Case Study: Peer39

Using Upsolver to Contextualize Billions of Pages for Targeting and Analytics

INDUSTRY: Ad targeting and digital marketing
SOLUTION: Page-level intelligence under GDPR/CCPA compliance
TECHNICAL USE  CASE: Ad targeting and digital marketing

Peer39 is an innovative leader in the ad and digital marketing industry that provides page level intelligence for targeting and analytics. Each day, Peer39 analyzes over 450 million unique webpages holistically to contextualize the true meaning of the page text/topics. Their products allow advertisers to optimize their spend while placing ads in the right place, at the right time, with the right audience while complying to rigorous privacy regulations such as GDPR/CCPA.

Peer39 had been using IBM Netezza for the past 10 years, and as a result they faced many limitations. They had to endure a long development time because the traditional MPP system was so rigid. As the system approached end of life support, Peer39 decided to reevaluate their technology stack. Their legacy process and technology stack presented many challenges:

  • Limited data availability: onboarding any new data was nearly impossible due to incompatibility with other systems.
  • Lack of business accuracy: inconsistent data being produced and the inability to perform proper data cleansing resulted in poor analytics and intelligence.
  • Lost business agility: this can lead to excessive or insufficient targeted ads, with significant business implications.

Upsolver provided Peer39 with a cloud-native, data stream processing platform with an easy-to-use UI that enabled Peer39 to deploy a modernized data stack in the cloud on time and under budget.

90%

reduction in Athena query latency

17B

events processed daily

20X

better Time-to-Analytics vs. Spark

  • Upsolver compute engine: the team dramatically improved the performance of ad targeting and digital marketing campaigns.
  • Automated and scalable workflow: utilizing built-in back end platform automation and orchestration, Peer39 easily achieved reliable data transformation and delivery that enabled advertisers to substantially reduce spending thanks to S3 storage optimization.
  • 90% reduction in Athena query latency with Upsolver’s out-of-box features such as compaction, queuing, guaranteed message delivery, and optimization.

Upsolver is unifying teams across data scientists, analysts, data engineers and traditional DBAs, enabling Peer39 to speed go-to-market with existing staff. After deciding to move to the Cloud, it was important for Peer39 to compare different solutions, including Spark. Peer39 started seeing challenges with Spark in the POC phase:

  • Significant risk: Spark required Peer39 to write a considerable amount of customized code that put deadline at risk. Ultimately, it became a build vs. buy decision for Peer39’s R&D leadership team.
"It was a classic decision to build vs. buy. With Spark, we need to build and understand every technical detail and Upsolver worked out-of-the-box."
  • True cost of ownership: finding quality Spark engineers is expensive and difficult. Upsolver is the perfect fit for the skills of the existing staff, who possess extensive SQL database (IBM Netezza) experience.
  • Scalability: onboarding new publishers and data providers within minutes instead of weeks.
  • Frequent changes: every small implementation can largely impact the end result, in turn resulting in long development cycles. It was a risk Peer39 was unwilling to take.
  • 20X better time-to-analytics vs. Spark.
  • Millions of dollars saved.
  • 17 billion events processed daily.
“I chose Upsolver because time-to-analytics over Amazon S3 is 20X faster compared to Spark. Our existing staff deployed a production-ready solution within one month, which eliminated the risk of not being able to replace IBM Netezza on schedule.”