CS-01 Alibaba Teaser Delta Audit — Quick-Scan Readiness Check

Resolved / Still Open

Three categories flagged June 30: title underleveling, boundary control, public/private proof confusion. Status as of tonight's fetch (last-modified July 3, 2026, 22:16 UTC):

Title and metadata framing — resolved. Live <title> reads: Alibaba.com — Redesigning B2B Trust at $50B+ GMV — Juno Chen. Mandate-forward. Names the company, the design problem, the scale, the designer. A hiring manager scanning a row of browser tabs reads a leadership claim. Done.

OG and social unfurl tags — resolved, one minor flag. og:title matches the mandate framing. og:description names the Head of Design & Research role, the structural trust gap, 32 interviews, executive buy-in, three cross-functional sprints, +20% daily transactions, +2.2pt NPS. Twitter card runs summary_large_image with a shorter description swapping NPS for −47% security concerns. Unfurl image is the logged-out homepage redesign. All strong. The og:url still points to the .html extension while Vercel 308-redirects to the extensionless path. Functional but sloppy. Low priority. Fix when you're in the file anyway.

Client-side gating on the full case — still open. Still the biggest problem on the site. A fresh fetch of cs-01-alibaba.html returned HTTP 200 with the complete case content: diagnostic sections, sprint sections, outcome data, agentic future-state material. Everything. The localStorage check fires after delivery, redirecting unauthenticated browsers to /login.html. But the HTML is already in the response body. View-source defeats the gate. This was flagged June 30. It remains unresolved, and it contaminates the trust narrative the teaser is trying to establish. More on this below.

Teaser density — newly flagged. The teaser page is 162KB of HTML carrying approximately 1,515 words of visible text, 23 images, and a full UI-components section that includes navigation specs, filter drawers, product cards, supplier cards, price-tier tables, color tokens, type scale, spacing scale, and border-radius values. That is a design-system documentation page wearing a teaser's clothes. I cannot confirm from available records whether this section was present on June 30 and went uncalled or was added between issues. If it was there and I missed it, that's on me. If Juno added it since, it signals she's actively building out the page, which means she's likely to act on this audit too.

The Walkthrough

The audience: a hiring manager with 40 browser tabs open, deciding which candidates deserve a closer look. Thirty seconds.

Browser tab. Alibaba.com — Redesigning B2B Trust at $50B+ GMV — Juno Chen. In a tab strip, this reads immediately as senior work at a known company with a quantified outcome. High confidence: doing its job.

Slack unfurl. A recruiter pastes this link into a channel with a hiring manager. They get: title that says trust redesign at scale, description that names the Head of Design & Research role and the research case behind executive buy-in, image that shows a real product screen. Three signals in one paste. The description is dense enough that even if Slack clips the tail, the truncated version still leads with role, structural trust gap, and research methodology. Mandate, scale, and outcome survive the cut. Not worth restructuring for.

First screen before scroll. The hero identifies the work as Alibaba.com B2B Platform Redesign 2024. Leads with the diagnostic insight: a consumer-browser experience serving procurement users. States that Juno named the gap, built the research case, secured the mandate, redesigned the platform. Metadata rows surface company, role (Head of Design & Research, North America), scale ($50B+ GMV, 200K+ suppliers, 25M+ desktop sessions), and outcomes (+7% DAU, +20% transactions, +2.2pt NPS, −47% security concerns).

Mandate, scale, outcome. All three land before scroll. A hiring manager who reads nothing else gets the message: this person diagnosed a structural problem at a major platform, got executive buy-in, led the redesign, and moved business metrics. Correct message.

One adjustment. Four outcome metrics is one too many for a scan. +20% transactions and −47% security concerns directly prove the trust thesis. +2.2pt NPS is meaningful but less immediate. +7% DAU measures engagement volume, and engagement volume is orthogonal to the trust claim this page is making. Moderate confidence: Drop DAU from the hero row. Keep it in the full case. Three metrics scan faster than four, and the two you keep are the two that prove the specific thesis.

