# Usage & Cost Optimization ## Overview Euno's usage and optimization features help you understand how your data resources are actually being used, identify waste, reduce costs, and improve performance across your entire data stack. **Key Question:** *You have hundreds of thousands of tables, dashboards, and models—but which ones actually matter?* *** ## Why Usage Data Matters ### The Challenge Most data teams face these problems: **🤷 Uncertainty** * "Is anyone actually using this table?" * "Which dashboards can we safely deprecate?" * "Are our expensive transformations even being queried?" **💸 Hidden Costs** * Tables consuming storage that nobody queries * Expensive dashboards with zero viewers * dbt models that take 30 minutes to build but are never used **⏰ Wasted Time** * Maintaining unused resources * Optimizing the wrong things * No visibility into what's actually important **❓ Questions You Can't Answer** * What's our most critical data asset? * Which tables justify their compute costs? * Where should we invest optimization efforts? ### The Solution Usage data answers all of these questions by tracking: * **Query patterns** - What's being queried and how often? * **Cost attribution** - Which resources are most expensive? * **User engagement** - Who's using what and when? * **Performance metrics** - What's slow and needs optimization? * **Adoption trends** - Are new dashboards being adopted? *** ## Usage-Driven Optimization Strategies ### 1. **Identify & Deprecate Unused Resources** **The Opportunity:** 15-30% of data resources typically go unused, yet they consume: * Storage costs * Maintenance time * Mental overhead * Documentation burden **How Euno Helps:** **Step 1: Find Candidates** ``` Query: Tables with 0 queries in last 60 days Result: 156 tables found Potential Savings: $8,400/year in storage ``` **Step 2: Verify Safety** * Check downstream dependencies ([Impact Analysis](/using-euno/impact-analysis.md)) * Verify with owners (automated [Workflow](/using-euno/metadata-activation/workflows.md) notifications) * Confirm no business-critical processes **Step 3: Deprecate** * Archive tables instead of dropping * Document deprecation reason * Monitor for unexpected queries **Expected Impact:** * Cost reduction * Cleaner, more maintainable catalog * Improved data discovery ### 2. **Optimize High-Cost, High-Usage Resources** **The Opportunity:** Your most-queried tables are often not optimized, leading to: * Excessive compute costs * Slow query performance * Poor user experience **How Euno Helps:** **Step 1: Identify High-Value Targets** ``` Query: Tables with >1000 queries/month AND >$500 monthly cost Result: 23 tables Optimization ROI: Very High ``` **Step 2: Analyze Patterns** * Query frequency and timing * User base and access patterns * Current storage structure **Step 3: Apply Optimizations** * Add clustering keys (Snowflake) * Add partitioning (BigQuery) * Create materialized views * Implement caching strategies **Expected Impact:** * Cost reduction on targeted tables * 2-10x query performance improvement * Better user experience ### 3. **Convert Views to Materialized Tables** **The Opportunity:** Complex views that are queried frequently waste compute on every query. **How Euno Helps:** **Step 1: Find Expensive Views** ``` Query: Views with >$100 monthly query cost Result: 12 views ``` **Step 2: Calculate ROI** * Current compute cost: $1,500/month * Materialization cost: $200/month (storage + refresh) * Potential savings: $1,300/month **Step 3: Implement** * Convert view to materialized table * Schedule incremental refreshes * Update downstream dependencies **Expected Impact:** * Cost reduction * Consistent query performance * Reduced warehouse contention ### 4. **Eliminate Unused Dashboards** **The Opportunity:** Dashboards consume: * Scheduled refresh compute * Developer maintenance time * Clutter in BI tools **How Euno Helps:** **Step 1: Find Abandoned Dashboards** ``` Query: Dashboards with 0 views in last 90 days Result: 47 dashboards ``` **Step 2: Validate with Owners** * Automated [Workflow](/using-euno/metadata-activation/workflows.md) notifications * Give owners 30 days to respond * Document reasons for keeping **Step 3: Deprecate or Archive** * Disable scheduled refreshes * Archive workbooks * Document for historical reference **Expected Impact:** * Reduced BI tool clutter * Lower compute costs (scheduled refreshes) * Improved user experience (easier to find relevant content) ### 5. **Optimize dbt Build Times** **The Opportunity:** Slow dbt models delay: * CI/CD pipelines * Data freshness * Developer productivity **How Euno Helps:** **Step 1: Find Bottlenecks** ``` Query: dbt models with average_build_time_60d > 15 minutes Result: 8 models Total time: 180 minutes per run ``` **Step 2: Analyze Causes** * Complex joins * Large data scans * Inefficient SQL * Missing incremental logic **Step 3: Optimize** * Implement incremental models * Add filters/limits for development * Optimize SQL logic * Parallelize where possible **Expected Impact:** * 50-80% reduction in build times * Faster CI/CD pipelines * Improved developer experience *** ## Next Steps 1. **Enable Usage Collection** * Configure your integrations * [Integration Setup Guides](/sources.md) 2. **Explore Your Usage Data** * Use AI Assistant to query * Sort and filter in UI * Identify quick wins 3. **Implement Your First Optimization** * Start with deprecated resource cleanup * Measure impact * [Optimization Strategies](#usage-driven-optimization-strategies) 4. **Set Up Monitoring** * Create usage-based [Workflows](/using-euno/metadata-activation/workflows.md) * Get alerted to issues *** ## Related Documentation * [Impact Analysis](/using-euno/impact-analysis.md) - Understand change impact * [Workflows](/using-euno/metadata-activation/workflows.md) - Automate usage monitoring * [Source Integrations](/sources.md) - Connect your data tools * [AI Assistant](/ai-assistant.md) - Query usage data with natural language --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.euno.ai/using-euno/usage-optimization.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.