every one; she nearly quits. | specificityconsensusopen-loop solution-aware | Last beat before the brand. The low makes the brand's entrance a relief, and explains why the narrator did not find it herself. | The arithmetic wobbles: three bottles, three shops and a marketplace as well. The closing line is stock. | keep-structure The shelf result, stated as the published failure rate, not as a private test. c1 |
| 50 56.4% | product-reveal h2 · 9 w | A subhead with a weekday and an hour credits her daughter with the solution. | specificitystory-transportopen-loop solution-aware | The product's entrance is credited to a family member, which insulates the reveal from the suspicion of selling. | The day and hour are kept from the template; the sibling page carries the same stamp with a different clause after it. Spacing in the row shows the last words set apart inline. | rewrite-claim A time stamp only if the story is true or disclosed as a composite (CL-14). |
| 51 56.9% | image img | An image slot (370 by 247) under that subhead; by position it most likely shows the daughter, the bottle, or both. | story-transport solution-aware | A photograph ahead of the name lets the bottle be seen as a gift before it is read as a brand. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | keep-structure Our reveal image is the softgel and the pack with its callouts. c3 |
| 52 58.2% | product-reveal 2× p · 38 w | The daughter, cast as the family's researcher, listens for a month, spends two days comparing every option against the Japanese standard, and returns with a single bottle. | story-transportborrowed-authorityease solution-aware | Just before the brand name. A daughter's recommendation is the most trusted channel this reader has, and it contradicts nothing she believes about advertising. | A contradiction: the byline sells a researcher of thirty years, and here the daughter holds that role in the family. The motif was planted at 1%, where a commenter in the hero collage orders because her daughter insisted. | swap-proof The comparing is done in public, by third-party tests we publish for every batch, not by an invented relative. Our reader spans 35 to 65, so she may be the daughter as easily as the mother. c3c1 |
| 53 59.4% | product-reveal 4× p · 42 w | The product is named for the first time, Rovina Natural Astaxanthin, followed by ticked lines that answer the three tests in order: 12 mg, microalgae source, coconut-oil softgel, each with a bracketed jab at the failing alternative. | loop-closedspecificitycontrastreason-why solution-aware | At 59%. She holds the cause (21%), the criteria (38%), the ingredient (46%) and the tests (53%). The brand comes last, as the single product that meets all of them, laid out as a specification and not as a pitch. | Label style: a field name, ticks, a repeated one-word confirmation. The bracketed asides carry the voice. The store link behind the buttons gives the product a different name (Sunglow), a small inconsistency. | rewrite-claim Our scorecard: CellScience Astaxanthin 12 mg, softgel, natural Haematococcus pluvialis, closed bioreactor, a certificate for every batch. The carrier-oil tick is dropped (b8), and so is any talk of the bag (c6). b7c4c5c3r16 |
| 54 60.8% | product-reveal 3× p · 28 w | Two more ticks (independent testing; non-GMO and American-made), then a line positioning the bottle as the nearest American equivalent of what a Tokyo pharmacy stocks. | borrowed-authorityspecificityloop-closed product-aware | The three tests are passed; the extra ticks answer the quieter question of safety. The closing line joins the brand to the desire built at 42 to 50%. | The ticks thin out to bare labels with no asides. The comparison is hedged (nearest, not the same), a rare modest claim. | swap-proof Our badges as the gate lists them: cGMP, made in USA, gluten-free, no stimulants, and the certificate behind the QR code. r18c3 |
| 55 61.8% | how-to-use p · 14 w | She begins the same morning: a single softgel and a water rinse, nothing else. | easecommitmentloop-closed product-aware | Between reveal and timeline. It repeats the Osaka vignette (water, one softgel), so the narrator steps into the life the reader was shown at 50%. | Four fragments. The callback is exact and unannounced. It implies she put her routine aside, which works against a reader who means to keep hers. | rewrite-claim One softgel a day with a meal that has some fat. Ours must never suggest she drops her routine: The Cell Shield is what makes it hold. b9r31 |
| 56 62.2% | results-timeline 2× h2 · 2× p · 61 w | The timeline opens on an admission: two days in, almost nothing, and she nearly doubts, recalling that it has to accumulate. By days three to seven dryness eases; a forgotten night cream and skin still soft next morning. | candorspecificityreason-whyfuture-pacing product-aware | Directly after she starts. The admission of nothing comes first so that a buyer who feels nothing in her first days keeps going instead of asking for a refund. The accumulation idea planted at 48% is the excuse. | The admission is the most trust-building line in the timeline. The accidental skipped cream works as an unplanned experiment, which persuades more than a claimed result would. | rewrite-claim The honest slow start transfers. The dates do not: our skin results are those of the human trials, measured over weeks. Missing claim: any skin change in the first seven days. b1 |
