Case study · Commerce
Making a mid-market e-commerce brand shoppable inside AI assistants
Client: Mid-market e-commerce brand (anonymised)
Challenge
Where they started.
Strong catalog and APIs, but invisible when shoppers asked ChatGPT, Claude, Gemini, Meta AI or Perplexity for product recommendations. Zero revenue from AI channels.
What the FDE team did
Inside their stack.
- Built an MCP server over catalog, inventory, pricing, cart and order-status APIs
- Published agent-ready product feeds and schema markup
- Implemented llms.txt and AEO content
- Enabled agentic checkout flows
- Added AI-channel attribution to analytics
- Tested discoverability across ChatGPT, Claude, Gemini, Meta AI, Grok, Perplexity and Cursor
Architecture
How it fits together.
Outcomes · illustrative placeholders
The scoreboard.
[X]%
Share of AI-assistant answers for target product queries
Illustrative · placeholder
[X]%
New revenue from AI channels within 90 days
Illustrative · placeholder
[X] wks
Time to first AI-driven order
Illustrative · placeholder
ChatGPTClaudeGeminiMeta AIPerplexityGrokCursor
[Client quote placeholder — awaiting approval]
Stop building technology. Start engineering results.
Start with an outcome audit. We’ll map the systems, measures and FDE pod needed to move from AI ambition to production impact.