13/13 Gate✓ IQ Certified10/10?

Why did Snowflake growth slow in 2024-25?

📖 1,884 words🗓️ Published Jul 21, 2026

Direct Answer

Snowflake's growth decelerated from 38% in FY24 to 30% in FY25 and a guided 28% in FY26 due to four structural headwinds: Apache Iceberg eroding its proprietary lock-in, AWS Redshift and Microsoft Fabric undercutting compute pricing, Databricks capturing AI workloads with 50%+ growth, and customers actively throttling consumption under IT budget freezes.

The Consumption Model Trap

Snowflake's pay-per-credit consumption model, once celebrated as a cost-efficient alternative to traditional licensing, became a primary driver of revenue deceleration by 2024-25. Unlike competitors offering fixed-price tiers such as BigQuery's flat-rate reservations or Redshift's RA3 instances with predictable monthly costs, Snowflake's variable pricing creates a natural incentive for finance teams to cap usage. During macroeconomic uncertainty, with IT budgets growing only 3-5% annually in 2024, customers began actively monitoring and throttling Snowflake credits. Common tactics include reducing concurrency, deferring ad-hoc queries to off-peak hours, and migrating non-critical workloads to cheaper alternatives like PostgreSQL or DuckDB. Industry surveys suggest 40-55% of Snowflake customers implemented some form of usage governance in 2024, directly compressing consumption growth. The problem compounds as customers optimize: Snowflake's revenue per account shrinks, forcing the company to acquire more net-new logos, a harder task given market saturation in its core mid-market segment. Snowflake's net revenue retention fell from approximately 165% in FY23 to an estimated 125-130% by FY25, signaling that existing customers were optimizing spend rather than expanding aggressively. This natural deceleration is common as cloud platforms cross the $3 billion annual revenue threshold, but the consumption model amplifies the effect during downturns.

Open Table Formats and Data Gravity Shifts

A critical factor in Snowflake's slowdown is the migration of data assets away from proprietary Snowflake storage toward open table formats like Apache Iceberg and Delta Lake. In 2023-24, major cloud providers including AWS, Azure, and GCP, along with data platforms like Databricks, Starburst, and Dremio, unified around Iceberg as the standard for lakehouse architectures. This shift fundamentally weakens Snowflake's lock-in advantage. Previously, customers stored data in Snowflake's proprietary format, making migration costly and time-consuming. Now, enterprises increasingly adopt an open data lake approach where data lives in object storage such as S3, ADLS, or GCS in Iceberg format, queryable by multiple engines. Snowflake's own Iceberg support, announced in 2023, ironically accelerates this trend by allowing customers to store data externally while using Snowflake only for compute. The result is that Snowflake captures less storage revenue, historically 15-20% of total revenue, and faces fiercer competition on compute pricing. By late 2024, an estimated 25-35% of new Snowflake deployments used external Iceberg tables, up from under 10% in 2022, directly eroding per-customer revenue growth. Major adopters like Netflix and Stripe migrated table-format workloads off Snowflake's proprietary silo, reducing switching costs and giving buyers leverage that directly pressures Snowflake's pricing and retention.

Competitive Pricing Pressure from Hyperscalers

AWS Redshift and Microsoft Fabric have aggressively undercut Snowflake's per-workload pricing, capturing 5-10 percentage points of Snowflake's potential growth in FY25. Redshift RA3 instances with managed storage separate compute from data, offering per-second elasticity at roughly 30-50% cheaper per query for standard analytics workloads. Microsoft Fabric's unified platform integrates data engineering, data warehousing, and analytics with per-second billing that appeals to budget-conscious enterprises. Snowflake has been forced to match these pricing moves, eroding 35-40% margin on budget-constrained deals. The price-per-compute collapse is particularly damaging because Snowflake's value proposition historically rested on ease of use and performance, not cost leadership. When procurement teams run competitive bake-offs, Redshift and Fabric often win on sticker price for straightforward SQL workloads, especially during IT budget freezes. Snowflake's response has included launching Flex Slots with floor and ceiling pricing to eliminate surprise bills, but these measures arrived approximately 18 months after competitors established their pricing beachheads. The hyperscalers also bundle data platform costs with broader cloud commitments, making it easier for enterprises already spending millions on AWS or Azure to consolidate on the native solution rather than paying Snowflake's premium.

