Download PDF
Case Studies > Attraqt: Enhancing Online Shopping Experiences with Google Cloud

Attraqt: Enhancing Online Shopping Experiences with Google Cloud

Technology Category
  • Infrastructure as a Service (IaaS) - Cloud Storage Services
  • Platform as a Service (PaaS) - Application Development Platforms
Applicable Industries
  • Consumer Goods
  • Retail
Applicable Functions
  • Procurement
  • Sales & Marketing
Use Cases
  • Real-Time Location System (RTLS)
  • Track & Trace of Assets
Services
  • Cloud Planning, Design & Implementation Services
The Challenge

Attraqt, a leading expert in merchandising-led shopping experiences, was faced with the challenge of scaling its services to meet the increasing demands of its 300 plus clients. The company's platform, which combines AI with human creativity, is used by more than 50% of France’s biggest ecommerce sites. However, the platform needed to handle large volumes of API queries per second and continually optimize the potential of customer data in real time. In 2017, this volume was an average of 600 prediction requests per second, tripling to 2,000 per second in the sales season. Additionally, the platform needed to rapidly scale during the sales season or when a big client comes on board. As the platform's customer base grew, it needed to grow its ability to scale further, while reducing the amount of operational work for its teams.

About The Customer

Attraqt is a leading expert in merchandising-led shopping experiences, enabling international brands to deliver exceptional online shopping experiences by combining the power of AI with human creativity. The company serves more than 300 clients, including some of the world’s leading brands, helping them optimize their economic performance through structure, scale, and the use of smart data. Attraqt's platform is used by more than 50% of France’s biggest ecommerce sites. The company acquired French predictive marketing start-up Early Birds and its award-winning AI-driven personalization platform in 2019, thereby expanding its capabilities in the form of the Experience Orchestrator, or XO platform.

The Solution

Attraqt partnered with Google Cloud to enhance its Experience Orchestrator (XO) platform's capabilities and scalability. The initial solution was based on a virtualized environment using Compute Engine to test algorithms and analyze consumers' histories. It also used Cloud Storage for automatic backups and Cloud CDN to display personalized content relevant to the consumer on a country-by-country basis and in near real time. As the platform's customer base grew, Attraqt switched to managed services, using Pub/Sub as the central messaging service for a microservices architecture, as well as Dataproc and Dataflow for batch processing, leveraging BigQuery for its data warehouse and Cloud Bigtable for column-orientated storage. Finally, the team implemented Google Kubernetes Engine to be able to scale up quickly and release resources when no longer needed. The XO team uses managed services such as Vertex AI and Cloud Vision to develop, maintain, and evolve its own algorithm.

Operational Impact
  • The partnership with Google Cloud has enabled Attraqt to deliver cutting-edge AI capabilities and scale to meet the traffic demands created by its clients. The company has managed to differentiate itself from its competitors by helping well-known retailers understand what makes their ecommerce journeys work and how to improve them, thereby increasing their customer order values and conversion rates. The use of Google Cloud-managed services has allowed the XO team to focus on product development and innovation. Furthermore, Attraqt's clients can access all the data brought together by Attraqt through BigQuery, with connectors that are native to Google. The company is now looking forward to expanding to new markets across Europe and North America.

Quantitative Benefit
  • Handled traffic spikes smoothly, serving 2.9 billion requests over a five-day shopping peak.

  • 100% uninterrupted service over peak trading periods.

  • Managed to triple the volume of prediction requests per second from 600 to 2,000 during the sales season.

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.