Google Cloud Platform Case Studies Arpeely: Scaling an Innovative Data Science Platform Globally with a Small Local Team

Edit This Case Study Record

	Arpeely: Scaling an Innovative Data Science Platform Globally with a Small Local Team Google Cloud Platform

Arpeely: Scaling an Innovative Data Science Platform Globally with a Small Local Team

Google Cloud Platform

Technology Category	Analytics & Modeling - Machine Learning Infrastructure as a Service (IaaS) - Cloud Computing
Applicable Industries	Education Equipment & Machinery
Applicable Functions	Sales & Marketing
Use Cases	Predictive Maintenance Traffic Monitoring
Services	Cloud Planning, Design & Implementation Services Training
Challenge	Arpeely, an Israeli ad-tech startup, aimed to revolutionize the media-buying process by leveraging machine learning and feature engineering techniques. The company sought to process billions of ad impressions daily and cherry-pick traffic based on in-app or post-conversion behavior KPIs. However, as a bootstrapped startup launched in 2017, Arpeely faced the challenge of managing global ad operations with a small team. The company needed a solution that would allow it to scale up quickly without having to invest heavily in developing complex services or expanding its team. Read More
About Customer	Arpeely is a data science startup based in Israel that uses machine learning and feature engineering to discover hidden opportunities in online advertising. The company processes dozens of billions of predictions daily and cherry-picks traffic based on in-app or post-conversion behavior KPIs. Arpeely is connected to the world’s largest advertising exchanges and achieved multimillion-dollar revenues in its first year of trading. By now, every single user in the U.S. has passed through Arpeely’s servers at some point, and it processes 20 billion ad impressions a day, while delivering millisecond predictions per ad view. Read More
Solution	Arpeely turned to Google Cloud to launch and scale its innovative ad-tech platform. Google Cloud provided the necessary tools and support for Arpeely to build and scale its operations, leveraging compute, data, and machine learning solutions. Arpeely started with App Engine, which allowed it to quickly iterate in the early days. As volumes scaled up, Arpeely moved on to the fully managed Google Kubernetes Engine (GKE) and BigQuery. BigQuery became the heart of Arpeely’s data warehouse, aggregating all its analytics, business metrics, and third-party integrations. The system automatically scaled up to deploy several hundred nodes as demand grew. Meanwhile, AutoML scaled the training of the machine learning models that Arpeely created, allowing it to continually fine-tune its bidding systems. Read More Log in to view content
Contents

Technology Category

Analytics & Modeling - Machine Learning

Infrastructure as a Service (IaaS) - Cloud Computing

Applicable Industries

Education

Equipment & Machinery

Applicable Functions

Sales & Marketing

Use Cases

Predictive Maintenance

Traffic Monitoring

Services

Cloud Planning, Design & Implementation Services

Training

Challenge

Arpeely, an Israeli ad-tech startup, aimed to revolutionize the media-buying process by leveraging machine learning and feature engineering techniques. The company sought to process billions of ad impressions daily and cherry-pick traffic based on in-app or post-conversion behavior KPIs. However, as a bootstrapped startup launched in 2017, Arpeely faced the challenge of managing global ad operations with a small team. The company needed a solution that would allow it to scale up quickly without having to invest heavily in developing complex services or expanding its team.

About Customer

Arpeely is a data science startup based in Israel that uses machine learning and feature engineering to discover hidden opportunities in online advertising. The company processes dozens of billions of predictions daily and cherry-picks traffic based on in-app or post-conversion behavior KPIs. Arpeely is connected to the world’s largest advertising exchanges and achieved multimillion-dollar revenues in its first year of trading. By now, every single user in the U.S. has passed through Arpeely’s servers at some point, and it processes 20 billion ad impressions a day, while delivering millisecond predictions per ad view.

Solution

Arpeely turned to Google Cloud to launch and scale its innovative ad-tech platform. Google Cloud provided the necessary tools and support for Arpeely to build and scale its operations, leveraging compute, data, and machine learning solutions. Arpeely started with App Engine, which allowed it to quickly iterate in the early days. As volumes scaled up, Arpeely moved on to the fully managed Google Kubernetes Engine (GKE) and BigQuery. BigQuery became the heart of Arpeely’s data warehouse, aggregating all its analytics, business metrics, and third-party integrations. The system automatically scaled up to deploy several hundred nodes as demand grew. Meanwhile, AutoML scaled the training of the machine learning models that Arpeely created, allowing it to continually fine-tune its bidding systems.

Impact #1	The use of Google Cloud has allowed Arpeely to manage global ad operations with a small team. The company has been able to keep its team lean despite its global reach by testing large matrices of machine learning models on live traffic using GKE and Cloud ML Engine. The lean infrastructure built on Google Cloud has enabled Arpeely to concentrate on its unique value proposition, with many Google Cloud tools running quietly in the background. As a result, Arpeely has been able to grow its business significantly without having to expand its team proportionally.

Impact #1

The use of Google Cloud has allowed Arpeely to manage global ad operations with a small team. The company has been able to keep its team lean despite its global reach by testing large matrices of machine learning models on live traffic using GKE and Cloud ML Engine. The lean infrastructure built on Google Cloud has enabled Arpeely to concentrate on its unique value proposition, with many Google Cloud tools running quietly in the background. As a result, Arpeely has been able to grow its business significantly without having to expand its team proportionally.

Benefit #1	Arpeely achieved multimillion-dollar revenues in its first year of trading.
Benefit #2	The company processes 20 billion ad impressions a day.
Benefit #3	Arpeely is on track to earn millions of dollars per month.

Benefit #1

Arpeely achieved multimillion-dollar revenues in its first year of trading.

Benefit #2

The company processes 20 billion ad impressions a day.

Benefit #3

Arpeely is on track to earn millions of dollars per month.

Download PDF Version

Overview

Arpeely: Scaling an Innovative Data Science Platform Globally with a Small Local Team

Operational Impact

Quantitative Benefit