Skip to the content

GraphQL scales and improves Netflix ad operations

As Netflix grew to cover 190 countries, their marketing team was faced with creating even more advertising that needed to be localised.

Imagine the same team tasked to create and quality control (QC) more ads, tailored for multiple concepts, A/B tests and formats, to be delivered correctly to billions of consumers.

The same ad with a different variable, such as language, could produce over 30 different ad variants; see below example of a US ad (left) localised for Japan (right).

Ad localised for Netflix

This challenge galvanised the Marketing Tech team at Netflix to create Monet, a dynamic ad creation and configuration platform built with Java and Groovy-based microservices.

Initially, Monet's React UI layer accessed traditional REST APIs powered by an Apache Tomcat server. However, use cases became more complex over time, and simple pages were drawing in so much data they caused network bandwidth bottlenecks.

To solve this issue, Netflix considered Falcor (Javascript library) as a possible solution, but in the end opted for a robust GraphQL (API query language) ecosystem with powerful third party tooling.

Post-implementation, core GraphQL benefits experienced by Netflix include:

  • Redistributing load and payload optimisation
  • Reusable abstractions
  • Chaining type systems
  • Improved developer user interface and experience

Read more about Netflix’s GraphQL solution

 

 

By Jo Lo
Jo Lo

About the author

Jo Lo

Jo is a digital producer at Pattern. Jo has a background in communication, and has worked as a producer and editor for organisations such as Les Mills International, Tailor, Auckland Libraries and XtraMSN/Telecom. From 2007-2009, Jo was the web editor for Taste and Metro magazines.