E-commerce

    Support Agent for a Fashion E-commerce Brand

    A conversational agent integrated into Shopify that resolves 73% of support queries without human intervention, increasing customer satisfaction by 45%.

    Back to case studies
    RAG
    Shopify
    E-commerce
    Support

    ModaVerde SL

    January 2026

    Challenge

    The support team was receiving 1,200 tickets per month, with an average response time of 18 hours. Seventy percent of the questions were repetitive order-status, return, and sizing requests.

    Solution

    We deployed a RAG agent trained on the company knowledge base and connected in real time to the Shopify API for order lookups. The agent escalates to a human only when it detects frustration or a complex case.

    Results

    73%
    Autonomous resolution
    Tickets resolved without human intervention
    <2min
    Response time
    Average agent response vs. previous 18h
    +45%
    CSAT
    Increase in customer satisfaction
    €4,800
    Monthly savings
    Reduction in support costs

    Architecture

    1

    Chat widget

    Embedded Shopify interface with brand styling

    React
    Shopify App
    WebSocket
    2

    RAG pipeline

    Knowledge base context retrieval

    Pinecone
    OpenAI Embeddings
    LangChain
    3

    Reasoning agent

    Decide whether to answer or query the Shopify API

    GPT-4o
    Function Calling
    ReAct
    4

    Shopify connector

    Real-time order and status lookup

    Shopify Admin API
    GraphQL
    5

    Smart escalation

    Frustration detection and handoff to a human

    Sentiment Analysis
    Zendesk API

    Want to build something similar?

    If this case sounds familiar, we can help you define a version adapted to your business.

    Talk about my case