PDP Quality Review · Pre-Build Audit

RevShift × Jessar · 2026-06-10 · Deployed canonicals audited: hero_pdp_sm_v2 · standard_pdp_sm_v1 · home_sm_v6 — against Caraway patterns, Soft Modern guidelines, the B2B-first mandate, and the catalog as we know it today · Internal
8.5/10
Hero PDP today
8/10
Standard PDP today
6+1
P0 upgrades + 1 red flag
0
Things to tear down

🚩 Red flag found during audit — conflicting heritage claims

The three deployed renders state three different founding years: hero PDP footer says "Since 1968" · standard PDP footer says "since 1987" · the homepage says "since 1996" in five places (hero, brand thesis, approach, credibility band, footer). Heritage is a core trust asset for the B2B story — it cannot ship with three numbers. Action: confirm the real founding year with Michael, then it becomes a single token used everywhere. (This is exactly the kind of detail a Linen Chest buyer notices.)

Verdict

The bones are excellent — keep them. The typography system, the italic discipline, the meta-row beam, the hairline build grid, the dark B2B specs panel, the oversized sub-brand mark: these are genuinely distinctive and already read more "considered" than most $100k retail sites. Nothing needs tearing down.

But the renders predate most of what we now know. They were designed May 5–7 — before canonical variant families, the no-brand-in-title rule, the features-gold pass, the 3D box renders, Cook & Fresca, the v3 taxonomy, and the facility video. Audited against Caraway and the actual business goal (win shelf space at Linen Chest / Stokes / Corbeil; build consumer trust; no cart), there are six structural upgrades that belong in the renders before a line of Liquid is written — because they change template structure, not styling.

The biggest single miss: the pages sell the product but under-sell the company. A retail buyer landing on this PDP must leave thinking "these people are a serious, certified, retail-ready partner." That trust layer — certifications, partners, heritage, packaging — is currently scattered or buried.

What stays — the approved craft (do not touch)

Cormorant Garamond + Inter system; italic-as-narrative rule (titles are data, stories get em)
Meta-row as hairline structural beam between lede and CTAs
4-col hairline build grid ("craftsman's diagram") + warm use-tiles
Dark inverse specs panel — the deliberate consumer-story / B2B-data split, with PDF + sample CTAs
Oversized Cormorant sub-brand mark + sub-brand dot system (sage/dusty/terra)
Dual CTAs everywhere: Find a Retailer (consumer) / Become a Dealer (B2B) — correct for a no-cart site
Bilingual EN/FR architecture (hero render) + annotation drawer with Hyper section mappings

P0 — six structural upgrades to re-render before the Liquid port

P0-1
Variant-family band — "Available in" (the signature Caraway move, and we're not doing it)
Caraway's PDP is organized around colourways. Jessar's catalog has exactly this shape — kettles ×4 colours, mills ×5, Dutch ovens ×3, lamps ×2–3, bulbs by pack/temp — but as individual listings (client convention). The hero render has two decorative swatches in a corner; the standard render has none.
Change: a real swatch/chip band directly under the title — each chip links to the sibling listing. Canonical titles make families derivable programmatically (same template, different token). Kitchen gets colour swatches; bulbs get pack-size / colour-temp chips ("2700K · 5000K · 4-pack · 6-pack"). One listing each, cross-linked — convention preserved, Caraway effect achieved.
P0-2
Trust layer — certifications, partner ribbon, heritage stats
B2B-first means the buyer's first question is "are these people legit?" Today: certifications appear only inside a collapsed accordion on the standard PDP; "Since 1968" is buried in the footer; the Costco/Linen Chest/Staples names exist only as body text in section 8.
Change: (a) cert badges (cULus · FCC · ICES) rendered as marks in the dark specs panel on both tiers; (b) a quiet stat band — "Since 1968 · 1,700+ products · 200+ retailers · Canada-wide" — high on the page (under hero or above dealer CTA); (c) partner wordmark ribbon ("Found at…") near the dealer CTA. Tasteful, hairline-separated, zero salesiness — the Soft Modern way.
P0-3
Packaging / "Retail-ready" block — our B2B answer to Caraway's unboxing
Retail buyers buy shelf presence, not just products. We now hold 107 branded 3D box renders (bilingual JS Gourmet packaging) plus box shots in the WP extras — an asset Caraway doesn't even need, and exactly what Linen Chest's buyer wants to see. No render currently shows the product in its packaging.
Change: a "Retail-ready" section (hero tier) / packaging gallery slot (standard tier): box render + callouts — bilingual packaging, UPC on box, inner/master case pack, shelf footprint. Sourced from jessar.qty_per_package/qty_per_container + box render media. This section does the B2B selling silently.
P0-4
Persistent B2B inquiry — sticky bar + unified mechanism on both tiers
Lead capture is this site's #1 conversion job, and Caraway's equivalent (sticky add-to-cart) is its hardest-working component. Today the hero buries "Become a Dealer" at section 8; the standard tier has an "Add to B2B Order Inquiry" button the hero doesn't have. Inconsistent, and nothing persists on scroll.
Change: a slim sticky bar that appears after the hero scrolls away — product name + SKU on the left, "Request price list" / "Add to inquiry" on the right. The inquiry-list pattern (buyer assembles SKUs → submits one form) becomes the site-wide B2B mechanism on both templates. Soft-Modern styled: canvas background, hairline top border, no shadowed monstrosity.
P0-5
Canonical-title display logic — split at the em-dash
The new title standard is Name — Variant, Size with no brand. Rendered flat, long titles read like data dumps ("Enameled Carbon Steel Whistling Kettle — Cherry Red, 2.1 L").
Change: the template splits the title at the em-dash: the name as the big serif line; the variant + size as a styled sub-line (chip or small-caps meta) that visually pairs with the variant-family band (P0-1). Brand identity comes from the sub-brand block — consistent with Michael's no-brand-in-title rule. The two PDPs render this identically.
P0-6
Real gallery + video-ready story
The hero render repeats one WP image four times; the standard render is placeholder boxes. Reality today: 1,163 SKUs average ~2.8 images (primary + WP extras + box render), and a professional facility video is being produced. The client explicitly wants a "very visual" site.
Change: gallery rendered with genuinely different angles + a dedicated packaging thumb; story section gets a video-capable block (16:9, poster frame, soft-corner) so the facility film — and future product loops from the parked AI-imagery program — drop in without retemplating. Banner imagery slots stay on home/collections (separate review).

