# CHL editorial r1 — rejection audit

## Disposition

**OPERATOR REJECTED.** Earlier internal `ready` and `SHIP` language documented implementation completion and self-review. It did not establish aesthetic acceptance by the operator. This archive diagnoses the rejected page; it does not redesign it, score it, claim conversion lift, or change Shopify.

## Root conclusion

Editorial r1 failed at the visual-concept level. Oversized serif headlines, cream and green bands, staged bottle photographs, and repeated split layouts form a familiar wellness template. They do not build an ownable Clean Health Lab world. The page gives too little visual information specific to this product, repeats the same product-display job in the hero and gallery, isolates people deep in the journey, and promises proof that resolves to facts plus an email request rather than available test evidence.

The problem is not the mere use of green, cream, serif type, or whitespace. It is the full composition: too few distinct visual ideas, too much space assigned to each, and too little brand-specific information carried through them.

## What the three independent judges actually said

Three isolated judges reviewed fresh captures without seeing prior verdicts or one another's reports. All three expressed favorable aesthetic opinions. Those opinions must remain in the record; this was not a unanimous aesthetic rejection. Their narrow comparisons also did not validate the operator's brief or replace operator acceptance.

| Lane | Positive judgment | Limiting judgment | Recorded verdict |
| --- | --- | --- | --- |
| Visual identity | Preferred the candidate's coherence, photography, typography, and editorial rhythm to v61. | Called the premium-wellness language interchangeable, the composition repetitive, and the human scene isolated. | Strong visual improvement, still insufficiently ownable. |
| Trust and persuasion | Preferred the readable label, transparent offer terms, restraint, and high-dose disclosure. | Found company, formulation, and dose authority thin; the 2,600-review count was not independently audited. | Credible presentation with material trust gaps. |
| Commerce and mobile | Preferred the aesthetics and learning arc. | Found purchase choices too late, human proof too deep, and the mobile path repetitive before the offer. | Revise before acceptance. |

Their standards were too internal and too narrow for the operator's question. They compared the candidate with the previous build and a small reference set, then treated relative improvement and implementation finish as evidence of success. That did not answer whether the concept itself was ownable, human, product-specific, or supported by strong proof.

## Ranked findings

1. **The core visual concept is anonymous for the brief.** The giant serif, green/cream blocks, staged bottles, and split photo/text modules could be reassigned to another premium wellness brand with minor substitutions. The bottle's iridescent circle never grows into a wider CHL visual language.
2. **The image sequence repeats jobs rather than building an argument.** The hero sells a staged bottle scene, then a separate gallery presents the bottle again. Later sections return to oversized headings and alternating image/information blocks. Product mechanism, ingredient behavior, use, provenance, and human context receive little distinct visual explanation.
3. **Space is allocated to repetition.** Formula, proof, routine, and reassurance ideas each receive large display areas while the useful content within them stays thin. The issue is the ratio of space to information, not whitespace by itself.
4. **Human evidence is isolated and late.** Fresh capture places routines around 5,759 px on desktop and 7,231 px on mobile—about six desktop screens and more than eight mobile screens down at the tested viewports. Three of five cards appear initially; two sit behind a reveal. On desktop, the five-column area leaves a large empty region. Capture geometry varies, but the late placement and 3-of-5 initial state do not.
5. **The proof promise is stronger than the available proof.** “Inspect the proof” provides facts, a readable label, source links, and an email request for testing records. It does not provide an inspectable test report. The label supports ingredient transparency; it does not establish company authority, formulation rationale, product effectiveness, or the review count. Do not invent doctors, tests, reviews, or certifications.
6. **Purchase hierarchy regressed.** V61's desktop decision region brought product, dose, a creator cue, and offers together. Editorial r1 places choices below the first 900 px desktop viewport. On mobile, a hero bottle leads into another gallery before choices begin; one independent capture measured the purchase start near 1,507 px. The incomplete v61 mobile hero capture cannot establish a mobile aesthetic comparison.
7. **Large reassurance copy carries too little product meaning.** “Care for the brain you count on,” “A daily habit…,” and “Nothing to hide” receive major display hierarchy while product identity and reasons to trust this brand and dose stay small. The pain-led opening names a useful lived tension but does not complete the CHL-specific case.

## Broader evidence: three different questions

The expanded research screened **20 brands** across supplements, beauty, food, apparel, home, accessories, and consumer technology, plus four user-named comparison anchors: Grüns, IM8, Seed, and Salt & Stone. Screening means that a documented source or current product path was reviewed. It does not mean every current site was visually inspected. The public report separately counts the **13 brands** with verified current screenshots.

The evidence has three scopes and they must not be blended:

1. **Vendor-reported commercial outcomes.** Murad reports a 56% conversion-rate increase in the year after a whole-store migration; Aje reports a 135% increase in the weeks after replatforming and mobile optimization; BODi reports a 10-percentage-point checkout-conversion increase at launch. These are contextual platform or store outcomes, not proof that a particular PDP composition caused conversion.
2. **Commercial scale or popularity.** Huel's brand-published FY24 results, Oura's SEC filing, and other cited scale sources support why a brand is worth studying. They do not measure page design or PDP conversion.
3. **Direct visual references.** Current screenshots support observations about composition, information sequence, product explanation, image roles, and human presence. They do not establish sales lift. Recognition-only or selection-only brands remain aesthetic references unless stronger evidence is cited.

