Readers typically pay for compute credits used by queries plus storage and ancillary fees. The main cost driver is warehouse size and duration, which determines compute credits; storage adds ongoing monthly charges. The price range varies by region and service level.
Assumptions: region, query complexity, warehouse size, and run duration influence totals.
| Item | Low | Average | High | Notes |
|---|---|---|---|---|
| Compute (credits per query) | $0.10 | $0.25 | $2.50 | Depends on warehouse size and runtime; per-query cost scales with credits used. |
| Storage (per TB per month) | $12.00 | $25.00 | $60.00 | Active vs long-term storage; higher for hot data and frequent access. |
| Admin & Configuration (hours) | $0.50 | $2.00 | $8.00 | Routine setup, tuning, and monitoring time. |
| Taxes & Fees | $0.05 | $0.20 | $1.00 | Regional tax and service fees may apply. |
| Contingency | $0.10 | $0.50 | $1.50 | Budget cushion for unexpected usage spikes. |
Overview Of Costs
Compute costs are the largest component for most Snowflake users, driven by warehouse size and query duration. Storage costs accrue monthly and are separate from per-query compute. Pricing varies by cloud region (AWS, Azure, Google Cloud) and by the Snowflake edition (Standard, Enterprise, etc.). This section provides total project ranges and per-unit ranges with brief assumptions to help planners estimate budgets.
Cost Breakdown
Understanding the cost mix helps identify where savings are realistic. The table below uses a mix of per-query and monthly elements. Assumptions: a mid-sized data set, routine nightly loads, and a 24/7 warehouse on a standard edition.
| Component | Low | Average | High | Assumptions |
|---|---|---|---|---|
| Compute (credits used per query) | $0.10 | $0.25 | $2.50 | Short to moderate-running queries on X-Small to Small warehouses; 1–5 minutes per query. |
| Storage per TB per month | $12.00 | $25.00 | $60.00 | Active dataset with moderate access frequency. |
| Labor & Admin | $0.50 | $2.00 | $8.00 | Setup, monitoring, and optimization work. |
| Taxes | $0.05 | $0.20 | $1.00 | Regional taxes and fees. |
| Contingency | $0.10 | $0.50 | $1.50 | Spike handling and unexpected usage. |
Factors That Affect Price
Key price drivers include warehouse size, query duration, data volume, and region. The main formula is: compute cost = credits used × price per credit; storage is per TB per month; administration varies with governance requirements. Additionally, cloud region and edition tier affect the per-credit rate and features available.
Cost Drivers
Two niche-specific drivers are important: 1) Warehouse configuration (size and auto-suspend) impacts credits per hour; 2) Data loading patterns (streaming vs batch) influence concurrent credits and storage traffic. For example, an X-Small warehouse might cost less per hour but run longer for complex joins, changing the per-query cost profile.
Ways To Save
Several practical tactics reduce Snowflake costs without sacrificing performance. Use auto-suspend and auto-resume to minimize idle compute, keep frequently accessed data in hot storage but leverage clustering keys to speed queries, and apply caching and result reuse where appropriate. Consider consolidating workloads to fewer, larger warehouses during peak hours to improve efficiency.
Regional Price Differences
Prices vary by region and cloud provider. In the U.S., three common patterns emerge: Region A (West Coast) often has higher peak-hour rates due to demand, Region B (Central) usually sits near average credits pricing, and Region C (East) can show slightly lower monthly storage costs due to data residency choices. Expect roughly +/- 10% to 25% deltas between regions depending on currency, taxes, and service tier.
Real-World Pricing Examples
Three scenario cards illustrate typical costs for common usage patterns.
-
Basic: 1 TB stored, nightly ETL, 1 tiny query per day, X-Small warehouse, auto-suspend enabled.
Assumptions: region, data volume, 12 hours of monthly warehouse time.Compute: 0.5 credits/day × 30 days ≈ 15 credits/month. Storage: 1 TB ≈ $12-$15/mo. Total ≈ $15-$40/month before taxes and contingencies; per-query ≈ $0.25–$0.75.
-
Mid-Range: 50–100 GB daily load, 100 GB in storage growth, a mix of joins and aggregations, Small to Medium warehouse, 6–8 hours across peak times.
Assumptions: region and edition standard.Compute: 100–200 credits/month; Storage: 0.5–1 TB/mo. Total ≈ $60-$260/mo; per-query ≈ $0.50–$2.50.
-
Premium: 5 TB storage, high-frequency analytics, large complex queries, Medium to Large warehouse with high concurrency.
Assumptions: optimized clustering and frequent bursts.Compute: 1,000–2,500 credits/month; Storage: 5 TB/mo. Total ≈ $500-$1,500/mo; per-query ≈ $2.00–$10.00.
Pricing FAQ
Common questions touch on credits, storage, and regional variation. How many credits does a typical query consume? It depends on the warehouse size, run time, and concurrency. How is storage priced? Per TB per month, with differences for hot vs cold storage. Do taxes apply? Yes, taxes and regional fees can affect the final bill.
data-formula=”compute_credits_per_query × price_per_credit”>