Caraway pattern audit — where we match, where we deliberately diverge

Caraway patternOur statusCall
Colourway selector on PDPMissingP0-1 variant-family band (cross-linked listings)
Sticky add-to-cartMissingP0-4 sticky inquiry bar (B2B translation)
Trust badges / press ribbonBuriedP0-2 certs + partners + heritage stats
Editorial storytelling bandsStrongKeep — story/build/made-for are already better-written than Caraway's
Materials transparencyStrongKeep — build grid + features-gold data (9,304 validated bullets)
Unboxing / what's includedPartialP0-3 packaging block (better than unboxing for B2B)
Comparison table ("vs other guys")HoldP2 — client's own deck positions vs Le Creuset; needs a legal/tone check. Frame as materials education if used.
Reviews / UGCN/ANo direct sales → no reviews. B2B substitute: retailer logos + heritage + certifications (P0-2). Optional later: retailer quotes.
"Complete the set" cross-sellPartialP1 — wire to product_line first (Inspiration Italia, Cook & Fresca), then category

Corrections after auditing the deployed standard PDP (standard_pdp_sm_v1)

The deployed Soft Modern standard PDP is stronger than first assessed (score raised 7 → 8): it is already fully bilingual, already has the PDF / sample / dealer actions on the dark specs panel, shows features openly instead of hiding them in accordions (the right Soft Modern call), and carries a clean meta-strip (SKU · UPC · master carton). Two earlier P1s are withdrawn as already solved. The six P0s all still apply to it — no variant band, no sticky inquiry, no cert badges, no packaging block.

Homepage quick-scan (home_sm_v6) — good slots, stale story

Already built for the client's asks: a banners "action surface" (§9), a 4-up vertical-video reels section (§8 — drop-in ready), a credibility band (§10), press strip (§11), and a "Two Paths" consumer/B2B split (§12). The facility video has a natural home in the Approach section (§7) and/or hero. The visual machinery the client wants exists — it's waiting on assets, not redesign.

But the brand story is stale on three counts: (1) "Ten brands" framing (hero, thesis, footer) — reality is 4 house brands + product lines, with Veraroma promoted to top-level and iBake under it per Sean's latest direction, Cook & Fresca now a real launched line (36 products), and Pop Fluff empty; (2) the category section predates the v3 three-level taxonomy; (3) the "since 1996" claim ×5 (see red flag). The portfolio section needs a re-render once the brand hierarchy is confirmed — it's the homepage's centrepiece and currently tells a story the data contradicts.

