Jump to content

How I Automated a Full SaaS Workflow with Nothing but Python Scripts

From JOHNWICK

From user onboarding and database updates to email reports and billing — here’s how I replaced entire microservices using Python, a few APIs, and a lot of sleep deprivation.

1. Why I Ditched No-Code and Wrote Python Instead Let me be blunt: no-code tools are great until you hit a wall. For my SaaS side project, I started with Zapier, Airtable, and Retool. But things got messy. Workflows failed silently. Debugging was a nightmare. I needed precision. That’s when I replaced every moving part with Python. Now, my SaaS operations run through a single repo:

  • Signup → database insert
  • User welcome → transactional email
  • Daily analytics → Slack report
  • Stripe billing → webhook processor
  • Admin dashboard → Streamlit app

Everything. In. Python. Let’s break it down.

2. User Signup → PostgreSQL Insert + Welcome Email As soon as a user signs up through the frontend, a webhook hits my Flask endpoint. Here’s how I handle it: 

from flask import Flask, request
import psycopg2
import smtplib
from email.mime.text import MIMEText

app = Flask(__name__)

@app.route('/signup', methods=['POST'])
def signup():
    data = request.json
    email = data['email']
    name = data['name']

    # Insert into PostgreSQL
    conn = psycopg2.connect("dbname=saas user=admin password=secret")
    cur = conn.cursor()
    cur.execute("INSERT INTO users (name, email) VALUES (%s, %s)", (name, email))
    conn.commit()
    conn.close()

    # Send welcome email
    msg = MIMEText(f"Welcome {name}! You're in 🚀")
    msg['Subject'] = "Welcome to Our App"
    msg['From'] = "[email protected]"
    msg['To'] = email

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

    return {"status": "ok"}

No third-party automation platform. Just a clean webhook, raw SQL, and SMTP.

3. Daily Metrics Dashboard → Slack Notification Every morning, I want to know:

  • New signups
  • Active users
  • Churned accounts
  • Stripe revenue

I wrote a Python script that queries my database and posts a summary to Slack: 

import psycopg2
import requests
from datetime import datetime, timedelta

def get_metrics():
    conn = psycopg2.connect("dbname=saas user=admin password=secret")
    cur = conn.cursor()

    cur.execute("SELECT COUNT(*) FROM users WHERE created_at >= now() - interval '1 day'")
    new_users = cur.fetchone()[0]

    cur.execute("SELECT COUNT(*) FROM users WHERE active=true")
    active = cur.fetchone()[0]

    conn.close()
    return new_users, active

def post_to_slack(new_users, active):
    msg = {
        "text": f"*Daily Report — {datetime.now().strftime('%Y-%m-%d')}*:\n• New Users: {new_users}\n• Active Users: {active}"
    }
    requests.post("https://hooks.slack.com/services/XXX", json=msg)

n, a = get_metrics()
post_to_slack(n, a)

I run this script with cron every day at 9 AM. It never fails.

4. Stripe Webhook Listener for Billing Events I didn’t want a full backend server — just a Flask app that receives Stripe events. Here’s how I listen for failed payments: 

from flask import Flask, request
import stripe
import logging

app = Flask(__name__)
stripe.api_key = "sk_test_..."

@app.route("/webhook", methods=["POST"])
def webhook():
    payload = request.data
    sig_header = request.headers.get("stripe-signature")

    try:
        event = stripe.Webhook.construct_event(
            payload, sig_header, "whsec_..."
        )
    except ValueError as e:
        return "Invalid payload", 400
    except stripe.error.SignatureVerificationError:
        return "Invalid signature", 400

    if event["type"] == "invoice.payment_failed":
        email = event["data"]["object"]["customer_email"]
        logging.warning(f"Payment failed for {email}")
        # Notify or flag the user

    return {"status": "received"}

It’s saved my life twice. Worth every line.