Allbirds is directional evidence only: the cited case says conversion rose after ship-from-store expanded inventory availability but supplies no percentage. It must not be presented as one of the quantified examples. Gymshark's current captures were blocked by a modal/navigation state, and HexClad's by a challenge page; they remain evidence-selection entries rather than visual references.

## What the directly reviewed examples show

- **Grüns:** a benefit-annotated packshot, attributed testimonial gallery, and lab-test entry near the buy box give product and proof specific visual jobs. Mint, yellow, green, a mascot, and strong outlines create a recognizable system. The visible UGC row was blocked at the embed level, so its playback was not evaluated.
- **Seed:** translucent materials, microscopic textures, and exploded-capsule callouts explain mechanism and vary composition while staying within a green/cream palette. This shows why palette alone is not the problem. Only the two usable desktop captures support this observation; the challenge-state captures were excluded.
- **IM8:** gallery frames change the argument across kit contents, benefits, and people; clinician context appears near purchase; saturated burgundy, full-bleed named-ambassador imagery, and early support keep human proof from becoming one isolated deep section. The desktop capture is labeled by its cookie-cleared state rather than assumed page position.
- **Graza:** illustrated packaging continues into the page's color, linework, harvest explanation, usage imagery, and delivery sequence. The 24-pound-versus-13-pound olive comparison and punchy-versus-mellow distinction turn the product difference into visual information.
- **Dr. Squatch:** packaging graphics, product texture, hard contrast, and ingredient imagery carry physical product character through the page. It is a continuity model, not a tone for CHL to imitate.
- **Jones Road:** real women, usage imagery, and skin-tone filters make human fit visible close to the product. Some how-to-use video panes were blank in the capture and are excluded from the finding.
- **Huel:** the confirmed clean top capture combines annotated nutrition, real flavor textures, and a concrete use case in the product gallery.
- **Aje:** full-body, close-up, and editorial photographs serve different product-understanding jobs while the purchase panel remains stable at the right. Its reported migration result establishes study relevance, not causation by this layout.
- **Anker:** illustrated specifications, a real hand for scale, dimensions, and a selection summary reduce choice ambiguity. The captured “top” filename is a mid-purchase gallery state, so the report labels the visible content rather than inferring scroll position from the filename.
- **Oura:** the product-overview page uses tactile wood and metal, controlled shadows, sensor close-ups, and a technical specifications table with footnotes to make a small device physically credible. It is not a buy PDP, and its long hero is not a model for CHL's purchase architecture.
- **Ridge:** the available captures are homepage evidence, not a wallet PDP. They support only the observation that material-finish close-ups and a named expert endorsement can create specificity.
- **Gymshark:** a visible 3.4 rating, a frank “not 100% squat proof” limitation, fit guidance beside sizes, and model images from different angles answer risk at the decision. The later collection and recommendation row is not a product-story section.
- **Murad:** large product-use frames show dispensing, application on a real person, and cream texture; numbered steps and educational articles continue the explanation. Its mobile decision state joins product name, reviews/questions, benefit tags, size, and subscription. The vendor-reported conversion outcome is a whole-store migration result, not proof of this composition.

These are transferable patterns, not designs to copy: give each image a different explanatory job; make the product's mechanism, use, and choice visible; place real authority near the decision; and carry distinctive product assets through the whole visual system.

## Preserve from the rejected candidate

- Readable Supplement Facts and product label treatment.
- Transparent first-payment, shipping, renewal, cancellation, one-time, guarantee, and high-dose terms.
- Real, usable CHL product and routine imagery.
- The pain-led opening as a useful premise, provided a future direction completes the brand-and-proof argument.

## Forensic correction

The initial research archive includes 14 references: seven with usable visual evidence, two blocked captures, and five text/indexed-only references. IM8 was text-only and Grüns was absent, so two important user-named benchmarks were not visually studied. Its 34 screenshots include failed captures. Screenshot quantity is not deep comparative analysis.

The final internal review checked five implementation details—kickers, text floors, tracking, insets, and overlay—and then recorded `SHIP`. The earlier round called the concept approved and the creative ceiling reached. That was self-referential internal direction, not operator acceptance. Fifty-five browser checks establish function, not commercial or aesthetic success.

The Impeccable detector reported 12 findings: three warnings and nine advisories. Its cream-palette and overused-font signals are secondary familiarity evidence. Palette choice, hidden modal borders, and color advisories do not explain the page's failure. No Impeccable critique snapshot was stored because its schema requires a numeric score and this scoped audit did not run a fresh full heuristic test; this synthesis is the authoritative record.

## Concrete brief for a future revision

1. One distinct product-and-purchase opening.
2. A gallery in which each frame has a different job: identity, readable label and dose, capsule scale, use, and verified quality.
3. All five verified creators visible or browsable early, without an empty grid.
4. Attributed real reviews placed near the objections they answer; no invented quotations.
5. Company provenance and an accessible real quality document, if available.
6. Product-specific art direction drawn from the existing packaging palette, iridescent circle, and real photographs.
7. Varied, purposeful image scale and customer-problem language constrained to verified claims.

Preserve the honest purchase terms. Avoid giant numbers, empty carousels, and aspirational copy presented as evidence. Compare desktop and mobile acceptance frames side by side before another full build. Preserve the rejected immutable PDP; any next revision must be a new fork.