# UGC placement and motion advisory

This consultation evaluates an unrendered proposal against current-page screenshots. It is not a page rescore, prototype test, or measurement of conversion, engagement, performance, or accessibility conformance.

## Current state and proposal

The current page presents the complete hero offer, then formula and supporting sections, then “Their routines. Their own words.” before customer reviews. A smaller Eliza story-proof control already appears in the hero.

The proposal would move the existing 20-card routines section after the complete hero offer and before “Two Ingredients, Nothing to Hide,” while retaining the smaller hero control. Continuous automatic horizontal motion and stationary manual navigation were assessed separately.

## Independent recommendations

| Lens | Earlier placement | Continuous automatic motion | Manual swipe/drag |
| --- | --- | --- | --- |
| Craft | Conditional | Oppose | Support |
| Journey | Conditional | Oppose | Support |
| Trust | Conditional | Oppose | Support |

All three judges recommend the same local sequence if the proposal is rendered: retain the small hero story control; keep the complete buy box and its disclosures together; add a compact, stationary routines row; then transition to the formula. Remove the former later instance rather than duplicate it.

Their shared conditions are:

- Lead with a contributor other than Eliza so the moved row adds breadth instead of immediately repeating the hero.
- Show four cards across on desktop in roughly an 1120–1200 px row, and one roughly 260–264 px card plus a next-card preview at 390 px.
- Keep quotations, attribution, source labels, play controls, visible previous/next controls, and an accurate position count readable.
- Make swipe/drag additive to visible controls. Keep the row stationary until visitor input; do not autoplay row motion, video, or audio.
- Keep reduced-motion behavior stationary and use immediate position changes without animated settling.
- Preserve the current placement if a compact render cannot keep the formula timely, preserve attribution, or create a clear boundary between adjacent green sections.

The main tradeoff is earlier experiential proof versus delaying ingredient explanation and creating a longer green passage immediately after the offer. The screenshots show the current page only, so sizing and behavioral recommendations remain conditional until rendered and observed.

## Evidence and raw judgments

The evidence packet contains 10 current-page frames: `evidence/frame-001.jpg` through `evidence/frame-010.jpg`. Frames 001–002 and 006–007 show hero/offer context; 003 and 008 show formula context; 004–005 and 009 show the current UGC section; 010 shows downstream review context.

- [Craft verdict](runs/consult-r1/craft/verdict.json)
- [Journey verdict](runs/consult-r1/journey/verdict.json)
- [Trust verdict](runs/consult-r1/trust/verdict.json)
- [Run receipt](runs/consult-r1/RUN-RECEIPT.json)
- [Input manifest](input/INPUT-MANIFEST.json)
- [Current-order record](input/CURRENT-ORDER.json)

All three fresh GPT-6 Astra/high sessions completed once and passed the advisory schema and citation checks. No numeric score was requested or produced.