How At-Bay Reduced Manual Snowflake Management with Automated Cost Optimization and End-to-End Lineage
At-Bay’s solution combines technology, insurance, and cybersecurity expertise to help businesses manage digital risk. As data usage expanded across the company, the data team needed a more efficient way to manage their Snowflake environment – reducing the time spent maintaining internal monitoring tools, investigating cost and performance issues, and assessing the impact of infrastructure changes.
With Seemore Data, At-Bay moved from manual monitoring and investigation to continuous, automated Snowflake management. Cost optimization runs in the background, while proactive alerts bring the team in only when attention is needed. When changes or investigations are required, end-to-end lineage and actionable insights help the team quickly understand dependencies, identify root causes, and move forward with confidence. The result: less manual work for the data team, continuous automated cost optimization, and greater control over a complex Snowflake environment.
86
Days added to At-Bay’s Snowflake contract runway
91.4%
Of captured savings generated by compute automations
6%
Of the Snowflake bill saved by optimizing a single query with Seemore Assistant
Challenges solved for At-Bay
Cyber insurance and security platform serving 35,000+ policyholders across 100+ industries
United States and Israel
340+ employees globally
14 data engineers
At-Bay’s data team manages a complex Snowflake environment used by teams across the organization. Before Seemore Data, maintaining visibility required the team to build and manage its own DBT dashboards and views, repeatedly analyze query history, and proactively monitor spend. At the same time, changes to its underlying data infrastructure required careful investigation to understand downstream dependencies and avoid disrupting the teams relying on them.
- Time-Consuming Manual Monitoring: The team built and maintained internal DBT dashboards, views, and analyses to track Snowflake usage, costs, and performance.
- Reactive Cost Investigation: Identifying the source of rising costs or performance issues required manually exploring query history and tracing individual workloads.
- Complex Downstream Dependencies: Changes to At-Bay’s CDC architecture could affect layers of tables, views, and business users, making it difficult to understand the impact before implementation.
- Continuous Oversight Requirements: Keeping Snowflake spend under control required proactive checks, even when everything was operating normally and no immediate action was needed.
Seemore’s compute automations continuously optimize At-Bay’s Snowflake environment in the background. Instead of repeatedly analyzing warehouse configurations and making manual adjustments, the team can rely on automations that are simple to operate, transparent, and require minimal ongoing attention.
- Always-On Optimization: Compute configurations are continuously optimized without requiring the team to monitor and adjust them manually.
- Minimal Operational Effort: The automations reduce the time previously spent analyzing warehouse usage and identifying optimization opportunities.
- Transparent Execution: At-Bay can clearly understand how the automations work and where savings are being generated.
- More Contract Runway: Automated savings have already added 86 days to At-Bay’s Snowflake contract runway.
Before Seemore, At-Bay’s team built and maintained its own DBT dashboards and views, analyzed Snowflake query history, and proactively checked whether spending remained within a reasonable range. Seemore replaced much of this manual effort with processed insights, weekly updates, and alerts that bring the team in only when attention is required.
- Less Internal Maintenance: The team no longer needs to rely on internally developed dashboards and views for ongoing Snowflake monitoring.
- Fewer Manual Investigations: Processed insights reduce the need to repeatedly explore raw query history.
- Proactive Notifications: Weekly Slack updates provide an immediate view of spending without requiring the team to open the platform.
- Management by Exception: Alerts notify the team when spending changes or an anomaly is detected, allowing them to investigate only when necessary.
At-Bay’s CDC architecture includes multiple layers of materialized tables and views used by Analytics, BI, and R&D teams. Seemore’s end-to-end lineage helps the data team understand these dependencies before making changes, reducing uncertainty and allowing affected stakeholders to prepare in advance.
- Downstream Impact Analysis: The team can identify which tables, views, and users will be affected by a proposed infrastructure change.
- Safer Schema Changes: Before removing a column or modifying a foundational table, At-Bay can understand the full downstream impact.
- Better Change Planning: Seemore’s lineage gives the team the context needed to plan complex changes before implementation.
- Proactive Communication: Impacted teams can be notified in advance and make any required adjustments on their side.
When a cost or performance issue does require investigation, Seemore Assistant and MCP help At-Bay move quickly from a high-level problem to the workloads driving it. Instead of manually piecing together raw Snowflake data, the team can access contextualized insights and identify concrete optimization opportunities.
- Rapid Bottleneck Identification: Seemore helps isolate the specific factors causing an expensive query to become slower over time.
- Actionable Recommendations: The team receives the context needed to implement targeted changes rather than continuing a broad manual investigation.
- Measurable Query Savings: One investigation helped At-Bay optimize a large query and save 6% of their Snowflake spend.
- Natural-Language Access: Seemore Assistant and MCP makes it easier to investigate lineage, usage, and query behavior without navigating across multiple dashboards.