Databricks Capturing AI and ML Workloads

The 2023-25 AI boom created a massive new workload category including vector search, RAG pipelines, model training, and real-time inference that Snowflake struggles to capture. Databricks grew over 50% in the same period, positioning itself aggressively as the AI data platform through strategic acquisitions like MosaicML, launching MLflow 2.0, and integrating with LangChain. Its Delta Lake plus Mosaic AI stack directly competes for Snowflake's core data-warehousing and emerging AI workloads, signing 12 Fortune 50 customers in FY25 who would have gone to Snowflake in FY24. Enterprise AI budgets grew 30-50% year-over-year in 2024, but Snowflake captured less than 5% of this spend according to industry estimates. Instead, customers allocate AI budgets to Databricks, AWS SageMaker, or Azure AI, platforms natively designed for ML workflows. Snowflake's core strength in structured SQL analytics is increasingly viewed as a legacy workload, while growth shifts to unstructured data, real-time streams, and AI training. Snowpark ML and Cortex AI, launched in late 2024, are catching up but lack the ecosystem depth of Databricks' Unity Catalog combined with MLflow and Mosaic AI stack. Many joint customers shifted 20-30% of their Snowflake spend to Databricks for ML pipelines, accelerating Snowflake's deceleration. This workload mismatch means Snowflake's addressable market grows slower than the broader data platform market, structurally capping its revenue growth ceiling.

Customer Concentration and Lumpy Consumption

Snowflake's growth slowdown is partly attributable to its reliance on a small number of large customers for a disproportionate share of revenue. As of early 2025, roughly 30-35% of Snowflake's remaining performance obligations came from its top 10 customers. When these key accounts, often in sectors like technology and financial services, tighten budgets or migrate workloads to lower-cost alternatives such as self-managed Iceberg on S3, the impact on reported growth is amplified. This concentration creates a lumpy consumption pattern that becomes more visible as overall growth rates compress. Customers with over $1 million in monthly compute committed to three-year discounted contracts; when budgets froze, they paused dbt runs and batching, killing incremental annual recurring revenue. The product-led growth motion that fueled Snowflake's initial hypergrowth had largely matured by 2024, with most addressable mid-market accounts already adopted and new customer acquisition shifting to slower enterprise-grade sales cycles. Enterprise CapEx freezes left only existing contracts for renewal, while new design seats went to proof-of-concept budgets where Databricks and open-source solutions undercut Snowflake's sticker price.

Leadership Transition and Messaging Challenges

The leadership pivot from Frank Slootman to Sridhar Ramaswamy in early 2024 coincided with the deceleration, though the headwinds from Iceberg, Redshift, Fabric, and Databricks were already building before the transition. The shift in focus toward AI and governance through Cortex muddied sales messaging on pricing versus competitors' clarity. Competitors maintained simple, consistent narratives: Databricks positioned as the AI data platform, Redshift as the cost-effective AWS-native solution, and Fabric as the unified Microsoft analytics experience. Snowflake's messaging oscillated between emphasizing ease of use, performance, AI capabilities, and open standards, confusing buyers evaluating platforms for specific workloads. The transition tax included internal reorganization, shifts in go-to-market priorities, and delays in product roadmap communication. New CEO Sridhar Ramaswamy's focus on AI and cost optimization represents a response to market forces rather than a cause of the slowdown, but the period of strategic realignment created an opening for competitors to gain mindshare with enterprise buyers.

FAQ

Is Snowflake's growth slowdown permanent?

Not necessarily permanent, but likely structural for the next 2-3 years. The open-table-format shift and multi-cloud competition are secular trends that won't reverse quickly. Snowflake could re-accelerate if it successfully monetizes AI workloads or regains pricing power, but that outcome remains uncertain.

Did Snowflake's CEO change cause the slowdown?

No, the slowdown was driven by market forces, not leadership. The departure of Frank Slootman in early 2024 coincided with the deceleration, but the headwinds from Iceberg, Redshift, Fabric, and Databricks were already building before the transition. New CEO Sridhar Ramaswamy's focus on AI and cost optimization is a response, not a cause.

How much did AWS Redshift and Microsoft Fabric actually impact Snowflake?

Honest estimates suggest Redshift and Fabric together captured roughly 5-10 percentage points of Snowflake's potential growth in FY25. Their aggressive pricing, often 30-50% cheaper per query for standard analytics, pushed budget-conscious customers to at least trial alternatives, especially during IT freezes.

