Challenge

Thrasio had a strategy to grow with breakneck speed by evaluating more than 5,000 FBA (Fulfillment by Amazon) companies and acquiring dozens of top-rated brands.  This would result in the need to manage thousands of category-leading products.

Thrasio had started this without a technology and data platform - managing a majority of their processes manually.

Proxet was brought in to implement a greenfield project.

Solution

Designed and implemented scalable infrastructure:

• AWS, Kubernetes, Airflow, Terraform, Databricks 
• Thousands of containers
• Hundreds of Airflow pipelines

Build ability to collect performance metrics and raw pricing data from multiple online marketplaces such as Amazon seller central, Walmart, Shopify, direct store sources, and proxy vendor API targets such as Rainforest, Fivetran, Rakuten, etc -  scraping tens of millions of products and SERP (Search Engine Results Page) results daily.

Designed and implemented a brands and product metadata management platform. 

Created broad organization visibility by generating analytics from the collected data, and implement advanced BI solutions with 100s of relevant dashboards .

We transitioned an Ad-Hoc research process with scattered (and unclean) data, into standardized processes, with structured data and built multiple ML models to allow for quantitative decision making.  This solution started with AWS SageMaker and transitioned to Databricks.

Outcome

• Built a brands and product metadata management platform.  Provided Thrasio with a reliable data-driven foundation for decisioning, investment and operational support.
• Allowed M&A, Marketing, SP&A (Strategic Planning & Analysis), and Brand Management teams to utilize a single source of Truth. 
• Enabled Thrasio to proactively manage and expand its portfolio of brands at scale, acquire 100s of brands and manage 14k products, achieving Unicorn status in under 2 years ( fastest in US history at the time).