| 57 64.1% | results-timeline 2× h2 · 2× p · 58 w | Weeks three to four: the daughter notices first and remarks on radiance; the tone once faked with makeup is simply present. Week six: the mirror no longer makes her wince, the grey has gone, cheeks look fuller, eyes feel clearer. | social-proofspecificityloop-closedhope product-aware | The tone symptom, third in the cause section, improves first in the story, and someone else reports it before she does, which guards against the charge of wishful seeing. The dread of the mirror from the opening is resolved here. | One quoted line from the daughter, the only speech she gets. A stray benefit (eyes) slips in from the ingredient's other uses. | rewrite-claim The Astaxanthin Glow and the measured trial outcomes; a reviewer's voice only when the review is real. b5b1 |
| 58 66.0% | results-timeline h2 · p · 33 w | Weeks eight to ten: the slackness at her jaw is firmer without having vanished, soft and tight in the right places, and the tired look has lifted. | candorspecificityhope product-aware | Sagging was the first symptom explained and is the last to improve: the hardest promise is held for the point where trust is highest, and is stated as partial. | The concession at the start of the first sentence is what makes the rest credible. One balanced sentence of paired opposites is more written than spoken. | rewrite-claim Elasticity improved against placebo in the human trials; we say that, in those terms. b1 |
| 59 67.0% | results-timeline h2 · 3× p · 41 w | Under a week label repeated out of sequence, a stranger at a market takes mother and daughter for sisters; she gives both ages, 60 and 38, lets the mistake stand, and takes her softgel the following day as ever. | social-proofstory-transportspecificityloop-closed product-aware | Last item of the timeline. It cashes the picture the researcher planted at 44%. A third party is needed because the narrator's own mirror proves nothing. | A heading error: the label goes back to week six after weeks eight to ten, and the closing letter dates the same scene at twelve weeks. The two-sentence line of ages is the best rhythm on the page. Fabrication: an anecdote nobody could verify, carrying the page's largest implied claim. | rewrite-claim The move transfers (someone else notices); the claim of looking 22 years younger does not. Ours waits for a true customer remark from the panel. Missing claim: any years-younger outcome. r17b1 |
| 60 68.5% | image img | An image slot after the market scene; by position it most likely shows mother and daughter, or a before and after. | social-proof product-aware | It follows the peak, where the reader most wants to see her. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. If it is a before and after, it is one nobody could verify. | drop No before-and-after imagery: a policy risk the angle file names. The softgel and its colour carry the visual. |
| 61 69.7% | future-pace h2 · 3× p · 102 w | The narrator turns to the reader: the hope-and-wait routine can end; age is no bar and nothing is beyond repair; once cells are guarded from within they recover, with collagen, barrier and radiance returning; aging is halted and undone. | hopefuture-pacingobjection-preemptvindication product-aware | Directly after the story's peak, while the reader is moved. Too late is the first objection of a sixty-year-old, so it is answered before any terms. | The subhead quotes the patience speech from entry 7, a loop closed on a phrase. Three short sentences in a row promise outcomes. The claim climbs from protecting to undoing, the strongest and least defensible on the page, and it contradicts the partial result admitted in entry 58. | rewrite-claim We promise what the trials measured: wrinkle depth, elasticity, hydration. Never reversal, never collagen rebuilding; ours holds what she builds. Our reader spans 35 to 65, so the objection is too early as often as too late. Missing claim: any reversal of damage. b1m1r24 |
| 62 72.5% | image img | An image slot between the promise and the two paths; its content cannot be inferred from the lines around it. | none product-aware | A visual pause marks the border between story and selling. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | keep-structure A product photograph at the border between story and close. |
| 63 73.6% | close h2 · 2× p · 103 w | A two-path close. One: more top-layer serums, the same mirror, and half a year on the same face with several hundred dollars gone. Two: try what reaches the right depth, allow 8 to 12 weeks, learn the Japanese habit. | contrastloss-aversionfuture-pacingcommitment product-aware | After the promise and before the terms: she decides in principle first and hears the deal after. A half-year horizon for doing nothing is set against a shorter one for trying. | Path one repeats one opening verb four times. The wall-and-foundation picture returns. Path two is worded as curiosity, not commitment. The count in the heading is a false binary. | rewrite-claim Our two paths cannot be her serums against our softgel. Ours: the same routine with and without The Cell Shield under it, and the weeks the trials took. b9m1b1 |
| 64 76.2% | offer h2 · p · 41 w | A subhead promises the order steps; the paragraph instead states the deal: a second bottle free with the first, justified as the supply needed for it to have time to work. | reason-whyanchoringease product-aware | At 76%, the first commercial terms, after the choice is made in principle. The reason (it needs time) converts the slow start into an argument for the larger quantity. | No price appears anywhere on the page. The heading is inherited from the sibling, where a step-by-step list follows; here nothing of the kind follows, a promise broken at heading level. A distancing phrase keeps the narrator outside the company. | swap-offer BRAND.md §3: 1 bag $29.99, 3 bags $59.97 (buy 2, get 1 free), 5 bags $89.95 (buy 3, get 2 free, the lowest price ever), free shipping, three free guides. On a prelander the tiers live on the PDP; the page names the deal and sends her there. x4r20 |
