The Business Challenge
The business situation the organisation faced before the project began.
The restaurant handled all orders manually via phone calls, requiring staff availability during peak hours. Missed calls meant lost orders and inconsistent customer experience.
How the Solution Was Delivered
Discovery, design, development, testing and rollout — the journey, not the tooling.
- 1Designed the conversation flow and menu structure for the voice agent
- 2Built and configured the Vapi assistant with GPT-4.1 as the language model
- 3Created a custom ElevenLabs voice and connected it to Vapi
- 4Built the n8n webhook workflow to receive order data and write to Airtable
- 5Deployed the menu frontend on Cloudflare Pages
- 6Tested end-to-end call flow across multiple order scenarios
Business Outcomes & Impact
The measurable outcomes from the underlying AI Deployment — with the story behind them.
Enabled round-the-clock order taking, eliminating 100% of missed calls during unstaffed hours and reducing order processing time from 3 minutes average to under 60 seconds.
Business Outcome Categories
The story behind the numbers
Enabled 24/7 automated order taking with zero staff involvement. Eliminated missed calls during peak and unstaffed hours. Standardized the ordering experience across all customer interactions.
Lessons Learned
What surprised the team, what worked well, and what would be done differently.
- Vapi conversation flow design requires explicit fallback handling for unclear speech inputs to avoid dead ends
- ElevenLabs voice latency affects perceived call quality; selecting a lower-latency voice model improves the experience significantly
- n8n webhook timeout settings must be adjusted when Vapi call duration exceeds default limits