How Erewhon Automated 70% of Customer Service Tickets and Saved $40K Annually with Zapier
Erewhon, a luxury grocery chain with 10 stores across Los Angeles, deployed a 39-step Zapier workflow connecting HelpScout, ChatGPT, a vector knowledge store, and BigQuery to automate customer service across its locations. Built by a single self-taught employee, the system automates 70% of tickets without human modification, saves 1,500 labor hours per year, and delivers $40K in annual savings—representing a 5.5x return on Zapier investment.
Impact
70%
Customer service tickets automated
$40K
Annual labor savings
1,500 hours
Labor hours saved per year
5.5x
ROI on Zapier investment
Challenge
Erewhon’s 10 Los Angeles stores each operated separate HelpScout inboxes with no shared institutional knowledge or member prioritization, requiring staff to spend 20+ hours per week on repetitive manual ticket responses with inconsistent quality across locations.
Solution
A 39-step Zapier workflow connecting HelpScout, ChatGPT, a vector knowledge store, and BigQuery was deployed to check membership status, draft personalized responses from institutional knowledge, surface purchase history, and route tickets—automating 70% of CS responses without human modification.
Tools & Technologies
What Leaders Say
“Our company is run on two things: Vishal, our lead programmer, and Zapier. If one of those fails, we’re done.”
Sign up to read complete case studies, access detailed metrics, and unlock all use cases.
Full Story
Erewhon operates 10 high-end grocery stores across Los Angeles, each with its own customer service inbox in HelpScout. The chain had built a loyal and high-value membership base, but its customer service operation was entirely manual: each location handled every ticket independently, with no ability to prioritize premium members, maintain consistent response quality across stores, or avoid redundant work on common inquiries. Staff were spending more than 20 hours per week on repetitive responses.
The core problem was fragmentation. Ten separate inboxes meant ten separate workflows, no shared institutional knowledge, and no way to automatically identify which customers warranted elevated attention based on their spending history or membership status. As ticket volume grew with Erewhon’s expansion, the team needed a scalable solution that could personalize at scale without proportional headcount growth.
Chris Morrison, Erewhon’s Business Analyst and self-taught AI lead, built a 39-step Zap that orchestrates the entire customer service response pipeline. When a ticket arrives in HelpScout, the system checks the customer’s membership status against Erewhon’s database, routes members to a dedicated AI agent configured with a warmer tone while non-members receive standard responses, drafts a reply using a vector store of institutional knowledge about membership policies and store procedures, and surfaces the customer’s purchase history so store managers can prioritize high-spend members. To validate quality, Morrison also built a second AI—an “English Tutor” agent that grades each AI draft against the final human response, scoring how much the agent changed.
The system now handles 70% of inbound customer service tickets without any human modification to the AI-drafted response. Across 10 stores, this translates to 1,500 labor hours saved per year and approximately $40,000 in annual labor cost savings. The investment in Zapier delivers a 5.5x return.
Erewhon’s customer service automation has become a core operational infrastructure, described by Morrison as one of only two things the company’s operations depend on. The architecture demonstrates how small teams at retail businesses can deploy multi-model AI pipelines—without engineering resources—to achieve measurable, scalable outcomes.
