Internal commercial review · Re-audit
The same thirty-phase standard applied to the rebuilt page destined for pro.jimmycoco.email, scored side-by-side against the original funnel entry page from the first audit. Headline: the overall score moves from 4.6 — commercially underperforming to 8.0 — strong. Every critical finding from round one is resolved on this page. What remains between 8.0 and world-class is specific, short, and listed below — most of it is proof and plumbing, not design.
Part one · The comparative scorecard
The one negative delta is honest bookkeeping, not regression: the old audit scored Shopify's polished checkout (7.5); this page's equivalent is the trial form, which looks excellent but is wired to mailto: — see finding 1. The two lagging categories, trust and form UX, share the same root: this page is finished as a design and unfinished as plumbing.
Part two · The first audit's critical findings, revisited
| Round-one finding | Status | How |
|---|---|---|
| 1 · Raw Liquid error in cart | N/A here / open on shop | This page has no cart. The error still exists on jimmycoco.co.uk and remains the single most urgent fix in the wider funnel. |
| 2 · Hero asks for nothing | Resolved | Outcome headline, two CTAs (trial / calculator), proof strip. The five-second test now returns: who it's for, what it is, why care, what to do. |
| 3 · B2B and B2C in one funnel | Resolved | Salon-only audience; retail reframed as the salon's second revenue line at RRP with no consumer buy buttons; consumer site linked only in the footer. |
| 4 · Flagship with zero social proof | Partially resolved | No product is asked to carry £79 unproven here; the mitt's 20 reviews are used. But salon-side proof is still absent — see finding 2. |
| 5 · Sales story trapped in images | Resolved | All six deck slides are native, readable, indexable page sections; only decorative photography remains as imagery. |
| — · Claims language | Retained by decision | Owner's call, recorded and excluded from scoring: site claims vocabulary kept verbatim pending updated brand guidelines. |
Part three · What still caps the score
Problem: "Post me a trial" opens the visitor's desktop email client — jarring on desktop, frequently broken on mobile, invisible to analytics, and every abandoned half-written email is a lost lead you'll never know existed.
Fix: wire it to a real endpoint. The natural home already exists: the repo's /api/campaigns/enroll → Supabase pipeline, which would drop every trial request straight into the outreach system and its onboarding sequences. A form service is the quick interim. This is the one change standing between the page and its purpose.
Problem: the page proves Hollywood credibility but offers zero peer proof — no stockist quote, no salon name, no "we switched because…". Salon owners trust other salon owners above celebrities.
Fix: even one authentic line from an existing stockist with a salon name and city, placed between the calculator and the application guide, would lift trust a full point. Collect it via the trial follow-up sequence that already exists in the campaign system.
Problem: no visible labels or aria-labels; screen readers announce empty fields, and placeholders vanish on focus (WCAG failure, and genuinely harder for everyone over 45 — a relevant share of salon owners).
Fix: proper floating or top labels; ten-minute change.
Problem: no analytics on the page — calculator engagement (the single most telling intent signal available) is unmeasured, as are scroll depth and CTA clicks.
Fix: lightweight analytics + events on slider use and form submission. Calculator-users who don't submit are the retargeting audience.
Problem: the story section reuses campaign imagery of a very recognisable person from the consumer deck. Rights that cover consumer marketing don't automatically cover a trade recruitment context.
Fix: confirm scope, or swap to Jimmy-at-work imagery, which arguably sells the method better to professionals anyway.
Problem: no FAQ: equipment compatibility, patch-test policy, delivery coverage (IE? mobile pros?), minimums. Each unanswered question is a reply-email of friction.
Fix: a five-question FAQ above the close. · Also low: the 1.1MB single-file build is fine for launch but should move to hosted images with lazy-loading when it becomes a real site; custom focus states would complete keyboard polish.
From 4.6 to 8.0 — the design work is done. What's left is proof and plumbing.
Would this page convert salons today? Yes — materially better than the original, which buried its offer in an image carousel behind a biography. The narrative now runs: recognition → formula → economics → simplicity → second revenue line → free, risk-free next step. That is the correct trade-sales arc.
The single highest-impact next change: replace mailto: with the enroll endpoint. Second: one real stockist quote. Those two alone are realistically worth another point of overall score — and more importantly, actual captured leads.
Forecast (same assumptions as round one): against the original page, expect multiples rather than percentages on trial-request rate — the original had no form and no visible ask at all. Measured honestly from launch: calculator engagement above 40% of visitors would confirm the centrepiece is working.
Don't touch: the calculator-echo mechanism, the dark close section, the litre facts strip, the "no lock-in" posture. They're the page's personality now.