Snowflake Cost Optimization: 10 Ways to Manage Runaway Costs

Key Takeaways 

• When used as intended [a traditional data warehouse] Snowflake is a great, cost-effective tool
• Unfortunately, it can get very expensive, especially for SaaS providers
• We explain Snowflake’s pricing model & why it can get very expensive
• Learn ten smart Snowflake cost optimization strategies

Snowflake’s Data Cloud enables organizations to unite siloed data, discover and securely share data, and execute diverse analytic workloads. Unfortunately, some Snowflake users encounter difficulties with high prices, and the pricing may be incompatible with SaaS providers’ business model.

We’ll describe the Snowflake pricing model, explain why it can become very expensive, and list ten smart Snowflake cost optimization strategies.

Snowflake’s Pricing Model

Snowflake was originally designed as a lower cost, alternative solution to meet traditional data warehouse requirements. Most of the costs associated with traditional data warehouses such as Oracle or Teradata were upfront license fees of six or even seven figures. Snowflake, on the other hand, gets users in the door with zero infrastructure costs upfront and a more cloud-oriented, utilization-based ‘pay as you go’ model.

Warehouses are available in sizes x-small to 6X-large, with each tier doubling in cost and compute power. The cost model is both linear and exponential. If you retain the same compute size, with all jobs being relatively small and no heavy lifting, your costs will be linear.

Similar to public cloud infrastructure, your expenditure scales with your usage. It’s logical to increase the price as value increases, but only if the increases are proportionate. Unfortunately, that doesn’t seem to always be the case with Snowflake. Once you require more power, the jump is exponential, doubling at each tier. Too often, people don’t realize this until they receive the unpleasant surprise of a five-digit bill for a single month.

Why is Snowflake So Expensive?

If you use Snowflake as it was designed to be used (a traditional data warehouse), it’s quite cost-effective. When you run big queries just a few times a week, it’s pretty cheap. One response to a query on PeerSpot states Snowflake, “can reduce costs in comparison to other options, such as Oracle.” Well, sure… a Rolls Royce is less expensive than a Bugatti Veyron, but it’s still a tad pricey for picking up groceries.

Bills Can Quickly Spiral Out of Control

Traditionally, data warehouses are used only occasionally. But, with any great tool, Snowflake often gets used heavily, and people begin relying on it for other uses, such as transactional database operations. This results in continuous operations. From a cost perspective, it’s like having your lights (or dishwasher or AC) running ALL.THE.TIME, which then results in a very expensive utility bill. For example, Instacart paid Snowflake $51 million in 2022, as revealed in their IPO filing.

Great Rates in Some Use Cases

A traditional data warehouse model in which you load data every night might require one hour, equaling one credit. If you shut off Snowflake when that hour was complete, which is typically done, you would pay about $90 per month.

Snowflake could be a great fit in this enterprise BI use case, where an analyst loads data from a single source of truth once daily and prepares additional executive reports quarterly. Other low performance requirements such as testing scenarios are also typically well-served by this model.

And Potentially Not-So-Great Rates for SaaS Providers

In our first example, you could expect significant quantities of idle time, but that goes out the window when your app must be available almost instantly to everyone. SaaS providers aim to increase the usage of their apps. As we wrote in “Making the Cut,” empowering users to create their own dashboards, and making your app useful to more employees within each organization boosts the value of your app.

With such changes, your app can become stickier, increasing customer retention, but also making utilization less predictable. With the need for real-time, interactive, and embedded analytics that is always on and performs well, delivering responsive queries requires larger environments. What was once reasonable Snowflake storage costs could potentially jump multiple thousands of dollars per month with a Snowflake database.Use our Snowflake Pricing Calculator to discover your potential savings for a Snowflake warehouse used for ad-hoc queries.

Why Snowflake is Economically Incompatible with Embedded Self-Service Analytics

When SaaS companies use Snowflake to power embedded analytics, the usage often rises dramatically. Particularly with self-service analytics and/or a large, global customer base, usage can become near continuous as users run queries and load data. 

This leaves SaaS providers facing a conundrum:

• Good news: A large, engaged user base, leveraging the app and gaining value from it, increasing stickiness
• Bad News: Costs rising exponentially

It’s no fault of Snowflake’s. It truly is a fantastic data warehousing tool and a high-performance database. But it’s simply not the right tool for embedded analytics.

Should You Query Snowflake Directly for Embedded Analytics?

Generally speaking, most companies will find this type of use case cost-prohibitive in the long run. This can be a good method to test a concept or show certain customers a prototype. However, to release a full-scale embedded analytics solution within your SaaS application, Snowflake tends to fall down for multi-tenant analytics use cases in both costs and functionality.

A great way to reduce high Snowflake costs is to structure your analytics so you’re not continuously querying the Snowflake data engine. That’s where Qrvey can help. You can cache the data and sync to Qrvey’s data lake, so you’re not continuously using Snowflake, periodically pause it, continuously using Qrvey which doesn’t have same utilization costs from data perspective.

You can configure your Snowflake utilization so it doesn’t eat away at your profit margins. This alleviates the cost of continuously running Snowflake. Qrvey’s solution takes you back to the way Snowflake was intended to be used.

