How Claroty Achieved Precise Customer Cost Attribution with Attribute

The Challenge:

• Complex, multi-tenant and technologically advanced architecture
• Lack of visibility into customer-level costs
• Inability to easily and effectively allocate costs
• Limited engineering bandwidth

The Solution:

Attribute provided:

• Capturing of real-time, transaction-level data with an eBPF sensor
• Identifying customer-specific resource usage across Kafka, RDS and S3 components
• Precise cost attribution for each customer interaction
• Actionable and granular reports of customer costs across regions

The Results:

• Granular visibility into resource usage per customer
• Insights into customer profiles
• Improved forecasting accuracy
• Budget optimization for production infrastructure
• Customer target strategy refinement
• Zero disruption to operations though fast and seamless deployment

As Claroty’s customer base and cloud operations expanded rapidly, they embarked on a multi-year journey to refine theur cloud cost management strategy. Having achieved cost allocation, chargeback, and basic forecasting, Claroty’s leadership sought to elevate their approach by aligning cloud cost metrics with business growth. In a fiercely competitive market, Claroty needed deep insights into gross margin, profitability, and cost efficiency for each customer, region, and customer profile to remain agile and strategically driven.

Claroty’s state-of-the-art SaaS platform operates on a complex, multi-tenant architecture powered by Kubernetes clusters, Apache Kafka, Apache Spark, and numerous other technologies. With shared resources across services, gaining visibility into precise customer-level costs posed a formidable challenge. Claroty needed a way to allocate costs accurately across its microservices architecture without the overhead of extensive tagging or the development of resource-intensive internal tooling.

The task was further complicated by the limited bandwidth of the engineering and DevOps teams, making manual solutions or internal development impractical. After evaluating several potential solutions—both external and in-house—Claroty’s Center of Cloud Excellence concluded that no solution provided the granularity, scalability, and cost-effectiveness needed to meet their growing demands.

That’s when Attribute stepped in.

Attribute’s technology was deployed seamlessly across Claroty’s production regions, each containing multiple Kubernetes clusters, hundreds of nodes, and thousands of vCPUs. The implementation was remarkably fast and required minimal intervention from the DevOps team, ensuring zero disruption to ongoing operations.

Using Attribute’s advanced eBPF-based sensor, the solution was able to capture real-time, transaction-level data. Attribute identified customer-specific resource usage—whether it involved Kafka message streams, RDS database queries, or S3 storage operations—and provided precise cost attribution for each customer interaction.

Within days, Claroty began receiving granular, actionable reports detailing customer costs across all regions. This level of visibility enabled Claroty to break down resource usage by customer, offering insights into customer profiles, identifying more profitable verticals, and isolating high-cost users. These reports became instrumental in improving forecasting accuracy, optimizing budgeting for production infrastructure, and refining customer targeting strategies.

Looking ahead, Claroty plans to deepen its partnership with Attribute, using its technology to further refine cost attribution down to individual customer actions and feature usage. By gaining this level of insight, Claroty will enhance its ability to forecast profitability, identify efficiency opportunities, and drive strategic initiatives that improve customer satisfaction and operational efficiency.