ONE BISON

Turning a business directory into a transaction-ready marketplace

Role: UX Designer
Stage: Pre-launch commerce
Focus: Seller activation · Marketplace UX · Transaction design
Partners: Founder · Engineering · QA

One Bison had a growing network of vetted businesses, but no infrastructure for sellers to manage products or for buyers to complete a purchase. I designed the commerce experience across seller setup, product management, discovery, checkout, and post-purchase.

FROM DIRECTORY → MARKETPLACE

The system I was designing

The system I was designing

Every transaction depends on all three sides staying in sync.

SELLER

SELLER

Set up the business and fulfill orders

PLATFORM

PLATFORM

Approve, monitor, and support marketplace activity

BUYER

BUYER

Discover, evaluate, purchase, and track

PROBLEM 1

Sellers shouldn’t have to manage the same catalog twice.

Sellers shouldn’t have to manage the same catalog twice.

Many sellers already manage their products in platforms like Shopify. Recreating and maintaining the same catalog in One Bison adds unnecessary work.

I separated stable user needs from unresolved technical decisions.

WHAT WE KNEW

• Sellers needed easier product entry
• Manual entry still needed to exist
• Shopify should be optional
• CSV would support additional sellers
• Sellers should choose which products to submit
• We should reuse existing One Bison product patterns

WHAT WAS STILL UNCERTAIN

• Final integration architecture
• Some inventory-sync behavior
• Variants
• Multiple inventory locations
• Future product-feed feasibility
• Some technical connection states

DECISION

Don't wait for the perfect answer.
Design what we know and expose what we don't.

Don't wait for the perfect answer.
Design what we know and expose what we don't.

NOW

1. Product entry methods
2. Shopify connection + product selection

NEXT

3. Inventory review
4. CSV import

LATER

5. Buyer education
6. Tutorials

BLOCKED

7. Product feed

7. Product feed

SOLUTION

Choose a method & Connect Shopify

Multiple paths instead of forcing every seller into one workflow.

Multiple paths instead of forcing every seller into one workflow.

Explain the connection and keep the seller in control.

Explain the connection and keep the seller in control.

Select products or Upload CSV

Connection ≠ automatic publication

Connection ≠ automatic publication

We could move the seller experience forward without committing to technical assumptions that were still changing.

We could move the seller experience forward without committing to technical assumptions that were still changing.

PROBLEM 2

The product had multiple generations of UI.

The product had multiple generations of UI.

Older Figma files, developer-built pages, and newer e-commerce work had created inconsistent patterns across the product.

Older Figma files, developer-built pages, and newer e-commerce work had created inconsistent patterns across the product.

01

Live production UI

02

Older Figma designs

03

Previous developer-built pages

04

New e-commerce designs

Standardizing the experience without creating unnecessary rework

Not every inconsistency needs the same solution.

Not every inconsistency needs the same solution.

  1. KEEP — Internal functional patterns

  1. KEEP — Internal functional patterns

Broader status colors can help internal teams scan information quickly—as long as each color has a consistent meaning.

  1. FIX — Customer-facing drift

  1. FIX — Customer-facing drift

When the intended pattern already exists in the design system, align the implementation.

  1. DECIDE — Undocumented legacy pattern

  1. DECIDE — Undocumented legacy pattern

Determine whether the pattern should become part of the system or be phased out.

Consistency doesn’t mean making everything identical. It means making patterns intentional, predictable, and reusable.

Consistency doesn’t mean making everything identical. It means making patterns intentional, predictable, and reusable.

Consistency doesn’t mean making everything identical. It means making patterns intentional, predictable, and reusable.

PROBLEM 3

Requirements describe the transaction. Users experience everything that can go wrong.

Requirements describe the transaction. Users experience everything that can go wrong.

The main transaction was straightforward. The complexity appeared in the states around it.

The main transaction was straightforward. The complexity appeared in the states around it.

Define exception and recovery states while mapping the workflow—not after development.

Define exception and recovery states while mapping the workflow—not after development.

SOLUTION

01

Cart changed

02

Seller application / incomplete status

03

Order tracking / delivery confirmation

04

Report a problem/dispute

The successful path is only one product state.
Recovery is part of the experience too.

The successful path is only one product state.
Recovery is part of the experience too.

Pre-launch momentum

Business growth is building ahead of the marketplace launch.

Business growth is building ahead of the marketplace launch.

6×

6×

Growth in business signups

Growth in business signups

+32%

+32%

Growth in engaged sessions

Growth in engaged sessions

+45%

+45%

Growth in search usage

Growth in search usage

What commerce success means
Seller activation → published inventory → checkout conversion → completed orders

These are company-level pre-launch signals, not post-launch commerce results.
They showed that One Bison was entering launch with growing seller and buyer interest—and made activation, discovery, and transaction readiness increasingly important.

Where AI could reduce marketplace friction next

With the underlying commerce flows established, I explored where AI could reduce repetitive work without taking consequential decisions away from users.

With the underlying commerce flows established, I explored where AI could reduce repetitive work without taking consequential decisions away from users.

Seller activation

Seller activation

→ AI-assisted listing

→ AI-assisted listing

Buyer discovery

Buyer discovery

→ intent-aware search

→ intent-aware search

Post-purchase

Post-purchase

→ AI support + human escalation

→ AI support + human escalation

What I'd do differently

Instrument earlier

Establish baseline events before redesigning critical funnels.

Validate with sellers sooner

Bring sellers into onboarding tests earlier instead of relying primarily on stakeholder knowledge.

Model marketplace complexity earlier

Account for multi-seller orders, fulfillment, and post-purchase states sooner in the transaction architecture.

Next case study

Repositioning a niche e-commerce brand for growth

Repositioning a niche e-commerce brand for growth

Repositioning a niche e-commerce brand for growth

View Case Study

View Case Study