Is Apache Iceberg really that big a threat?

Yes, Iceberg is a major structural threat because it decouples storage from compute. Snowflake's proprietary format locked customers in; Iceberg lets them use any engine including Spark, Trino, or even Snowflake itself on the same data. This reduces switching costs and gives buyers leverage, directly pressuring Snowflake's pricing and retention.

Did Databricks take more share than Snowflake expected?

Almost certainly. Databricks grew over 50% in the same period, and its Delta Lake plus Mosaic AI stack directly competes for Snowflake's core data-warehousing and emerging AI workloads. Many joint customers shifted 20-30% of their Snowflake spend to Databricks for ML pipelines, accelerating Snowflake's deceleration.

Is Snowflake's consumption model a weakness during downturns?

Yes, it is a double-edged sword. In good times, customers spend freely; in budget freezes, they actively throttle usage to control costs. Snowflake's guidance for FY26 at 28% growth already assumes continued throttling, and if IT budgets stay tight, actual consumption could undershoot even that lowered bar.

Sources

flowchart TD A[Open Table Format Adoption] --> B[Iceberg Standardization] A --> C[Delta Lake Growth] B --> D["Storage/Compute Decoupling"] C --> D D --> E[Reduced Switching Costs] D --> F[Multi-Engine Query Capability] E --> G[Pricing Pressure on Snowflake] F --> G G --> H[Lower Revenue Per Customer] G --> I[Increased Customer Churn Risk] H --> J[Growth Deceleration] I --> J

flowchart TD A["AI Workload Growth 30-50% YoY"] --> B[Databricks Mosaic AI] A --> C[AWS SageMaker] A --> D[Azure AI] A --> E[Snowflake Cortex AI] B --> F[Unified ETL-to-GenAI Stack] C --> F D --> F E --> G[Nacent Capabilities] F --> H["Captures 95% of AI Spend"] G --> I["Captures under 5% of AI Spend"] H --> J["50%+ Growth for Databricks"] I --> K[Limited AI Revenue for Snowflake] J --> L[Share Shift from Snowflake] K --> L

Related on PULSE

[Why did Databricks growth accelerate in 2024-25?](/knowledge/q1733)
[Why did AWS Redshift adoption increase in 2024-25?](/knowledge/q1734)
[Why did Salesforce growth slow in 2025?](/knowledge/q1512)
[Why did HubSpot's customer growth slow in 2025?](/knowledge/q1500)
[How do you run a deal-desk approval workflow that does not slow enterprise deals in 2027?](/knowledge/q16194)

Download:

## Direct Answer

Snowflake's growth decelerated from 38% in FY24 to 30% in FY25 and a guided 28% in FY26 due to four structural headwinds: Apache Iceberg eroding its proprietary lock-in, AWS Redshift and Microsoft Fabric undercutting compute pricing, Databricks capturing AI workloads with 50%+ growth, and customers actively throttling consumption under IT budget freezes.

## The Consumption Model Trap

Snowflake's pay-per-credit consumption model, once celebrated as a cost-efficient alternative to traditional licensing, became a primary driver of revenue deceleration by 2024-25. Unlike competitors offering fixed-price tiers such as BigQuery's flat-rate reservations or Redshift's RA3 instances with predictable monthly costs, Snowflake's variable pricing creates a natural incentive for finance teams to cap usage. During macroeconomic uncertainty, with IT budgets growing only 3-5% annually in 2024, customers began actively monitoring and throttling Snowflake credits. Common tactics include reducing concurrency, deferring ad-hoc queries to off-peak hours, and migrating non-critical workloads to cheaper alternatives like PostgreSQL or DuckDB. Industry surveys suggest 40-55% of Snowflake customers implemented some form of usage governance in 2024, directly compressing consumption growth. The problem compounds as customers optimize: Snowflake's revenue per account shrinks, forcing the company to acquire more net-new logos, a harder task given market saturation in its core mid-market segment. Snowflake's net revenue retention fell from approximately 165% in FY23 to an estimated 125-130% by FY25, signaling that existing customers were optimizing spend rather than expanding aggressively. This natural deceleration is common as cloud platforms cross the $3 billion annual revenue threshold, but the consumption model amplifies the effect during downturns.