6. Weekly Usage Reports → PDF via ReportLab Each Sunday, users get a PDF report of their app usage. I generate it with reportlab and send it using Gmail's SMTP. 

from reportlab.pdfgen import canvas

def generate_report(username, usage):
    filename = f"{username}_report.pdf"
    c = canvas.Canvas(filename)
    c.drawString(100, 800, f"Weekly Report for {username}")
    c.drawString(100, 750, f"Total Logins: {usage['logins']}")
    c.drawString(100, 700, f"Actions Taken: {usage['actions']}")
    c.save()
    return filename

Users love these PDFs. They make the product feel polished and premium.

7. Admin Dashboard with Streamlit I didn’t build an admin UI from scratch. I used Streamlit. 

import streamlit as st
import psycopg2
import pandas as pd

conn = psycopg2.connect("dbname=saas user=admin password=secret")
df = pd.read_sql("SELECT * FROM users", conn)

st.title("Admin Dashboard")
st.dataframe(df)

Zero frontend code. Just Python + SQL + beauty.

8. Google Sheets Sync for Financial Reports My co-founder lives in spreadsheets. So I gave her an auto-updating Google Sheet. 

import gspread
from oauth2client.service_account import ServiceAccountCredentials

scope = ["https://spreadsheets.google.com/feeds"]
creds = ServiceAccountCredentials.from_json_keyfile_name("creds.json", scope)
client = gspread.authorize(creds)

sheet = client.open("Finance Dashboard").sheet1
sheet.update("A2", [[ "2025-07-27", "$3220", "15 new users" ]])

No more copy-pasting from dashboards. It just… updates.

9. Customer Feedback Parser + Keyword Clustering We get tons of feedback via Typeform and Intercom. I built a tool that clusters them by themes: 

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans

feedback = ["Loved the UI!", "Needs dark mode", "Slow on mobile", "More integrations!"]
vectorizer = TfidfVectorizer(stop_words="english")
X = vectorizer.fit_transform(feedback)

model = KMeans(n_clusters=2)
model.fit(X)

for i, label in enumerate(model.labels_):
    print(f"Cluster {label}: {feedback[i]}")

Now I can group issues and prioritize what matters.

10. User Behavior Tracking with Session Logs Want to see which features are used the most? I log every button click and event. 

import json
from flask import Flask, request

app = Flask(__name__)

@app.route("/log", methods=["POST"])
def log_event():
    with open("session_logs.json", "a") as f:
        json.dump(request.json, f)
        f.write("\n")
    return {"status": "logged"}

I later analyze this with pandas to build heatmaps and flowcharts.

11. Task Scheduler Using APScheduler Need a function to run every hour? Don’t use cron — use this: 

from apscheduler.schedulers.background import BackgroundScheduler
import time

scheduler = BackgroundScheduler()

def hourly_task():
    print("Running hourly task...")

scheduler.add_job(hourly_task, "interval", hours=1)
scheduler.start()

while True:
    time.sleep(60)

My SaaS heartbeat. Never fails. Easy to test. Logs to file. Final Thoughts: Python Is My Backend-as-a-Service Every time I thought I needed a complex backend… I wrote a 40-line Python script instead. The beauty of Python automation is that it scales with you:

  • Start with a cron script
  • Evolve into Flask or FastAPI
  • Add Slack, Stripe, or Google Sheets integrations
  • Wrap with Streamlit for instant UI
  • Deploy with Docker if needed

If you’re building a SaaS — you don’t need to over-engineer. Just start automating. One Python file at a time.

Pro Tip:
Bundle everything into a /scripts and /services folder. Add a Makefile for easy make backup, make report, make dashboard. Suddenly your side project feels like a real product.

Read the full article here: https://medium.com/top-python-libraries/how-i-automated-a-full-saas-workflow-with-nothing-but-python-scripts-e94f370830a2

Retrieved from "https://johnwick.cc/index.php?title=How_I_Automated_a_Full_SaaS_Workflow_with_Nothing_but_Python_Scripts&oldid=2988"