How to add analytics to Python

This guide walks you through adding server-side analytics to any Python application. You'll install the OpenPanel SDK, configure it with your credentials, track custom events, and identify users.

Server-side tracking gives you complete control over what data you collect and ensures events are never blocked by browser extensions or ad blockers. The Python SDK works with Django, Flask, FastAPI, and any other Python framework or script.

PrerequisitesCopy Anchor Link

• A Python project
• An OpenPanel account
• Your Client ID and Client Secret from the dashboard

Install the SDKCopy Anchor Link

Start by installing the OpenPanel package from PyPI.

pip install openpanel

If you're using Poetry, you can run poetry add openpanel instead.

Initialize OpenPanelCopy Anchor Link

Create a shared module for your OpenPanel instance. This approach lets you import the same configured instance throughout your application.

# lib/op.py
import os
from openpanel import OpenPanel

op = OpenPanel(
    client_id=os.getenv("OPENPANEL_CLIENT_ID"),
    client_secret=os.getenv("OPENPANEL_CLIENT_SECRET")
)

Server-side tracking requires both a client ID and client secret for authentication. Add these to your environment variables.

# .env
OPENPANEL_CLIENT_ID=your-client-id
OPENPANEL_CLIENT_SECRET=your-client-secret

You can also pass global properties during initialization. These properties are included with every event automatically.

op = OpenPanel(
    client_id=os.getenv("OPENPANEL_CLIENT_ID"),
    client_secret=os.getenv("OPENPANEL_CLIENT_SECRET"),
    global_properties={
        "app_version": "1.0.0",
        "environment": os.getenv("ENVIRONMENT", "production")
    }
)

Track eventsCopy Anchor Link

Use the track method to record events. The first argument is the event name, and the second is an optional dictionary of properties.

from lib.op import op

# Track a simple event
op.track("button_clicked")

# Track with properties
op.track("purchase_completed", {
    "product_id": "123",
    "price": 99.99,
    "currency": "USD"
})

When tracking events in request handlers, you'll typically pull data from the request and track it alongside your business logic. Here's an example in a Django view.

from lib.op import op

def signup_view(request):
    if request.method == 'POST':
        email = request.POST.get('email')
        name = request.POST.get('name')
        
        user = create_user(email, name)
        
        op.track("user_signed_up", {
            "email": email,
            "source": "website"
        })
        
        return JsonResponse({"success": True})

The same pattern works in Flask and FastAPI. Import your OpenPanel instance and call track wherever you need to record an event.

You can also track events from background tasks. This is useful for monitoring async jobs, email delivery, and scheduled tasks.

from celery import shared_task
from lib.op import op

@shared_task
def send_email_task(user_id, email_type):
    try:
        send_email(user_id, email_type)
        op.track("email_sent", {
            "user_id": user_id,
            "email_type": email_type
        })
    except Exception as e:
        op.track("email_failed", {
            "user_id": user_id,
            "error": str(e)
        })

Identify usersCopy Anchor Link

The identify method associates a user profile with their ID. Call this after authentication to link subsequent events to that user.

from lib.op import op

def login_view(request):
    user = authenticate_user(request)
    
    op.identify(user.id, {
        "firstName": user.first_name,
        "lastName": user.last_name,
        "email": user.email,
        "tier": user.plan
    })
    
    op.track("user_logged_in", {"method": "email"})
    
    return JsonResponse({"success": True})

To track an event for a specific user without calling identify first, pass the profile_id parameter to the track method.

op.track("purchase_completed", {
    "product_id": "123",
    "amount": 99.99
}, profile_id="user_456")

You can increment numeric properties on user profiles. This is useful for counters like login count or total purchases.

op.increment({
    "profile_id": "user_456",
    "property": "login_count",
    "value": 1
})

Verify your setupCopy Anchor Link

Run your Python application and trigger a few events. Open your OpenPanel dashboard and navigate to the real-time view. You should see your events appearing within seconds.

If events aren't showing up, check that your client ID and client secret are correct. Server-side tracking won't work without the client secret. Review your application logs for any error messages from the SDK.

Next stepsCopy Anchor Link

The Python SDK reference covers additional configuration options like event filtering and disabling tracking. If you're also tracking client-side events, you might want to read about cookieless analytics to understand how OpenPanel handles privacy without cookies.