Databricks is preparing for what could be one of the largest tech IPOs in recent memory, and in doing so, it is quietly tightening its grip on territory that Confluent has spent years trying to lock down – real-time data pipelines that connect storage, processing, and AI workloads.
The IPO That Changes the Competitive Map
Databricks has not officially set an IPO date, but the company has been making moves that signal serious preparation: hiring senior finance executives, filing confidential paperwork, and expanding enterprise sales infrastructure. Going public is not just a capital event for Databricks – it is a strategic signal to the market that the company intends to operate at a different scale and compete in categories it previously left to partners or adjacent players.
Confluent built its business on Apache Kafka, the open-source streaming platform originally created at LinkedIn. It turned real-time event streaming – the plumbing that moves data between systems the moment something happens – into a managed cloud product. That positioning made Confluent the default answer for engineering teams that needed low-latency data pipelines, and the company went public in 2021 on the strength of that story. For a while, it looked like a durable moat. Kafka expertise is hard to build, and Confluent layered governance, connectors, and enterprise support on top of a technology that was already deeply embedded in production systems.
What changed is that Databricks stopped treating data pipelines as someone else’s problem. Delta Live Tables, its managed pipeline product, handles incremental data loading, schema evolution, and data quality enforcement inside the Databricks environment. That means teams running AI and analytics workloads on Databricks have less reason to introduce a separate Confluent layer. The data can move from source to lakehouse to model without leaving the platform. For engineering teams trying to reduce tool sprawl, that is a genuinely attractive proposition.
Databricks’ IPO push accelerates this pressure in a specific way. Public market investors will want to see total addressable market expansion, and the company will likely frame data ingestion and streaming as part of its core platform story rather than a peripheral feature. That framing – repeated in analyst calls, investor presentations, and product marketing – shifts how enterprise buyers think about what Databricks is actually for. It stops being a query engine sitting on top of a data lake and becomes infrastructure for the entire data journey.
Where Confluent’s Ground Becomes Contested
Confluent’s competitive strength has always been depth rather than breadth. Its Kafka-based architecture handles extremely high message volumes with strong ordering guarantees, making it the right choice for financial transaction processing, real-time fraud detection, and event-driven microservices architectures. These are workloads where sub-millisecond latency matters and where operational reliability is non-negotiable. Databricks does not directly compete in those scenarios – and probably never will.
The problem for Confluent is that many of its customers use it for something less demanding: moving data from operational databases into analytics systems. That use case – often called change data capture or CDC – is increasingly covered by Databricks’ native connectors and pipeline tools. When a company is already paying Databricks for compute and storage, paying Confluent separately to shuttle data between systems becomes a harder budget conversation to win.
Databricks has also been aggressive about partnerships and acquisitions that close capability gaps. Its acquisition of Arcion brought native CDC functionality directly into the platform, targeting exactly the kind of database replication workflows that Confluent’s connectors handle. That is not a peripheral feature addition – it is a direct move into Confluent’s bread-and-butter revenue territory. Confluent sells connectors as a meaningful part of its platform value; Databricks just made many of those connectors redundant for customers already in its ecosystem.
The IPO dynamic adds financial pressure on top of competitive pressure. Once Databricks is public, it will have access to stock-based currency for further acquisitions, and it will face Wall Street pressure to grow faster in every segment it touches. That creates an incentive structure where the company will actively pursue customers currently sitting in Confluent’s install base, offering consolidation deals that trade pricing flexibility for platform lock-in. Confluent, which is already navigating its own path to profitability as a public company, has less flexibility to engage in prolonged price wars.
There is also a talent and narrative dimension that does not show up directly in revenue figures. Databricks has been extraordinarily successful at positioning itself as the AI data company – the place where machine learning, analytics, and data engineering converge. That narrative attracts enterprise architects who are designing new systems rather than maintaining old ones. When a company is building a new data platform from scratch, choosing Databricks as the center of gravity means Confluent enters the conversation as an add-on rather than a foundation. That shift in architectural decision-making compounds over years.
Confluent has responded by investing in Tableflow, a feature designed to make Kafka topics directly readable as Apache Iceberg tables – the same open format that Databricks’ Delta Lake ecosystem supports. The idea is to position Confluent not just as a streaming layer but as a participant in the broader lakehouse architecture. Whether that integration story is compelling enough to keep enterprise buyers from consolidating onto Databricks alone is the question Confluent’s sales team faces on nearly every competitive deal right now.
What the Numbers Cannot Fully Capture
Confluent’s last reported annual recurring revenue crossed $900 million, and the company has maintained strong net revenue retention, meaning existing customers keep spending more. That is a real business with real staying power. The risk is not that Confluent disappears – it is that its growth ceiling gets compressed as Databricks absorbs the marginal data pipeline budget that might otherwise have flowed to Confluent. In enterprise software, ceiling compression is often more damaging than direct customer loss, because it changes the valuation story that investors are willing to pay for.
The outcome may ultimately depend on how many enterprises decide they want best-of-breed streaming infrastructure versus how many decide they want fewer vendors and cleaner contracts. Databricks is betting heavily that consolidation wins. If the IPO raises the capital and the profile the company is angling for, it will have significant resources to make that bet pay off – and Confluent will spend the next several years defending the ground it built its entire public market story on.
