Starzdata Builds a Scalable Data Activation Platform with OVHcloud — for ¼ the Cost

Key Takeaways

Cost Predictability: For usage-based business models, OVHcloud’s infrastructure cost predictability supports financial planning and service pricing.
Latency Optimization: In data-intensive operations, OVHcloud’s strategic placement of computing resources can dramatically improve performance and service delivery times.
Vendor Independence: Building on OVHcloud’s open-source stacks rather than proprietary services provides greater flexibility and reduces lock-in risks.
Regional Considerations: Using OVHcloud’s Local Zones can provide both technical and market advantages when serving clients in specific regions.

Company Overview

StarZData is a French data platform that enables organizations to connect and activate distributed datasets using external APIs and curated data sources. Built for marketing, strategy, and commercial efficiency use cases, Starzdata simplifies the orchestration of market segmentation, data extraction, and
CRM enrichment. The platform supports both custom datasets and third-party intelligence via a usage-based model aligned with real business needs. The platform supports a modular approach: Clients can use Starzdata to store and enrich their own datasets with near-zero costs, and only consume credits when connecting to external providers.

Challenge

As a platform that connects internal datasets to external providers and APIs, Starzdata required a cloud architecture capable of supporting unpredictable workflows. The company initially turned to a large hyperscale cloud provider, offering serverless functions and managed services.

However, they soon encountered several significant challenges:

•   Unpredictable and Escalating Infrastructure Costs
Despite low or inconsistent traffic, server costs remained high and fluctuated unpredictably, undermining financial forecasting and creating budget volatility.
•   Vendor Lock-in
Reliance on proprietary cloud services restricted architectural flexibility and made migrating or adapting to new solutions costly and time-consuming.
•   Inflexible Cost Structure Limiting Business Innovation
Starzdata’s usage-based pricing model required equally predictable backend costs. The existing infrastructure made it difficult to bundle features into a credit-based system, stalling product innovation.
•   Performance Bottlenecks Impacting Cost and Delivery
Large-scale data enrichment and generative AI workflows involving millions of records were slowed by latency issues. These delays directly increased compute time and, consequently, operational costs.
•   Lack of Responsive Customer Support Increased Downtime
With no access to real-time human support, the Starzdata team spent excessive hours debugging avoidable issues, diverting valuable engineering resources from core product development.

Solution

Initially, Starzdata moved from their prior vendor to a combination of OVHcloud for front-end, back-end, and core platform services; and another cloud provider platform for data storage, key management systems, data processing, and job activation.

To address latency issues when working with data providers, Starzdata leveraged OVHcloud’s Local Zones to position computing resources closer to data providers to reduce network latency for API calls and improve batch processing performance for large data sets.

Starzdata first utilized three primary OVHcloud services:

•   Compute
For processing power and running their platform
•   Block Storage
For structured data storage needs
•   Object Storage
For storing data sets in Parquet and CSV formats

The company strategically deployed instances in OVHCloud Local Zones to minimize latency between their infrastructure and data providers. This was particularly crucial for batch processing operations where even small latency improvements could significantly impact overall processing time and cost.

Starzdata also began relying on OVHcloud’s services to create separate instances for each customer to address sensitive data management constraints, such as data sovereignty and confidentiality. The company also works with strategy consulting firms that need to manage sensitive information like CRM records and financial data, making OVHcloud’s scalable and flexible services particularly supportive to their daily processes.

Results

Starzdata’s migration to OVHCloud delivered substantial benefits, far beyond cost reduction. By moving core processing workloads to OVHcloud, the team gained the visibility and control required to support a more transparent, consumption-based business model.

Specifically, OVHcloud’s predictable pricing allow Starzdata to separate storage and compute costs from external data usage, enabling a model where clients only consume credits when activating external data, while enriching internal datasets at minimal or no cost.

Key impacts included:

• 75% Cost Savings, Predictable Pricing, and Operational Efficiencies
StarZData reduced infrastructure costs to one-quarter of their previous expenses and gained the ability to forecast costs accurately using simple spreadsheet calculations. The company now has clear visibility into infrastructure expenses, without the confusion or surprise of unpredictable network egress costs. Crucially, StarZData can now project how costs might scale with business growth. “With OVHcloud, we're able to predict how the costs are going to evolve. ... As a startup founder, I'm not handling a multi-million-dollar turnover company. I still keep a close eye on how costs are evolving from one month to the other. I can do that with OVHcloud. It's crystal clear,” says StarZData founder Mathieu Colas.

•   Predictable Forecasting
With clear control over network, compute, and storage costs, Starzdata can directly integrate infrastructure costs into client-facing pricing.
•   Accelerated Processing Through Local Zones
Latency reductions of up to 90% on key API workflows allowed Starzdata to cut large data enrichment jobs from 2 days to under 4 hours.
•   Sensitive Use Cases Supported
With dedicated instances per client, Starzdata can now meet the data protection expectations of financial institutions, audit firms, and strategic consultancies.

The Future: Full OVHcloud Integration

StarZData's migration demonstrates how strategic cloud infrastructure decisions can not only reduce costs but also enable new business capabilities and improve service delivery.

The company is now planning a complete migration to OVHcloud to further optimize costs and reduce dependency on proprietary services. OVHcloud’s infrastructure is also powering StarZData’s ability to develop new AI capabilities for their platform. They also plan to leverage OVHcloud’s Local Zones for additional use cases, particularly for clients in the financial sector where latency is critical.

Ready to Get Started?