How to Create Automated Review Request Campaigns

When you automate review request campaigns via email and sms, you stop relying on customers to remember. I build these with a key twist: only ask customers who had a positive experience. Sending review requests to everyone is how you get negative reviews you could have avoided.

The system triggers after a successful transaction, filters by satisfaction signal, and sends a personalized request at the optimal time.

What You Need Before Starting

• Customer transaction data with email and phone
• Python 3.8+ with SMTP and SMS capabilities
• A satisfaction signal (CSAT score, support ticket status, repeat purchase)
• Direct review links for Google, Yelp, or Facebook

Step 1: Build the Trigger Logic

Only request reviews from happy customers:

def should_request_review(customer_id):
    recent_tickets = get_recent_tickets(customer_id, days=30)
    negative_tickets = [t for t in recent_tickets if t["sentiment"] == "negative"]
    if negative_tickets:
        return False

    csat = get_latest_csat(customer_id)
    if csat and csat < 4:
        return False

    already_requested = check_recent_request(customer_id, days=90)
    if already_requested:
        return False

    return True

Step 2: Generate Personalized Requests

import anthropic

client = anthropic.Anthropic()

def generate_review_request(customer, transaction):
    response = client.messages.create(
        model="claude-sonnet-4-20250514",
        max_tokens=200,
        messages=[{
            "role": "user",
            "content": f"""Write a short, personal review request email.

Customer name: {customer['name']}
Service/product: {transaction['description']}
Date: {transaction['date']}
Business name: [Your Business]

Rules:
- Under 80 words
- Reference their specific purchase/service
- Make it easy (include the direct link)
- Do not beg or over-ask
- Casual, grateful tone"""
        }]
    )
    return response.content[0].text

Step 3: Set Up the Review Link Router

Route customers to the platform where you need reviews most:

REVIEW_LINKS = {
    "google": "https://g.page/r/YOUR_PLACE_ID/review",
    "yelp": "https://www.yelp.com/writeareview/biz/YOUR_BIZ_ID",
    "facebook": "https://www.facebook.com/YOUR_PAGE/reviews"
}

def get_review_link(customer_id):
    """Route to the platform with the fewest recent reviews."""
    counts = get_platform_review_counts(days=30)
    lowest = min(counts, key=counts.get)
    return REVIEW_LINKS[lowest]

Step 4: Send via Email and SMS

import smtplib
from email.mime.text import MIMEText
import requests

def send_email_request(email, name, message, review_link):
    full_message = f"{message}\n\nLeave your review here: {review_link}"
    msg = MIMEText(full_message)
    msg["Subject"] = f"How was your experience, {name}?"
    msg["From"] = "[email protected]"
    msg["To"] = email

    with smtplib.SMTP("smtp.gmail.com", 587) as server:
        server.starttls()
        server.login("[email protected]", os.getenv("EMAIL_PASSWORD"))
        server.send_message(msg)

def send_sms_request(phone, name, review_link):
    message = f"Hi {name}, thanks for choosing us! If you have a moment, we would love your feedback: {review_link}"
    # Send via Twilio or your SMS provider
    requests.post("https://api.twilio.com/2010-04-01/Accounts/{sid}/Messages.json",
        auth=(os.getenv("TWILIO_SID"), os.getenv("TWILIO_TOKEN")),
        data={"To": phone, "From": os.getenv("TWILIO_NUMBER"), "Body": message})

Step 5: Schedule and Track

from datetime import datetime, timedelta

def schedule_review_requests():
    """Run daily. Find eligible customers from yesterday's completed transactions."""
    yesterday = (datetime.now() - timedelta(days=1)).strftime("%Y-%m-%d")
    transactions = get_completed_transactions(yesterday)

    sent = 0
    for txn in transactions:
        customer = get_customer(txn["customer_id"])
        if should_request_review(customer["id"]):
            review_link = get_review_link(customer["id"])
            message = generate_review_request(customer, txn)
            send_email_request(customer["email"], customer["name"], message, review_link)
            log_request(customer["id"], "email", review_link)
            sent += 1

    return {"sent": sent, "eligible": len(transactions)}

What to Build Next

Add A/B testing on request timing. Test sending requests 1 day, 3 days, and 7 days after the transaction. Track which timing generates the highest review completion rate for your specific business.

Related Reading

• AI for Email Marketing Automation - email automation patterns for review campaigns
• The Feedback Loop That Powers Everything - review requests closing the customer feedback loop
• The Measurement Framework That Actually Works - measuring review campaign effectiveness