From Browsing to Trading: Designing a B2B Marketplace Where Deals Move in Six Figures

Context

Elex is a B2B wholesale marketplace where businesses trade electronics in bulk — 150-unit lots of iPhones, 200-unit blocks of Galaxy flagships, laptop inventory by the carton. The buyers aren't consumers; they're distributors and trading companies moving hundreds of thousands of dollars per transaction, picking up from warehouses, and competing for inventory that disappears in hours.

The engagement ran the full arc: research into how existing wholesale marketplaces actually work, a structured study of the leading competitor's app, a design proposal, the complete mobile UI, and the brand identity.

The Problem

Consumer e-commerce patterns collapse the moment the cart holds $372,000.

• Wholesale buyers evaluate unit economics — price per unit, quantity, storage variant, packaging type — not product photography and lifestyle copy
• High-demand lots are scarce and fast-moving; hesitation means losing the deal to another trader
• Six-figure transactions can't run on card checkouts — they need pre-funded working capital and an auditable money trail
• Trust runs both ways: sellers need verified, compliant buyers; buyers need certainty about what's in the carton
• The existing marketplace experience treated traders like consumers — decorative, slow, and structurally wrong for the job

The core insight from the research phase: a wholesale marketplace isn't a shop with bigger numbers — it's a trading floor. The design has to optimise for decision speed, not browsing pleasure.

Behavioral Insight

Studying how traders actually operate — across existing marketplace behavior and the competitive landscape — surfaced three consistent patterns:

1. Traders think in deals, not products. The mental unit is "150 units of 15 Pro Max, 256GB, master carton, at $950/unit = $142,500" — a single decision object. Any layout that fragments those variables slows the decision.
2. Speed is the real currency. Good lots vanish. Traders shortlist, watch, and strike — which makes watchlists, reservation holds, and instant wallet payment core trading tools, not convenience features.
3. At this transaction size, money needs a paper trail, not a checkout. Buyers pre-load working capital and think in terms of balance, utilisation, and reconciliation — every debit, deposit, and adjustment must be traceable to a reference number.

The category default asked traders to shop. Elex had to let them trade.

Design Hypothesis

If we structure every surface around the deal as the atomic unit, compress the path from discovery to committed capital, and make money movement fully legible, then traders transact faster with more confidence — and the marketplace wins on the only metric that matters in wholesale: velocity with trust.

This meant a deliberate shift in the design language:

The Approach: Designing the Trading Loop

The app was structured around the questions a wholesale buyer actually asks, in order:

1. What's moving right now? — deals-first home with live lots and time-boxed offers
2. What are the economics? — unit price, quantity, variant, and packaging visible before any tap
3. Can I hold it while I decide? — watchlist and reservation windows
4. Can I pay instantly? — pre-funded wallet with visible balance and utilisation
5. Where's my stock and my money? — order states built for logistics, and a full transaction ledger

Key Design Decisions

The deal as the atomic design unit. Every listing carries its complete decision payload in one card: model, storage, colour, unit count, price per unit, packaging type (master carton vs individual box), and computed deal total. Brand-first navigation — Apple, Samsung, OnePlus, HP, Dell, Google, Oppo, Acer, Sony — mirrors how traders actually segment inventory, and search runs on SKU, brand, and model, because that's the language of the trade.

A wallet designed as working capital. The wallet is the financial heart of the product: available balance front and centre, deposits via bank transfer and net banking with 256-bit encryption messaging, and a transaction ledger where every entry — deposit added, order debit, return adjustment, tax correction — carries its reference number and timestamp. At checkout, the order shows wallet utilisation (what share of available capital this deal consumes) — turning payment into a capital-allocation decision the buyer can make in one glance.

Checkout as a deal memo. Confirmation reads like a trade document, not a receipt: pick-up warehouse address, item summary by lot and variant, and an itemised bill — subtotal across units, bulk discount, platform and processing fees, freight, and tax — down to the grand total, with wallet impact alongside. Nothing about a $372,335.40 commitment is left ambiguous.

Order tracking built for logistics, not deliveries. Orders live in operational states — To Pick, Picked — organised by transaction ID, unit counts, and SKU counts per order, because fulfilment here means trucks at a warehouse dock, not a courier at a doorstep.

Learnings

• B2B UX is decision-support design. The job is compressing a multi-variable trade decision into one legible object — density done right is a feature
• In marketplaces, money UX is product UX. A trustworthy ledger and visible capital utilisation do more for conversion than any promotional surface
• Scarcity must be real to be respected. Reservation timers backed by actual holds create urgency without eroding trust
• Trust at scale is infrastructure, not reputation — verification, compliance, and masked identities are what let large sums move between strangers
• Study the incumbent, then design for the behavior it ignores. The competitive teardown revealed less what to copy than where traders were being made to shop instead of trade