| 65 77.3% | guarantee p · 66 w | A refund promise running 90 days, with three named outcomes as the test, returnable even as an empty bottle; it ends by telling the reader she has come this far and has only another identical year to lose. | risk-reversalcommitmentloss-aversion most-aware | Immediately after the terms and before any button: the risk is removed ahead of the first ask. | The conditions are sensory (how her skin feels and looks), not procedural. The remark about having read this far is a consistency nudge. At 66 words, the densest selling paragraph. | swap-offer Ours is the 30-Day Money-Back Guarantee, by that name and that number everywhere (BRAND.md §3, CL-09): 30 days, full refund, keep the bag. Never 90. |
| 66 78.7% | objection h2 · 8 w | A subhead says the product is not sold on Amazon and that production runs in small batches, in the first person plural. | exclusivityscarcityobjection-preempt most-aware | It follows the guarantee because the next thought of a persuaded reader is where to buy. | A pronoun slip: the narrator has been a customer throughout, and this heading speaks as the maker. The founder voice of the sibling page leaks through. | keep-structure Sold on the official site only, said in the brand's voice and not the narrator's. r29 |
| 67 79.2% | image img | An image slot under that subhead; by position it most likely shows production or the bottle. | exclusivity most-aware | It sits between the claim of scarcity and its explanation. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | swap-proof A real photograph of cultivation or of the certificate. c5c3 |
| 68 80.3% | objection p · 33 w | The paragraph sets marketplace products (synthetic, weak, petroleum-derived) against the brand's version, and says it sells direct only to guarantee authenticity: a single source and no fakes. | common-enemyexclusivityreason-whyobjection-preempt most-aware | It restates the three tests one row before the first button, so the last thing read before the ask is why no other bottle will do. | Three disparaging modifiers pile onto one dismissive noun. It ends on three short noun phrases, a triad that carries the position. | rewrite-claim Official site only, natural not petrochemical, 12 mg. No mention of the carrier oil. r29r16b7 |
| 69 81.2% | cta cta · 7 w | The first button, at 81% of the page: it names the brand, the ingredient and the 90-day guarantee, and links to the product page. | risk-reversalcommitment most-aware | The page withholds every button until cause, ingredient, brand, story, terms and guarantee are all delivered. It is late: four fifths of the page pass without a way to act. | Six words, with the guarantee doing the persuading. No price and no quantity on the label. | keep-structure Our button goes to the PDP with its tracking and package parameters kept (L-06), and carries the 30-day guarantee. We place an earlier one after the scorecard. |
| 70 81.5% | urgency 4× p · 49 w | A postscript-style turn: the thread from the opening gains 200 women weekly, all searching for what the reader now has; then the single catch, that stock runs out and latecomers wait. | scarcitysocial-proofloop-closedurgency most-aware | It follows the first button, for the reader who did not click. The thread returns as a crowd competing for stock, and the reader is told she is ahead of them. | A two-word sentence rewards her for reading. Fabrication: here the thread lives on Facebook; on the sibling page the same numbered comment lives on Reddit. The scarcity has no count and no date. | drop An invented thread and an unevidenced stock-out do not transfer. Our only urgency is a true one: the launch-week code while it runs. |
| 71 83.1% | urgency 2× p · 107 w | She relays the company's answer on why it does not produce more: algae-grown astaxanthin resists mass production, every batch is verified, and they prefer delay to shipping a weak one. If it shows as available, that is the opening. | reason-whyscarcityborrowed-authority most-aware | Scarcity without a reason reads as a trick. The reason is supplied in the very next paragraph, in the company's quoted words, so the narrator stays a witness. | At 96 words the longest paragraph on the page, at the point where speed matters most. It sags. It holds the only direct quotation from the brand, and nobody could verify it. | rewrite-claim Our quality reason is documented and needs no stock-out: closed-bioreactor cultivation and a third-party certificate for every batch. c5c3 |
| 72 85.3% | image img | A taller image slot before the closing letter; by position it most likely shows the bottle. | scarcity most-aware | It divides the urgency block from the letter. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | keep-structure A pack shot. |