P1 — land during the build (no re-render needed)

ItemNote
P1Breadcrumb → v3 taxonomyStandard render is category-first already (good); deepen to the v3 drill-down (Dept → Category → Sub-cat) with the brand as a separate chip.
P1Cross-sell from real dataCards are placeholders; wire to product_line first (Inspiration Italia, Cook & Fresca), then category — with live images (imagery baseline now supports it).
P1Footer brand list refreshStandard PDP footer lists Pop Fluff (0 products) and omits Veraroma (now top-level). Footer nav should derive from the confirmed hierarchy, not a hand-typed list.
P1Copy provenanceRender romance copy was hand-written mockup text (invented specifics). Build pulls from validated product_content_v2 (0 known factual errors) — never freestyle at template level.
P1Auto per-SKU sell sheetsThe "Download catalog page (PDF)" links need a real target: generate branded PDF sell sheets from metafields (one per SKU) rather than hand-making them.

P2 — post-launch / optional

P2Materials-education comparison band (rubberwood vs ABS, enameled steel vs aluminum) — deck content exists; legal/tone check first.
P2Icon-led benefit row (Caraway-style) fed from features-gold — only if icons can stay tasteful; the typographic build grid may simply be better.
P2Product loop / orbit video per hero SKU — parked AI-imagery program (change-order scope).
P2Retailer testimonial quotes — the B2B answer to reviews, when client relationships allow.

Content-to-section audit — what data we hold vs. what the templates can show

Coverage of display-worthy data across the 1,667 primary SKUs (validated content, post-QA). This is what justifies — or kills — each proposed section. Principle confirmed with Sean: every section is dynamic and renders only when its metafield is populated. One section library serves both tiers; a product "upgrades toward hero" by gaining content, never by switching templates.

DataKitchen (608)Light-fix (434)Home/Bath (293)Bulbs (271)Elec (61)
4+ feature bullets (validated)95%97%96%99%80%
6+ feature bullets44%49%54%43%40%
Romance copy100% across all categories
Material94%63%90%22%
Capacity55%29%
Care instructions53%34%17%
Certifications3%*21%1%*61%18%
Warranty2%60%21%14%
Photometrics (W/lm/K/dim/base/life)partial~100%
UPC + master case (B2B block)~100% — always renderable

*Kitchen/Home cert coverage fills when Jessar returns cert_request.csv (pending).

The resulting section library (all metafield-gated)

SectionRenders whenTierStatus
Variant-family bandfamily metaobject existsBothP0 · new
Feature highlights (structured, top-3 + rest)4+ bullets — 96% of catalogBothP0 · upgrade
Certifications + warranty bandcerts/warranty populatedBothP0 · new
Photometrics block (the bulb "function display")wattage present — ~100% of bulbsBothP0 · new
Care & materialscare or material populatedBothP1 · promote from spec row
Capacity / size guidecapacity or dims populatedBothP1
"How it works" — appliance functionsjessar.functions populatedHero→BothNeeds data pass — ~64 appliance SKUs (toasters, air fryers, kettles): programs/modes/controls exist in features but unstructured. Small targeted extraction task.
Retail-ready / packagingbox render media existsHero (gallery slot on standard)P0 · new
Story · Build grid · Use-tiles · Sub-brand blockrespective metafieldsHeroExists — keep
Dark B2B specs + sticky dealer/inquiryalwaysBothExists + P0-4 sticky on standard

Net answer to "do we need more informational sections": yes — four new ones are justified by data we already hold (variant band, certs/warranty, photometrics, packaging), one needs a small data pass first (appliance functions), and the feature display deserves an upgrade since it's the single richest asset (96% coverage). Everything else: the data isn't there yet — sections would render empty, so they wait.

Recommended sequence

1. Sean validates this priority list (especially P0-1 variant band + P0-4 inquiry mechanism — both are site-wide patterns, not just PDP) and asks Michael two questions: the real founding year, and final brand-hierarchy confirmation (Veraroma top-level / iBake under it).
2. We re-render: hero_pdp_sm_v3.html + standard_pdp_sm_v2.html with the six P0 upgrades — the hero on a Cook & Fresca kettle (4-colour family, brand-new line: proves the variant band on real data) — plus a home_sm_v7 portfolio refresh once the hierarchy is confirmed.
3. Sean approves the new renders → Liquid port begins against approved structure. (Render-before-commit, per the quality bar.)

Estimated render effort: one working session for the two PDPs; the homepage portfolio refresh rides on the hierarchy confirmation. The cost of skipping this: retro-fitting variant bands + sticky inquiry into live Liquid mid-build — far more expensive than re-rendering now.

Generated 2026-06-10 · RevShift internal · Sources: hero_pdp_sm_v2.html, standard_pdp_sm_v1.html, home_sm_v6.html (deployed canonicals), brand_guidelines_soft_modern.html, PROJECT.md §3/§5/§6, product_content_v2 (1,703 live), features_gold, render3d index, Caraway PDP pattern reference.