End-to-End Data Discovery and Delivery

Dataset Marketplace

MarketplaceDataE-commerceAPIB2B

A B2B data marketplace built for serious buyers: developers training AI models, analytics teams sourcing external signals, and product teams integrating third-party data into their applications.

RoleLead Product Designer
PlatformWeb Application
ToolsFigma, Prototyping
ClientConfidential
Overview

Platform Scope

The design challenge was to build a marketplace that supports discovery, technical evaluation, transparent purchasing, and flexible delivery - all without requiring users to leave the platform or guess at what they are buying.

Dataset MarketplaceSearch-first discovery
Dataset Detail PagesTechnical specs & pricing
Interactive PlaygroundLive API testing
Purchase FlowZero ambiguity checkout
Data DeliveryCloud or direct download
Core Design Principles

Evaluate before you buy

Every dataset can be previewed, tested, and understood before purchase.

Surface technical truth early

File size, format, schema, and refresh rate are visible immediately - not hidden in documentation.

Reduce time from intent to usable data

Once purchased, data is delivered in the format and destination the user chooses.

Design for scale

The platform supports single purchases and enterprise-level integrations equally well.

Layer 1: Dataset Marketplace

Search-First Discovery

The marketplace landing page is search-first: prominent search bar, fast filters by platform (Zillow, GitHub, YouTube, Instagram), and dataset cards that surface key metadata immediately.

Dataset Marketplace
Search BarFull-text search across all datasets
Filter ChipsQuick filters by platform and category
Dataset CardsFormat, size, downloads at a glance
Custom RequestContact sales for custom datasets
Layer 2: Dataset Detail Pages

Decision Making Above the Fold

Every dataset has a dedicated detail page with comprehensive metadata: size, format, record count, refresh frequency, pricing, and source attribution. The goal is to provide enough information for a purchase decision without scrolling.

Dataset Detail Light
Dataset Detail Dark

Above the Fold

  • -Dataset name and source
  • -Price ($120 one-time purchase)
  • -Data-Fields count (21)
  • -Total Records (7.4M)
  • -File size (2.4 GB)
  • -Refresh frequency (Every Month)

Action Buttons

  • -Buy Dataset - Primary CTA
  • -Try in Playground - Test before buying
  • -Share - Quick link sharing
  • -Overview / Data Preview tabs
Layer 3: Interactive Playground

Live API Testing

Each dataset has a dedicated landing page with an interactive playground. Users can test the API with real parameters, preview JSON responses, and understand the data structure before committing to a purchase.

GitHub Dataset Landing Page
Playground Section
  • - Type/Owner/Repository filters
  • - Include options (AI Data)
  • - Time range selection
  • - Live search execution
Code Generation
  • - Results & Code toggle
  • - API request preview
  • - JSON response format
  • - Credit usage tracking
Feature Grid
  • - Repository Data
  • - Developer Profiles
  • - Trending Projects
  • - Commits Analysis
Layer 4: Purchase Flow

Zero Ambiguity Checkout

The purchase modal is designed to eliminate ambiguity. It recaps exactly what the user is buying: dataset name, size, format, downloads, and what is included (full download, playground access, commercial use rights).

Buy Dataset Modal Light
Buy Dataset Modal Dark
What You Get
+ Full dataset download (CSV)+ Immediate access in Playground+ Unlimited personal & commercial use
Layer 5: Data Delivery Configuration

Flexible Cloud Delivery

Post-purchase, users choose how they want their data delivered. The platform supports JSON/CSV format selection and multiple delivery methods including Amazon S3, Google Cloud Storage, Azure Blob Storage, or direct download.

Get Dataset Modal Light
Get Dataset Modal Dark
Amazon S3Direct delivery to S3 bucket with secure access credentials
Google Cloud StorageUpload to GCS bucket with service account authentication
Azure Blob StorageTransfer to Azure storage with connection string auth
Direct DownloadCompressed ZIP package with manifest and checksums
Design Rationale

Key Design Decisions

Information Architecture

  • +Metadata-driven cards reduce evaluation time
  • +Price visible on every listing - no hidden costs
  • +Technical specs above the fold on detail pages
  • +Playground accessible before purchase

User Experience

  • +Filter chips for rapid category navigation
  • +Modal-based flows reduce context switching
  • +Clear CTAs at every decision point
  • +Dark/light mode support throughout
Outcome

The final platform enables users to discover datasets, understand what they are buying, test before committing, and receive data in their preferred format - all within a single, cohesive experience.

DiscoveryEvaluationPurchaseDelivery

End-to-end data marketplace experience built for B2B buyers.

Project TypeB2B Data Platform
sadra targhi - marketplace case study