How to Automate Abandoned Cart Email Sequences

I built this to automate abandoned cart email sequences that adapt based on cart value. A $20 cart should not get the same treatment as a $500 cart. This adjusts urgency, offers, and follow-up count automatically.

Value-based abandoned cart sequences recover more revenue than one-size-fits-all approaches.

What You Need Before Starting

• Python 3.8+ with requests
• Anthropic API key
• Ecommerce API (Shopify, WooCommerce)
• ESP with trigger capability

Step 1: Capture Abandoned Carts

import requests
import os

def get_abandoned(shop_url, token):
    url = f"https://{shop_url}/admin/api/2024-01/checkouts.json"
    headers = {"X-Shopify-Access-Token": token}
    checkouts = requests.get(url, headers=headers).json().get("checkouts", [])
    return [c for c in checkouts if not c.get("completed_at")]

Step 2: Segment by Value

def segment_cart(cart):
    total = float(cart.get("total_price", 0))
    if total >= 200: return "high"
    if total >= 50: return "medium"
    return "low"

CONFIGS = {
    "high": {"emails": 5, "offer": "10% off", "timing": [1, 4, 24, 48, 72]},
    "medium": {"emails": 3, "offer": "5% off", "timing": [1, 24, 48]},
    "low": {"emails": 2, "offer": "free shipping", "timing": [4, 24]},
}

Step 3: Generate Cart Emails

import anthropic
from dotenv import load_dotenv

load_dotenv()
client = anthropic.Anthropic()

def generate_cart_email(cart, segment, email_num):
    config = CONFIGS[segment]
    items = ", ".join([i["title"] for i in cart.get("line_items", [])])
    message = client.messages.create(
        model="claude-sonnet-4-20250514",
        max_tokens=512,
        system=f"Abandoned cart email #{email_num}/{config['emails']}. Offer: {config['offer']}. Conversational, not pushy.",
        messages=[{"role": "user", "content": f"Items: {items}. Total: ${cart['total_price']}."}]
    )
    return message.content[0].text

Step 4: Schedule the Sequence

def schedule_sequence(cart, segment):
    config = CONFIGS[segment]
    for i, hours in enumerate(config["timing"]):
        email = generate_cart_email(cart, segment, i + 1)
        queue_send(cart["email"], email, delay_hours=hours)

What to Build Next

Add real-time monitoring so the sequence triggers within minutes of abandonment. The first hour has the highest recovery rate.

Related Reading

• Prompt: Write a Product Launch Email Sequence - how AI transforms email marketing operations
• AI for Email Marketing Automation - how AI transforms email marketing operations
• From Overwhelmed to Automated - framework for deciding what to automate first