How Often Are Snowflake Bills Generated?

If you’re facing Snowflake sticker shock, you may wonder how often bills are generated. Billing frequency impacts cost visibility.

Snowflake bills are generated monthly.

If you’re asking this, costs may be higher than expected. Here’s how to control Snowflake costs.

How often are Snowflake bills generated—and how does billing work?

Every month, Snowflake generates a Snowflake Order Form, a monthly usage statement for customers with at least one active contract. It shows credits consumed and total spend.

It also includes lifetime usage under your contract. Only ACCOUNTADMIN and ORGADMIN roles can view it.

How to interpret your Snowflake bill

Snowflake bills can be complex. Pricing includes multiple components that are hard to interpret. Understanding pricing helps interpret your bill.

Snowflake pricing is determined by three usage categories: data storage, data transfer, and compute resources.

Data storage

Snowflake charges a flat rate per TB for stored data, including tables and historical data.

Storage costs are based on average daily storage usage in your Snowflake account. The exact rate depends on whether your account is Capacity or On-Demand, and also varies by region (e.g. US vs. EU) and platform (e.g. AWS vs. GCP). 

Data transfer

Snowflake does not charge for inbound data but charges for cross-region data transfer, whether on the same or a different cloud platform. Include this in total cost calculations.

Compute resources

Compute usage consumes Snowflake credits. There are three types of said resources: 

→ Virtual Warehouse Compute consumes Snowflake credits based on seconds spent loading data, executing queries, and performing other DML operations. Warehouses scale across ten sizes, each doubling in cost. Credits apply only while running, not while suspended or idle, and credit usage per hour directly correlates to the number of servers in a warehouse cluster. Performance scales roughly linearly with size, so doubling a warehouse’s size usually results in a 50% reduction in a given task’s processing time for the same cost. 

→ Serverless Compute describes certain features like Search Optimization and Snowpipe which use Snowflake-managed compute resources instead of virtual warehouses. These resources are automatically scaled up or down for each workload. 

→ Cloud Services Compute describes the cloud services layer of Snowflake architecture which consumes credits for tasks like authentication, metadata management, API, SQL, query parsing and access control—if daily consumption of cloud services resources exceeds 10% of daily warehouse usage. 

Why is your Snowflake bill larger than expected?

If your Snowflake bill is higher than expected, you’re not alone. The cloud does not guarantee savings. Here are some things you need to watch out for to keep your costs under control:

• Over-provisioning. Unlike on-prem servers, the cloud offers virtually infinite scalability. You can scale down to reduce costs but the opposite is also true—and there’s no cap on how high you can go. So you’ve traded one over-provisioning problem for another that requires just as active and watchful attention as the previous. 
• Under-provisioning at Capacity. If you pre-purchase Snowflake credits for a discount (called Capacity) and go above that amount, you’ll incur a steep price-per-credit increase. 
• Lack of visibility into usage. If you don’t have a clear line of sight into usage and cost increases, it can be difficult to control them. That’s why we recently launched our free Snowflake Workload Intelligence tool to give you the visibility you need to manage Snowflake usage. 
• Relying on advice, not automated action. Optimization advice is just advice. Someone has to implement it. If you’re relying on a data engineering team that’s already stretched too thin to do this, you’re going to miss opportunities. That’s why we recommend using a Snowflake optimization automation tool like Keebo to do this in the background. 

It’s important to keep in mind that because Snowflake pricing is dynamic and often relies on factors outside your control, optimizing your costs isn’t a one-time endeavor. Even if you optimize for today’s state, things change. 

Dynamic, automated optimization is the only way to truly keep your Snowflake costs under control and prevent future sticker shock when your next bill comes in. 

Final thoughts on keeping your Snowflake bill under control

If you feel like your monthly Snowflake bills are out of control, remember: you don’t have to passively accept these changes. By taking the bull by the horns, as it were, you can proactively optimize your Snowflake spend and minimize costs.

This is exactly what Team Velocity did when they started implementing AI optimization with Keebo. With only 30 minutes of effort, they achieved 36% cost savings, seeing their first results within only 24 hours. 

Read the full Team Velocity case study, including how they allocated their freed-up resources, here.