Optimising Snowflake for Digital i Performance and Scalability
Helping a media insights leader future-proof their Snowflake data platform with performance and cost optimisation.
Butterfly Data partnered with Digital i to optimise the performance and scalability of their Snowflake data platform, ensuring it could meet growing data demands while controlling costs.
Over a focused engagement, we analysed usage patterns, identified inefficiencies, and provided strategic recommendations tailored to their unique data workflows.
The result is a more efficient, cost-effective Snowflake environment that positions Digital i to scale confidently and deliver even greater value to their media clients.
The Challenge
Digital i collects and provides powerful viewing analytics to leading broadcasters and streaming platforms. Their insights enable clients to understand content performance, audience behaviour, and competitive positioning so they can sharpen their market strategies and negotiate better deals.
With the addition of YouTube streaming analytics, Digital i experienced a rapid surge in data volumes ingested into their Snowflake environment. As they expected to continue to scale, the team recognised the potential risk of escalating compute costs and performance bottlenecks. They needed to assess their current setup, uncover inefficiencies, and ensure their Snowflake implementation remained both high-performing and cost-effective.
Butterfly Data conducted a comprehensive review of Digital i’s Snowflake usage using Snowflake’s built-in Account Usage views. Key areas analysed included Query History, Database Storage Usage, and Warehouse Metering to highlight inefficiencies across storage and compute operations.
The metrics from these views were aggregated and visualised using SQL, Snowsight dashboards, and interactive notebooks built with Plotly. These provided clear, accessible charts to illustrate areas for optimisation, which were critical in helping to communicate findings and prioritise next steps.
Reviewing Digital i’s Snowflake Usage
Our analysis revealed that storage costs were minimal compared to compute, shifting the focus to performance-based recommendations. We identified key opportunities to streamline both client-facing and ingestion pipeline queries.
Digital i received a tailored optimisation report and walk-through presentation. Armed with actionable insights, their team is now equipped to plan and prioritise implementation for enhanced performance and to scale with confidence.