Skip to main content
Skip to main content
Edit this page

Breaking the Rules: Success Stories

This guide is part of a collection of findings gained from community meetups. For more real world solutions and insights you can browse by specific problem. Need tips on debugging an issue in prod? Check out the Debugging Toolkit community insights guide.

ClickHouse as Rate Limiter (Craigslist Story)​

Conventional wisdom: Use Redis for rate limiting.

Craigslist's breakthrough: "Everyone uses Redis for rate limiter implementations... Why not just do it in Redis?"

The problem with Redis: "Our experience with Redis is not like what you've seen in the movies... weird maintenance issues... we will reboot a node in a Redis cluster and some weird latency spike hits the front end"

Test rate limiting logic using ClickHouse approach:

Results: "Running untouched for nearly a year without any alert" - a dramatic improvement over Redis infrastructure.

Why it works:

  • Incredible write performance for access log data
  • Built-in TTL for automatic cleanup
  • SQL flexibility for complex rate limiting rules
  • No Redis cluster maintenance headaches

Mobile Analytics: The 7-Eleven Success Story​

Conventional wisdom: Analytics databases aren't for mobile applications.

The reality: "People out in the factory floors... people out in health care facilities construction sites... they like to be able to look at reports... to sit at a computer at a desktop... is just not optimal"

7-Eleven's breakthrough: Store managers using ClickHouse-powered analytics on mobile devices.

The use case: "The person who runs a store they're going back and forth between the stock room out to the front into the register and then going between stores"

Success metrics:

  • Daily sales by store (corporate + franchise)
  • Out-of-stock alerts in real-time
  • "Full feature capability between your phone and your desktop"

Customer-Facing Real-Time Applications​

Conventional wisdom: ClickHouse is for internal analytics, not customer-facing apps.

ServiceNow's reality: "We offer an analytic solution both for internal needs and for customers across web mobile and chatbots"

The breakthrough insight: "It enables you to build applications that are highly responsive... customer facing applications... whether they're web apps or mobile apps"

Why this breaks conventional rules:

  • Real-time customer segmentation: "Give customers the ability to real-time segments the data and dynamically slicing"
  • User expectations: "In 2024 we have been very much trained to expect a certain degree of responsiveness"
  • Retention impact: "If that repeats often enough you're either not going to come back"

Success pattern: ClickHouse's speed enables customer-facing applications with sub-second response times, challenging the notion that analytical databases are only for internal use.

The Rule-Breaking Philosophy​

Common thread: These successes came from questioning assumptions:

  • "I asked my boss like what do you think of this idea maybe I can try this with ClickHouse" - Craigslist
  • "Mobile first actually became a big part of how we thought about this" - Mobile analytics pioneers
  • "We wanted to give customers the ability to... slice and dice everything as much as they wanted" - ServiceNow

The lesson: Sometimes the "wrong" tool for the job becomes the right tool when you understand its strengths and design around them.