| 73 86.8% | objection h2 · 2× p · quote · 56 w | A closing letter starts. 127 days earlier she hovered over this button for twenty minutes; a quoted thought says she always falls for things and cannot take another letdown; then the sum spent and her dermatologist's laser suggestions. | identificationmirroringobjection-preemptspecificity most-aware | After terms, guarantee and scarcity, what remains is neither price nor logic but self-protection. The page addresses it only now because it only becomes live with her finger near the button. | The twenty minutes echo entry 17. The day count is awkward and credible, yet it comes to about eighteen weeks, which fits none of the dates in the timeline. The dermatologist returns as the one urging laser, the first time the authority figure is used for pressure. | keep-structure The fear of one more disappointment is the true centre of skincare fatigue. We answer it with the guarantee and with proof, and her doctor stays an ally. p6 |
| 74 88.7% | close 2× p · 22 w | She sets a second thought beside the fear, that she cannot go on as before, and clicks anyway with shaking hands, sure she is making an error. | commitmentstory-transportidentification most-aware | The reader does not need to be convinced, only to act unconvinced. The narrator shows that doubt and purchase can sit together. | Fragments. One adverb pairing is a wrong note that no speaker would produce. | keep-structure Permission to try while unsure, backed by the 30-Day Money-Back Guarantee. |
| 75 89.4% | results-timeline 3× p · 134 w | The results retold in three dated scenes: day seven, the skipped moisturizer; week six, her own image in a store window at her skin's worst hour, and a text to her daughter; twelve weeks, the sisters remark. | specificitystory-transportsocial-prooffuture-pacing most-aware | Many readers scroll straight to the bottom. The letter is a page within the page: objection, decision, results, invitation. | The store-window scene is new and is the best-observed moment on the page, because it names the hour of day her skin looked worst. The rest repeats entries 56 to 59 beat for beat, and its dates conflict with the first telling. | rewrite-claim A short recap for the bottom-scroller, in trial terms and the Glow; one felt line she can say. b1b5 |
| 76 92.1% | close 4× p · 82 w | She lists quiet gains: no dread at the mirror, no planning the next purchase, no three-front fight; then meets the reader's suspicion that she was merely lucky, admits she shared it, and now sees a face she had forgotten. | reliefidentificationobjection-preemptfuture-pacing most-aware | After the results, the last objection standing is that the narrator is an exception. It is the last one answered. | Three fragments define relief by what is absent. The luck objection is put in the reader's own words. Largely shared with the sibling page. | keep-structure Relief from the juggling: one softgel in place of the next new thing, and a routine she can stop second-guessing. p6r23 |
| 77 94.3% | close 2× p · 9 w | A one-line sign-off tells the reader the narrator will be there whichever way she decides, then a first-name signature. | easeidentification most-aware | After two closes and a scarcity block, a line without pressure lowers resistance right above the button. | Seven words and a dash before the name: a letter's ending on a web page. The line is identical on the sibling, under a different name. | keep-structure A signed ending only under a disclosed composite or a real person (CL-14). |
| 78 94.9% | cta cta · 3 w | The second button, at 95%, invites her to check whether stock is available, in capitals with an arrow. | commitmentscarcitycuriosity-gap most-aware | It follows the letter, and its label depends on the stock-out claim made 12% earlier. | Two words and an arrow. The label names an action with no cost, which is why it is easy to press. | rewrite-claim A small-yes label that is true for us: see the offer on the official site. No availability claim. r29 |
| 79 95.3% | urgency p · 23 w | An update line reports a surge in demand, stock moving quickly, and the two-for-one deal running while supplies last. | scarcityurgencysocial-proof most-aware | It sits under the second button, where a news site puts a late addition, and reads as fresher than the article. | A capitalized label makes it look like an editor's note. It has no date, which an update needs. | drop No undated demand claims. |
| 80 96.0% | image img | A small square image slot (210 by 210) above the badge lines; by size and position it is most likely a guarantee seal. | risk-reversal most-aware | It heads the trust block at the foot of the page. | No alt text. Content inferred from position, not seen: our local capture ends at 7% of the page. | swap-offer Our own 30-day seal. |
| 81 97.9% | guarantee 4× p · 17 w | Four badge lines: the 90-day refund promise, quick shipping at no charge, a secure checkout, and simple returns. | risk-reversalease most-aware | It sits at the foot, between the second and third buttons, for the reader who is deciding on logistics alone. | Template strings, largely the same on the sibling page. | swap-offer 30-Day Money-Back Guarantee, free shipping on every tier, and the gate's badges. r18 |
| 82 99.2% | cta cta · 5 w | The third button, at 99%, frames the purchase as beginning a 90-day trial. | risk-reversalcommitment most-aware | It is the last ask. The three button labels escalate in a line: get the product, check the stock, begin the trial. | Four words and an arrow. The possessive makes the trial hers. | swap-offer Same recasting with our number: a 30-day trial under the 30-Day Money-Back Guarantee. |
| 83 99.7% | page-furniture p · 9 w | The offer bar from the top of the page, repeated as the last line. | exclusivityurgency most-aware | It is the last thing seen by a reader who scrolls to the end. | Identical to entry 1. | page-furniture As entry 1. |