Confluent S-1 Reveals ‘Reimagining of Business’ Theme
Kafka-backer Confluent yesterday published its S-1 with the SEC, which it previously kept confidential when it originally filed to go public back in April. The 198-page document reveals some details about Confluent, as well as the company’s overall view of where it fits into the future of business and technology.
First, the data.
We know that Confluent recorded $236.6 million in total revenue in fiscal year 2020 ended December 31, up from $149.8 million and $66.2 million in 2019 and 2018, respectively. Confluent recorded expenses of $394.3 million, $198.6 million, and $90.3 million in 2020, 2019, and 2018, representing operating losses of $233.2 million, $98.1 million, and $41.5 million for those years, respectively. As of March 31, the company reported having $280.1 million in cash and cash equivalents on hand. Total shareholder equity was reported to be $244.6 million at that time.
We know that, as of March 2021, the company has more than 2,500 paying customers, an increase of 1,500 in just 12 months. The company had 561 customers who were spending $100,000 or more per year at the end of March 2021, compared to 374 at the same time in 2020. Customers spending $1 million or more increased from 33 to 60 from March 2020 to March 2021, according to the S-1. The COVID-19 pandemic had, and continues to have, a “modest adverse impact on certain parts of our business…” the company said.
The Confluent Platform, the company’s self-managed offering for enterprises, represents the bulk of revenue for Confluent, bringing in $115.8 million and $177.2 million in 2019 and 2020, respectively. By comparison, Confluent Cloud, the company’s managed offering and what it hopes to be its primary growth vehicle going forward, brought in $14.4 million and $31.4 million in 2019 and 2020, respectively. Revenue from outside the United States accounted for 34% and 36% of our total revenue in 2020 and 2021, respectively, the company said.
Confluent hasn’t said what stock exchange it will be listed on, but we do know that it will use CFLT as its ticker symbol.
The S-1 shows that Confluent values its capability to help companies succeed in the “digital-first” paradigm. Being digital-first, the company says, is not just a matter of “adding an application or automating an existing process. It is an end-to-end reimagining of business.”
It’s about building new experiences to interact with customers on the front-end. Behind the UX, digital-first is about “transitioning to real-time, software-driven back-end operations.”
For retail companies, digital-first can help keep better track of inventory across multiple channels, the company says. A manufacturer, meanwhile, may want to harness “a real-time flow of data from….IoT sensors to deliver predictive maintenance and reduce downtime.”
Confluent hopes to help customer succeed with that “life or death” struggle for survival, largely on the back of the Apache Kafka message bus and associated add-ons. In the view of Confluent, traditional data management technologies can only handle data at rest, and aren’t cut out for the data-in-motion future that widespread digitization will bring.
“It is not just that companies are using more software–in a very real sense, they are actually becoming software,” the company writes. “As companies increasingly become software, they need a central nervous system that connects all of their disparate software systems, unifying their business and enabling them to react intelligently in real-time.”
As companies become software and build new data connections to bring in data from teams, applications, and sensors far and wide, they’re using more machine learning technology to act upon that data. “Today, many of the most powerful insights delivered by machine learning applications and systems depend on massive and continuous volumes of data in motion being processed, connected, and analyzed all at once,” the company writes in the S-1.
The growth of IoT and connected sensors is another factor fueling digitization. Confluent cites IDC data that predicts connected IoT devices will generate 73.1 zettabytes of data by 2025, nearly 4x more than 2019. “To capture this massive volume of real-time data and build solutions that deliver transformative impact, enterprises need a new foundational data infrastructure designed for data in motion,” Confluent writes.
The adoption of microservices that break up functions into smaller atomic chunks, as well as the movement to the cloud, with its nearly limitless computing capacity, are also trends that factor in favor of Confluent and its quest to help customers succeed in the age of digitization.
Confluent points out that 70% of the Fortune 500 are already Kafka users. Kafka, of course, is the core underlying technology upon which Confluent is built. Kafka, which the Confluent co-founders created at LinkedIn in the early 2000s, is mentioned 84 times in the 198 pages of the S-1 (not including charts at the end), while Confluent Cloud is mentioned 118 times and Confluent Platform 56 times.
ksqlDB, which Confluent dubs a “native data-in-motion database,” gets 15 mentions. But its value as a “bridge” technology perhaps is understated in the S-1. “We believe that the rise of real-time stream processing of data in motion is still in the early stages of adoption. Our investment in ksqlDB positions us to succeed in this emerging area as it gains adoption with customers,” the company writes in the S-1.
The S-1 also includes details of recent stock sales of the three co-founders of Confluent (who are also the three co-creators of Apache Kafka). Some of these sales occurred last summer as part of the Series E redeemable convertible preferred stock financing; