case study

Consumer Data Platform

Media & Entertainment

Business Impacts

5M

Web URLs Were Scraped Daily for Content

Automated

Insights Generation

Customer Key Facts

  • Location : North America
  • Industry : Media & Entertainment

Problem Context

The customer, a Media Consulting company, wanted to structure and analyze large unstructured datasets arising from brand and asset marketing. They wanted to provide direction to their partners on creative, media, messaging, and communication strategies for digital ad campaigns.

Challenges

 

  • Storing at least 14 TB of proprietary data coming from 11 different data sources
  • Proprietary data was unclean and had not been validated for over a year
  • Processing daily data of 2 billion records

Technologies Used

AWS Glue

AWS Glue

AWS Lambda

AWS Lambda

Amazon S3

Amazon S3

Amazon QuickSight

Amazon QuickSight

Amazon Athena

Amazon Athena

AWS Elastic BeanStalk

AWS Elastic BeanStalk

Developed an End-to-End Consumer Data Platform for Analysis

Solution

Quantiphi developed an end-to-end platform to inject multiple data sources into AWS Cloud and created a data lake on Amazon S3. Multiple different data sources were combined and transformed using AWS Glue to create a data-ready platform for analysis. An automated solution was also developed, that was able to scrape content from more than 5M web URLs daily to identify the content preference of users and generate insights on user behavior.

Result

  • Enables the creation of consumer pools
  • Helps generate key insights into consumer behavior, such as application usage, purchase behavior, and media consumed in real-time

Start Your Next Gen AI Journey Today

Discover how Quantiphi’s AI-powered solutions can transform your business. Fill out the form, and we’ll help you explore tailored AI strategies to unlock new opportunities for growth.

Thank you for reaching out to us!

Our experts will be in touch with you shortly.

In the meantime, explore our insightful blogs and case studies.

Something went wrong!

Please try it again.

Share