Confluent, Inc. Q3 FY2025 earnings call

• Strong Q3 results: Exceeded high end of guided metrics, subscription revenue grew 19% to $286M, Confluent Cloud revenue grew 24% to $161M, non-GAAP operating margin expanded 3 percentage points to ~10%. • Go-to-market focus: Tightening field alignment drove more use cases into production with over 40% sequential growth in late-stage pipeline progression; expanding large customer base with largest sequential net add in $100,000-plus ARR customers in 2 years. • Flink momentum: Q3 Flink ARR for Confluent Cloud grew >70% sequentially, with over 1,000 customers using Flink. • Partner ecosystem: Partners sourced well over 25% of new business in past 12 months. • AI context data: Confluent's data streaming platform provides context for AI agents/applications, crucial for turning AI prototypes into production systems. • WarpStream performance: 1-year anniversary with 8x consumption growth and multiple 6-figure deals closed.

Subscription revenue grew 19% to $286 million in Q3. Confluent Cloud revenue grew 24% to $161 million, representing 56% of subscription revenue. Subscription gross margin was 81.8%, above the long-term target threshold of 80%. Operating margin increased 340 basis points to 9.7%. Flink ARR for Confluent Cloud grew more than 70% sequentially in Q3, with over 1,000 customers using Flink. WarpStream has seen 8x growth in consumption over the past year.

• Fiscal Q4 2025: Subscription revenue expected to be $295.5M - $296.5M (18% growth), non-GAAP operating margin ~7%, non-GAAP net income per diluted share $0.09 - $0.10. • Fiscal Year 2025: Subscription revenue expected to be $1.1135B - $1.1145B (21% growth), non-GAAP operating margin ~7%, non-GAAP net income per diluted share $0.39 - $0.40, adjusted free cash flow margin ~6%. • Q4 cloud revenue modeled at ~$165M, ~20% growth, accounting for ~56% of subscription revenue.

• Market uncertainties: Impact of broader market conditions on customer spending and adoption. • Competition: Intense competition in data streaming and related spaces, including from CSPs and other vendors. • Operational challenges: Ensuring smooth migration of workloads from open source Kafka to Confluent and managing infrastructure effectively for customers.

Q: Brad Zelnick asked about learnings from go-to-market changes and conversion trends.

A: Edward Kreps said field alignment and DSP specialization are working well, with high confidence in late-stage pipeline progression as customer workloads reaching production drive future consumption.

Q: Sanjit Singh asked when growth might bottom.

A: Edward Kreps said cloud business is strong, with normalization in cloud growth rates when excluding impact of a particular customer, and tailwinds from GSP offerings like Flink.

Q: Mark Murphy asked about late-stage pipeline traction.

A: Edward Kreps said it's a metric tracking dollar value of new use cases moving into production, with strong execution in field driving sequential growth.

Q: Ryan MacWilliams asked about AI use cases in production.

A: Edward Kreps said customer support, anomalies investigations, IoT, manufacturing, retail, financial services, and insurance have active AI projects, with data flow and quality being key.

Q: Robbie Owens asked about CSP replacement opportunity.

A: Edward Kreps said TCO improvements and DSP capabilities are driving customer adoption, with tools for migration making it easier.

Q: Jason Ader asked about better performance due to sales execution vs macro tailwinds.

A: Edward Kreps said structural improvements in go-to-market and new products are contributing to performance.

Q: Aleksandr Zukin asked about cloud revenue trends.

A: Rohan Sivaram said Q3 drivers included momentum of new use cases, normalized optimization, and Flink growth, with long-term visibility from RPO and coverage improvement.

Q: Michael Cikos asked about month-over-month consumption trends.

A: Rohan Sivaram said month-over-month growth improved sequentially due to drivers like new use cases, optimization, and Flink.

Q: Howard Ma asked about Flink's contribution.

A: Rohan Sivaram said Flink has broad and deep adoption, with over 1,000 customers and significant growth, contributing to Confluent Cloud in FY 2026.

Q: Eric Heath asked about Flink wins and competition with Databricks.

A: Edward Kreps said Flink has real-time data pipeline and action-on-data use cases, and Confluent and Databricks are complementary partners serving different constituencies.