{
  "id": 20422663,
  "title": "Before You Renew Snowflake, Run These 6 Queries",
  "url": "https://urgent.news/2026/11/17/before-you-renew-snowflake-run-these-6-queries",
  "topic": "tech",
  "section": "Tech",
  "published": "2026-11-17T08:00:02.000Z",
  "source": {
    "name": "HackerNoon",
    "slug": "hackernoon",
    "url": "https://hackernoon.com/before-you-renew-snowflake-run-these-6-queries?source=rss"
  },
  "original_language": "en",
  "account": "When negotiating a Snowflake contract, it's crucial to dig into the usage data before signing. Here are six key queries to run, all requiring ACCOUNTADMIN or organization-level access:\n\n1. **Spend Breakdown**: Examine spending by service, not just compute. Compute is often the largest line item, but serverless features, storage, and data transfer also contribute. Analyze spend by month and service type over the last 12 full months. Use either organization_usage.usage_in_currency_daily for billing or account_usage.metering_daily_history for credits. This helps identify areas where savings can be made.\n\n2. **Run Rate Calculation**: The commitment should rest on the three-month average of credits billed. This figure provides a better long-term average than the full year's data, as it excludes shorter months. Watch the trend – if credits are growing, a commitment based on today's run rate might be too low. Remember, cloud services are only billed when usage exceeds 10% of the day's warehouse usage, so plan your commitment accordingly.\n\n3. **Idle Compute**: Warehouses continue to bill even when no queries are running. Snowflake now attributes this idle compute to individual queries, making it measurable. Check for idle credits burned by warehouses over the last 30 days. A BI warehouse often shows high idle compute due to frequent queries waking the warehouse. Check warehouse settings like auto_suspend and min_cluster_count to optimize idle time.\n\n4. **Query Usage by Tag**: Set query tags for every scheduled job and BI connection to group costs by purpose rather than service account. Look for queries that run repeatedly, as they often represent inefficiencies. Identifying and optimizing these queries can significantly reduce costs.\n\n5. **Unused Storage**: Examine tables where Time Travel, Fail-safe, and clone bytes account for more storage than the live data. These tables can quickly become costly. Consider setting tables to transient to drop Fail-safe data, and reduce DATA_RETENTION_TIME_IN_DAYS to minimize storage costs. Dropping unnecessary tables also prevents continued billing after they're deleted.\n\n6. **Final Contract Review**: Before finalizing the contract, review the remaining balance and the speed at which it's being consumed. This ensures you understand the financial commitment you're making and can plan for future renewals.",
  "summary": "Prepare for your Snowflake renewal with six SQL queries to analyze spending, identify idle compute, optimize storage, and forecast credit consumption.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}