02
Alibaba.com · B2B Platform Redesign · 2023

The signals before
anyone touched
a screen.

Desktop was 25% of traffic. Eighty percent of transaction value. The platform was built for the wrong audience — and no one had named it yet. I built the research case, secured the mandate, and redesigned it. $50B+ GMV ran through every decision.

01CompanyAlibaba.com
02ProjectB2B Site Refresh · 2023
03RoleHead of Design & Research, North America
04TeamDesign · Research · PM · Engineering
05Scale$50B+ GMV · 200K+ suppliers · 25M+ desktop sessions
+7%
Daily active users
+20%
Daily transactions
+47%
Organic search traffic
+2.2pt
NPS · $50B+ GMV platform
01
01The Diagnostic

A mobile-first company
sitting on a desktop problem.

32 cross-functional interviews, a Baymard audit, and gaze tracking on $500+ order sessions all pointed to the same conclusion: the desktop was built for consumer shoppers. The buyers using it were running procurement.

Three points of failure. Each one traceable to a single design choice.
Discovery 01
First Impression
"Flea market."
The page read like a consumer marketplace. B2B buyers — sourcing at scale, running procurement — saw no signal it was built for them. Sign-in rate ran 37 points below the app, on the same products, to the same buyers.
Discovery 02
Search Journey
73%
Sessions ended without engagement. No B2B filters, no card-level trust, no response-rate sort. Buyers opened 10+ PDPs to qualify a single supplier.
Discovery 03
Transaction Confidence
47%
Buyers cited security fears at payment. Price tiers behind inquiry wall. Trade Assurance below fold. Gaze paths showed tab-switching to price-check externally.
02
02Three sprints. Three bets.

Research signal → design bet → measured outcome.
That’s the structure. The work follows.

The three failures from the diagnostic each became a sprint. First impression, search, transaction — in that order.

Sprint 01
First Impression — The B2B Signal

B2B credibility
above the fold.

“This looks like I’m buying from a flea market, not a professional business partner.”

First-time buyers on $500+ orders called it a flea market. No distinction between a consumer browsing and a business sourcing at scale — no logged-out/logged-in state, no B2B signal above the fold. I designed the gate: B2B credibility above the fold, browse-first entry without a credential wall, and a personalized returning-buyer state. The PM worried sign-in would drop. It went up.

+4%Sign-up rate
+8%Pages per session
Before — Pre-2023 Consumer marketplace aesthetics · EIN gate on registration · No B2B value prop · Sign-in rate 37pt below app
alibaba.com
Pre-2023 Baseline · Homepage · Logged Out
Before — Old Alibaba.com homepage pre-2023 baseline
01Consumer framing — “Featured Suppliers LIVE” + “Just for you” tiles read as B2C. $500+ intent buyers called it a flea market.
02Sign-in friction wall — EIN gate at registration created a 37pt sign-in deficit vs. the same buyer population on app.
03No B2B value prop — Trade Assurance, verified supplier count, and sourcing scope were all below fold or missing.
After — 01a · B2B Homepage · Logged Out B2B credibility above fold · browse-first entry without credential gate · +4% registration
alibaba.com
Redesigned Homepage · Logged Out · Desktop
After — B2B Homepage Redesign · Logged Out
01B2B hero above fold — Warehouse imagery + “Source smarter” headline replaces the promotional carousel. Professional intent, not consumer browse.
02Why Source panel — Trade Assurance, verified suppliers, and sourcing protections surface immediately below the hero. Value prop is the page.
03+4% sign-up rate — Browse-first entry, no EIN gate. Sign-in became contextual, not a wall.
After — 01b · Homepage · Logged In · Personalized Returning buyer dashboard · personalized product grid · +8% user actions
alibaba.com
Redesigned Homepage · Logged In · Returning Buyer
After — Homepage Logged In · Personalized
01Personalized dashboard — Returning buyer lands in sourcing context: recent activity, saved suppliers, and recommended categories.
02Promotions in context — Injected into the personalized grid by purchase history — not a broadcast banner.
03+8% pages per session — Contextual personalization drove deeper sourcing engagement vs. the static logged-out state.
Sprint 02
Search Journey — B2B-Native Architecture

Trust signals
at card level.

“I can’t tell if a supplier is any good until I’ve read their entire page. And by listing 10, I’ve lost track of the first one.”

Search was built for browsing, not sourcing — no trust signals at the card, no B2B sort, MOQ buried three page-loads deep. I redesigned the card to carry what buyers need at a glance: Trade Assurance, response rate, OTD — as primary sort controls, not sidebar filters.