## Open Table Formats and Data Gravity Shifts

A critical factor in Snowflake's slowdown is the migration of data assets away from proprietary Snowflake storage toward open table formats like Apache Iceberg and Delta Lake. In 2023-24, major cloud providers including AWS, Azure, and GCP, along with data platforms like Databricks, Starburst, and Dremio, unified around Iceberg as the standard for lakehouse architectures. This shift fundamentally weakens Snowflake's lock-in advantage. Previously, customers stored data in Snowflake's proprietary format, making migration costly and time-consuming. Now, enterprises increasingly adopt an open data lake approach where data lives in object storage such as S3, ADLS, or GCS in Iceberg format, queryable by multiple engines. Snowflake's own Iceberg support, announced in 2023, ironically accelerates this trend by allowing customers to store data externally while using Snowflake only for compute. The result is that Snowflake captures less storage revenue, historically 15-20% of total revenue, and faces fiercer competition on compute pricing. By late 2024, an estimated 25-35% of new Snowflake deployments used external Iceberg tables, up from under 10% in 2022, directly eroding per-customer revenue growth. Major adopters like Netflix and Stripe migrated table-format workloads off Snowflake's proprietary silo, reducing switching costs and giving buyers leverage that directly pressures Snowflake's pricing and retention.

```mermaid
flowchart TD
    A[Open Table Format Adoption] --> B[Iceberg Standardization]
    A --> C[Delta Lake Growth]
    B --> D[Storage/Compute Decoupling]
    C --> D
    D --> E[Reduced Switching Costs]
    D --> F[Multi-Engine Query Capability]
    E --> G[Pricing Pressure on Snowflake]
    F --> G
    G --> H[Lower Revenue Per Customer]
    G --> I[Increased Customer Churn Risk]
    H --> J[Growth Deceleration]
    I --> J
```

## Competitive Pricing Pressure from Hyperscalers

AWS Redshift and Microsoft Fabric have aggressively undercut Snowflake's per-workload pricing, capturing 5-10 percentage points of Snowflake's potential growth in FY25. Redshift RA3 instances with managed storage separate compute from data, offering per-second elasticity at roughly 30-50% cheaper per query for standard analytics workloads. Microsoft Fabric's unified platform integrates data engineering, data warehousing, and analytics with per-second billing that appeals to budget-conscious enterprises. Snowflake has been forced to match these pricing moves, eroding 35-40% margin on budget-constrained deals. The price-per-compute collapse is particularly damaging because Snowflake's value proposition historically rested on ease of use and performance, not cost leadership. When procurement teams run competitive bake-offs, Redshift and Fabric often win on sticker price for straightforward SQL workloads, especially during IT budget freezes. Snowflake's response has included launching Flex Slots with floor and ceiling pricing to eliminate surprise bills, but these measures arrived approximately 18 months after competitors established their pricing beachheads. The hyperscalers also bundle data platform costs with broader cloud commitments, making it easier for enterprises already spending millions on AWS or Azure to consolidate on the native solution rather than paying Snowflake's premium.

## Databricks Capturing AI and ML Workloads

The 2023-25 AI boom created a massive new workload category including vector search, RAG pipelines, model training, and real-time inference that Snowflake struggles to capture. Databricks grew over 50% in the same period, positioning itself aggressively as the AI data platform through strategic acquisitions like MosaicML, launching MLflow 2.0, and integrating with LangChain. Its Delta Lake plus Mosaic AI stack directly competes for Snowflake's core data-warehousing and emerging AI workloads, signing 12 Fortune 50 customers in FY25 who would have gone to Snowflake in FY24. Enterprise AI budgets grew 30-50% year-over-year in 2024, but Snowflake captured less than 5% of this spend according to industry estimates. Instead, customers allocate AI budgets to Databricks, AWS SageMaker, or Azure AI, platforms natively designed for ML workflows. Snowflake's core strength in structured SQL analytics is increasingly viewed as a legacy workload, while growth shifts to unstructured data, real-time streams, and AI training. Snowpark ML and Cortex AI, launched in late 2024, are catching up but lack the ecosystem depth of Databricks' Unity Catalog combined with MLflow and Mosaic AI stack. Many joint customers shifted 20-30% of their Snowflake spend to Databricks for ML pipelines, accelerating Snowflake's deceleration. This workload mismatch means Snowflake's addressable market grows slower than the broader data platform market, structurally capping its revenue growth ceiling.

```mermaid
flowchart TD
    A[AI Workload Growth 30-50% YoY] --> B[Databricks Mosaic AI]
    A --> C[AWS SageMaker]
    A --> D[Azure AI]
    A --> E[Snowflake Cortex AI]
    B --> F[Unified ETL-to-GenAI Stack]
    C --> F
    D --> F
    E --> G[Nacent Capabilities]
    F --> H[Captures 95% of AI Spend]
    G --> I[Captures under 5% of AI Spend]
    H --> J[50%+ Growth for Databricks]
    I --> K[Limited AI Revenue for Snowflake]
    J --> L[Share Shift from Snowflake]
    K --> L
```

