Dogfooding at RudderStack: Our Data Stack | RudderStack

neub9
By neub9
3 Min Read

How we Leverage RudderStack at RudderStack: A Quick Guide

Overview

At RudderStack, we take full advantage of our Event Stream and ETL pipelines to gather data from various sources like apps, websites, and cloud tools. This enables us to have a comprehensive view across every digital touchpoint and understand the user journey and profile. We stream real-time data to cloud tools and dump behavioral data to Snowflake and S3 every 30 minutes.

Customer Data Pipelines

We leverage the Event Stream for analytics, conversions, and 3rd-party SDK management, as well as ETL for cloud data to enhance analytics. Additionally, our Reverse ETL sends enriched data and audiences back to the stack.

Pro Tips from Internal Teams:

Our marketing team uses the distinct Anonymous ID from RudderStack for accurate site traffic and conversions tracking. They also find it more convenient to send .track events to ad platforms compared to using Google Tag Manager.

Our product and marketing teams use Customer.io for messaging, and having both server-side and client-side events allows them to easily build segments for triggered emails and marketing campaigns.

ETL

We use ETL to pull data from multiple sources, including Salesforce, Customer.io, Google Ads, Facebook Ads, and Stripe. This allows us to load Salesforce data into Snowflake every 24 hours and inform other tools in the stack of lead and account status changes through Reverse ETL.

Reverse ETL

Reverse ETL keeps all our tools up to date by capturing customer data within Snowflake and updating other marketing tools. This is particularly useful for managing opt-out statuses for communication channels.

Real-Time Transformations

We’ve developed user transformations to support various processes, including mapping and renaming user traits, excluding anonymous page calls from flowing to Customer.io, and enhancing events before passing them to their final destination.

Destination Transformations

Our Destination Transformations in the data plane automatically translate RudderStack payloads into the proper format for each downstream tool. This makes it seamless to create users in multiple platforms with a single identify call.

System Observability via Grafana

We monitor the health of all our pipelines through Grafana, which provides detailed insights into the performance and status of our RudderStack Cloud instance. System performance metrics are stored in InfluxDB.

Share This Article
Leave a comment

Leave a Reply

Your email address will not be published. Required fields are marked *