+9%Search completion
+3%Conversion
Before — Search Baseline Consumer-inherited layout · No card-level trust · No B2B sort options · MOQ buried · 73% session abandonment
alibaba.com › search?q=humidifier
Pre-2023 Baseline · Search · Consumer Layout
Before — Old Alibaba.com search results pre-2023 baseline
01No card-level trust — TA, VS, and RTS badges absent. Every qualification required a separate PDP click.
02Consumer sort only — Best Match, Price, Newest. No Response Rate, no OTD. B2B’s #1 concern was invisible.
03MOQ hidden in PDP — The primary B2B viability filter required a full page load. Buyers opened 10+ PDPs per session.
After — 02 · B2B-Native Search Architecture Trust signals at card level · Response Rate & OTD sorts · supplier-view tab · +9% search completion
alibaba.com/search?q=humidifier&filter=ta%2Cvs&sort=response_rate
Search · B2B Filters · Response Rate Sort
After — B2B-Native Search Architecture
01Trust at card level — TA · Verified · RTS badges on every card. Supplier credibility assessed at browse, zero PDP clicks.
02B2B-native sorts — Response Rate + OTD as first-class controls. 34% of sessions sorted by Response Rate in the first month.
03Supplier view tab — Products / Manufacturers / Worldwide toggle shifts from SKU to company-level qualification without leaving search.
Sprint 03
Transaction Confidence — PDP Trust Architecture

The PDP as
the spreadsheet.

“I always build a spreadsheet before contacting anyone. I need to see the math before I commit to a conversation.”

Buyers were building spreadsheets before they’d commit to a conversation because the PDP couldn’t answer the question that mattered: does the math work? I put the math on the page — tier pricing inline, a live order calculator, Trade Assurance in the primary scan zone.

+3%Payment conversion
−47%Buyer-reported security concerns
Before — PDP Baseline No trust above fold · inquiry-only pricing · 47% security concerns · no order calculator
alibaba.com › product › 2024-polka-dot-halterneck-dress_987654.html
Pre-2023 Baseline · PDP · Inquiry-First
Before — Old Alibaba.com product detail page pre-2023 baseline
01Inquiry-only pricing — “Contact supplier for price” replaced actual tiers. Buyers couldn’t evaluate viability before committing to a conversation.
02Trust below fold — Response rate and OTD buried in the right sidebar. Not in the first scan zone.
03No payment protection — Trade Assurance not surfaced at the moment buyers made commitment decisions.
After — 03 · PDP Trust Architecture Guaranteed badge above fold · tier pricing inline · live order calculator · supplier trust in primary scan zone
alibaba.com/product/2024-polka-dot-halterneck-dress_987654.html
PDP · Ready to Ship · Trust Architecture
After — PDP Trust Architecture
01Guaranteed + tier pricing above fold — All price tiers visible before any scroll. Buyers see the math without a single message sent.
02Live order calculator — Qty × tier + shipping + coupon = total, in real time. The PDP became the spreadsheet.
03Trust in scan zone — Response Rate, OTD, and Trade Level alongside price in the primary read path. No sidebar scroll.
03
03If I were building this today

The same three surfaces.
A complete sourcing loop.

I’d design the loop from intent to transaction. Natural language in Chat becomes a structured brief — the agent parses what to buy, at what price, under what conditions — then traverses Search and PDP in real time, evaluating every trust signal against those parameters. When the buying signals clear, the transaction completes autonomously. The buyer defined the intent once. The AI decided what, when, and how to pay.

Surface 01 · AI Chat
NL → structured brief.
No form, no re-entry.
Procurement agent · sourcing brief
500× custom water bottles, 500ml stainless, custom logo
Lead time ≤20 days ≤$10.00/unit Trade Assurance required
AI
Brief parsed · querying search
qty = 500 · price_max = $10.00
lead = ≤20d · cert = trade_assurance
———————————————————
Scanning 12,483 results → filtering TA + RTS…
agent queries search with spec →
Surface 02 · Search
Card-level trust signals.
12K filtered, zero clicks.
stainless water bottle custom logo 500ml
Yiwu ProMetal Co., Ltd.
$5.40–$5.80
🛡 Trade Assurance ✓ Verified Supplier Ready to Ship
MOQ 100 pcs Response ≤3h OTD 99%
Hangzhou Steel Mfg.
$5.60–$6.10
🛡 Trade Assurance Ready to Ship
MOQ 50 pcs Response ≤6h OTD 97%
opens top result · evaluates PDP →
Surface 03 · PDP
Inline pricing.
Brief confirmed in ms.
🛡 Alibaba Guaranteed ✓ Verified Manufacturer
Custom Logo 500ml Vacuum Insulated Stainless Steel Water Bottle · BPA-Free
2–49 pcs
$10.40
50–1,999 pcs
$9.90
← brief.qty 500
2,000+ pcs
$8.90
🚚 FREE shipping capped at $20.00 · orders 50+ pcs
Lead time: 14 days ✓  ·  500 × $9.90 = $4,950  ·  budget: $5,000 ✓
Complete sourcing loop
Brief to order.
Every surface closed.
Chat parsed the spec. Search surfaced qualified suppliers. PDP confirmed the math. One loop — no spreadsheet, no platform exit, no unanswered question.
Brief parsed Supplier qualified Math confirmed Order
Yiwu ProMetal · 500 pcs · $9.90/unit
🛡 Trade Assurance  ·  Lead time 14d
Total $4,950  ·  budget $5,000 ✓

Every surface in the sourcing loop carries structured data, accurate trust signals, and inline pricing. The loop closes on the platform — no spreadsheet required.