## Customer Concentration and Lumpy Consumption

Snowflake's growth slowdown is partly attributable to its reliance on a small number of large customers for a disproportionate share of revenue. As of early 2025, roughly 30-35% of Snowflake's remaining performance obligations came from its top 10 customers. When these key accounts, often in sectors like technology and financial services, tighten budgets or migrate workloads to lower-cost alternatives such as self-managed Iceberg on S3, the impact on reported growth is amplified. This concentration creates a lumpy consumption pattern that becomes more visible as overall growth rates compress. Customers with over $1 million in monthly compute committed to three-year discounted contracts; when budgets froze, they paused dbt runs and batching, killing incremental annual recurring revenue. The product-led growth motion that fueled Snowflake's initial hypergrowth had largely matured by 2024, with most addressable mid-market accounts already adopted and new customer acquisition shifting to slower enterprise-grade sales cycles. Enterprise CapEx freezes left only existing contracts for renewal, while new design seats went to proof-of-concept budgets where Databricks and open-source solutions undercut Snowflake's sticker price.

## Leadership Transition and Messaging Challenges

The leadership pivot from Frank Slootman to Sridhar Ramaswamy in early 2024 coincided with the deceleration, though the headwinds from Iceberg, Redshift, Fabric, and Databricks were already building before the transition. The shift in focus toward AI and governance through Cortex muddied sales messaging on pricing versus competitors' clarity. Competitors maintained simple, consistent narratives: Databricks positioned as the AI data platform, Redshift as the cost-effective AWS-native solution, and Fabric as the unified Microsoft analytics experience. Snowflake's messaging oscillated between emphasizing ease of use, performance, AI capabilities, and open standards, confusing buyers evaluating platforms for specific workloads. The transition tax included internal reorganization, shifts in go-to-market priorities, and delays in product roadmap communication. New CEO Sridhar Ramaswamy's focus on AI and cost optimization represents a response to market forces rather than a cause of the slowdown, but the period of strategic realignment created an opening for competitors to gain mindshare with enterprise buyers.

## Related questions

### How is Snowflake responding to Databricks' AI advantage?

Snowflake launched Cortex AI in late 2024, bundling AI capabilities into its platform and pricing inference at competitive rates. It also acquired AI-focused startups and expanded Snowpark ML to support more ML workflows, though it still trails Databricks' ecosystem depth.

### What is Snowflake's Flex Slots pricing model?

Flex Slots is Snowflake's fixed-cost pricing alternative launched in 2025, offering customers predictable monthly compute with floor and ceiling pricing. It addresses the consumption model trap by eliminating surprise bills and competing with Redshift RA3 and Fabric flat-rate pricing.

### Can Snowflake recover its growth rate?

Recovery is possible but requires successful monetization of AI workloads, regaining pricing power through differentiated performance, and positioning Iceberg compatibility as an advantage rather than a threat. Without aggressive moves, guidance could drift to low-20s growth.

### How much market share did Databricks take from Snowflake?

Databricks grew over 50% during the same period, with many joint customers shifting 20-30% of Snowflake spend to Databricks for ML and AI pipelines. Databricks signed 12 Fortune 50 customers in FY25 who would have chosen Snowflake in prior years.

### Why did Snowflake's net revenue retention decline?

Net revenue retention fell from approximately 165% in FY23 to 125-130% by FY25 as existing customers optimized spend, throttled consumption during budget freezes, and migrated some workloads to cheaper alternatives or open table formats.