Scroll depth: the three work sections. Homepage/first-impression, search-journey, PDP trust. Each follows the same pattern: diagnostic stat, before state, redesigned state. The 37-point sign-in gap below the app fold. The 73% session abandonment. The 47% security-fear citation with Trade Assurance buried. Each section shows browser mockups, legacy versus redesigned.

The before/after mockups are the strongest craft signal on the page. They let a hiring manager see the delta without reading a word. A scan-optimized viewer can scroll through all three sections in under ten seconds and register that real design judgment happened at each layer of the product. High confidence: these sections are doing exactly what a teaser should do.

Scroll depth: the UI-components section. Here the page breaks its own contract.

The components section includes navigation specs, search bars, source tabs, filters, applied filters, price range, sort options, filter drawer, product cards, supplier cards, price-tier tables, order buyboxes, color/size pickers, confidence signals, spec tables, Q&A accordions, how-to-buy steps, onboarding steppers, role tiles, auth forms, business-info forms, welcome states, trust badges, buttons, pagination, breadcrumbs, color tokens, type scale, spacing scale, and border-radius scale.

For the hiring manager who just scrolled past three strong work sections, this wall of component documentation does one of two things. They leave because they already got what they needed. Or they start recalibrating: the page's altitude shifts from "I led a trust redesign at enterprise scale" to "I also built a component library." That second read is an IC-level signal. It pulls the narrative down precisely when it needs to stay at the leadership layer.

High confidence: Move the entire UI-components section behind the password gate. Replace it on the teaser with a single row showing 3–4 representative components at thumbnail scale, labeled something like "Design system: 40+ components, token architecture, multi-surface consistency," followed by the CTA. Craft-curious viewers get a reason to request access. The leadership narrative stays intact.

Exit point. The closing section restates the pattern (named the gap, secured the mandate, redesigned the platform), summarizes the three surfaces, repeats the CTA. Clean. Footer links to Equinox+ (backward) and Red Cross (forward), giving the engaged viewer a next case without friction.

Trust-surface coherence. Outside the gate problem (addressed below), the page's language, framing, and visual emphasis are coherent with the trust-design positioning throughout. The diagnostic stats, the before/after structure, the outcome metrics all reinforce the narrative that Juno identifies where trust breaks down and redesigns systems to restore it. Nothing on the public teaser contradicts that claim.

The Gate Fix

Two options.

Option A: Server-side gating. The full case page returns a login/request-access page to unauthenticated requests. The HTML never reaches the browser. This is the correct answer for this candidate. Your entire professional positioning is "I redesigned trust at enterprise scale." Your portfolio's own access boundary should be real.

Option B: Remove the implication. Change "Request the full case study" to "Continue to full case study." Treat the teaser as a summary layer and drop the access-control framing entirely. No gate, no pretense. This is the fallback if the hosting setup makes server-side gating impractical this weekend.

The current middle state, where the CTA implies protection and the server delivers everything, is the worst of the three. High confidence, unchanged from June 30. Fix this before any outreach.

What to Do This Weekend

Most hiring managers are watching fireworks tonight, reviewing portfolios on Monday. H2 headcount unlocks mid-July. You have a window to get this right before the next wave of outreach lands.

• Fix the full-case gating. Server-side if possible. Remove the implication if not. Do not leave theatrical gating on a case study about trust.
• Move the UI-components section behind the gate. Replace with a thumbnail strip and CTA. Single highest-impact change for teaser scan quality.
• Drop +7% DAU from the hero metrics. Three metrics. The two that stay prove the trust thesis directly.
• Update og:url to the extensionless path. While you're in the file.

The title, metadata, hero, and three work sections are strong. First-screen experience is landing the right signals for the right audience. The problems live below the three work sections and behind the CTA. Fix those and this teaser is ready for the July window.