Snowflake Cost Optimization: 10 Smart Strategies to Reduce Costs

If you’re managing cloud infrastructure, you’ve got a leg up in optimizing Snowflake costs. From observability to idle resources, many of the tactics for optimizing Snowflake costs are similar to the tactics for all types of cloud infrastructure. 

Here are ten cost reduction strategies we recommend:

1. Portion Data Based on Urgency

You can significantly reduce your Snowflake costs by using Qrvey and the data synchronization model. Qrvey’s solution includes “Live Connect,” which enables real-time data support, making the optimization of data sources like Snowflake and Amazon Redshift much easier, improving performance and lowering costs.

This gives you the flexibility to optimize your usage by deciding what will sync live versus a predefined schedule. If your utilization patterns are predictable, you can optimize by pulling data only during a designated sync period. (Try our Snowflake price calculator to see your reduction for yourself)

2. Fine-Grain Prioritizations

It’s best to allocate your most expensive resources to the areas where that expertise is truly needed. For example, your highly skilled engineers are likely to focus on the most technically complex, business-critical projects. Similarly, by using Qrvey, you can create a data optimization model that allows you to use tools like Snowflake and others to achieve the most cost-effective combination.

With Qrvey, you can achieve fine-grained distinctions, separating out data even down to the chart level.  

For product teams building embedded analytics, Qrvey can blend data sources on a single dashboard. This means you can source data from Snowflake AND from Qrvey’s built-in data warehouse on the same dashboard.

This gives you ultimate cost control to decide what visualization requires real-time data and which can tolerate some latency.

3. Select the Right Warehouse Sizes

Your compute costs relate to the duration and capacity of your warehouses in use. Selecting the proper warehouse size and efficiently managing your warehouses can help optimize Snowflake costs.

If you reduce and find that performance is still acceptable, consider reducing size yet again.

4. Reduce Storage Costs by Cleaning your ‘Closets’

The more data you’re storing, the higher your storage costs. Regularly monitor and delete redundant or outdated data to reduce your Snowflake storage costs. 

5. (Quickly) Auto-Suspend Idle Warehouses

With Snowflake – as with most cloud infrastructure  – if it’s running, you’re paying for it. Therefore, if you’re not using it, it shouldn’t be running. Fortunately, Snowflake – again, like most cloud services – realizes this and equips you with the tools to make this easy. 

A warehouse can be set to automatically resume or suspend, based on activity. By default, auto-suspend is enabled. Snowflake automatically suspends the warehouse if it is inactive for the specified period of time. Auto-resume is also enabled by default.

To further reduce Snowflake costs, you can reduce the time post-query prior to auto-suspension.

6. Watch Out for Long Running Queries

Snowflake provides parameters that let you control the behavior of your account, individual user sessions, and objects. The STATEMENT_TIMEOUT_IN_SECONDS parameter controls the amount of time, in seconds, after which a running SQL statement (query, DDL, DML, etc.) is canceled by the system. The default is listed as, “172800 seconds (2 days).” You can prevent excessively long queries by adjusting this timeout value. 

7. Use Resource Monitors

No organization wants to be surprised by bills that are exponentially higher than expected. But without observability, it’s a constant risk with cloud infrastructure.

Snowflake resource monitors can be used to monitor credit usage by virtual warehouses and the cloud services needed to support those warehouses. Additionally, you can set up a resource monitor to suspend a user-managed virtual warehouse when it reaches a credit limit.

8. Consolidate Warehouses

Once again, Snowflake challenges mirror cloud computing in general, with sprawl being a common problem. Warehouse sprawl results in each warehouse performing relatively fewer queries and therefore sitting idle more often. 

Idle warehouses incur only storage costs which are relatively cheap, particularly compared to compute credits. However, all of the aforementioned cost reduction techniques are more difficult with more warehouses. You end up with duplication of tasks and efforts, which results in additional labor costs. Consolidating to manage fewer warehouses enables you to retain better control of your Snowflake costs.

SELECT, the Snowflake optimization and cost management platform, suggests, “Where needed, create separate warehouses based on performance requirements versus domains of workload. For example, creating one warehouse for all data loading, one for transformations, and one for live BI querying will lead to better cost efficiency than one warehouse for marketing data and one for finance data.”

9. View the Snowflake Cost Insights

Snowflake identifies opportunities to optimize costs within a particular account. These insights are calculated and refreshed weekly, and you must be granted the ACCOUNTADMIN role to view them. Sample insights include rarely used tables, large tables that are never queried, rarely used materialized views, and rarely used search optimization paths.

10. Don’t Time Travel (Too Far)

Snowflake Time Travel enables accessing historical data at any point within a defined period. Snowflake maintains copies of all modifications and changes to a table made over your retention period. Therefore, the longer the period, the more data retention and the higher your costs.

The ability to restore data that has been accidentally deleted can avert a catastrophe, but to optimize costs, avoid enabling travel too far back.

How Qrvey Helps Optimize Snowflake Costs (with Cost Calculator)

Qrvey offers a cost-effective solution tailored to analytics in SaaS apps. Its embedded analytics layer enables the creation of fully custom dashboards.

Buying a third-party platform to embed into your SaaS product can bring numerous advantages over building all your functionality in-house. Use our Build vs Buy Calculator to uncover the hidden costs and find your ROI.

Find out how much money you can save with Qrvey. Try our Snowflake cost calculator today.