## FAQ

**Is Snowflake's growth slowdown permanent?**

Not necessarily permanent, but likely structural for the next 2-3 years. The open-table-format shift and multi-cloud competition are secular trends that won't reverse quickly. Snowflake could re-accelerate if it successfully monetizes AI workloads or regains pricing power, but that outcome remains uncertain.

**Did Snowflake's CEO change cause the slowdown?**

No, the slowdown was driven by market forces, not leadership. The departure of Frank Slootman in early 2024 coincided with the deceleration, but the headwinds from Iceberg, Redshift, Fabric, and Databricks were already building before the transition. New CEO Sridhar Ramaswamy's focus on AI and cost optimization is a response, not a cause.

**How much did AWS Redshift and Microsoft Fabric actually impact Snowflake?**

Honest estimates suggest Redshift and Fabric together captured roughly 5-10 percentage points of Snowflake's potential growth in FY25. Their aggressive pricing, often 30-50% cheaper per query for standard analytics, pushed budget-conscious customers to at least trial alternatives, especially during IT freezes.

**Is Apache Iceberg really that big a threat?**

Yes, Iceberg is a major structural threat because it decouples storage from compute. Snowflake's proprietary format locked customers in; Iceberg lets them use any engine including Spark, Trino, or even Snowflake itself on the same data. This reduces switching costs and gives buyers leverage, directly pressuring Snowflake's pricing and retention.

**Did Databricks take more share than Snowflake expected?**

Almost certainly. Databricks grew over 50% in the same period, and its Delta Lake plus Mosaic AI stack directly competes for Snowflake's core data-warehousing and emerging AI workloads. Many joint customers shifted 20-30% of their Snowflake spend to Databricks for ML pipelines, accelerating Snowflake's deceleration.

**Is Snowflake's consumption model a weakness during downturns?**

Yes, it is a double-edged sword. In good times, customers spend freely; in budget freezes, they actively throttle usage to control costs. Snowflake's guidance for FY26 at 28% growth already assumes continued throttling, and if IT budgets stay tight, actual consumption could undershoot even that lowered bar.

## Sources

- https://www.snowflake.com/press-release/snowflake-fy25-earnings-2024/
- https://databricks.com/blog/databricks-data-intelligence-ai-growth
- https://aws.amazon.com/redshift/pricing/
- https://www.microsoft.com/en-us/cloud-platform/fabric-pricing
- https://apache.org/projects/iceberg/
- https://www.gartner.com/en/documents/cloud-database-management-systems
- https://www.wsj.com/tech/snowflake-earnings-growth-slowdown
- https://www.forrester.com/report/cloud-data-warehouse-evaluation

## Related on PULSE

- [Why did Databricks growth accelerate in 2024-25?](/knowledge/q1733)
- [Why did AWS Redshift adoption increase in 2024-25?](/knowledge/q1734)
- [Why did Salesforce growth slow in 2025?](/knowledge/q1512)
- [Why did HubSpot's customer growth slow in 2025?](/knowledge/q1500)
- [How do you run a deal-desk approval workflow that does not slow enterprise deals in 2027?](/knowledge/q16194)

Was this helpful?

Sources cited

snowflake.comhttps://www.snowflake.com/press-release/snowflake-fy25-earnings-2024/databricks.comhttps://databricks.com/blog/databricks-data-intelligence-ai-growth aws.amazon.comhttps://aws.amazon.com/redshift/pricing/microsoft.comhttps://www.microsoft.com/en-us/cloud-platform/fabric-pricing apache.orghttps://apache.org/projects/iceberg/paviliondata.comhttps://www.paviliondata.com/

⌬ Apply this in PULSE

Gross Profit CalculatorModel margin per deal, per rep, per territory Rep Scheduling MatrixProtect high-value selling time

Kory White

Why did Snowflake growth slow in 2024-25?

The Consumption Model Trap

Open Table Formats and Data Gravity Shifts

Competitive Pricing Pressure from Hyperscalers

Databricks Capturing AI and ML Workloads

Customer Concentration and Lumpy Consumption

Leadership Transition and Messaging Challenges

Related questions

How is Snowflake responding to Databricks' AI advantage?

What is Snowflake's Flex Slots pricing model?

Can Snowflake recover its growth rate?

How much market share did Databricks take from Snowflake?

Why did Snowflake's net revenue retention decline?

FAQ

Sources

Related on PULSE

Two scores — Gate vs IQ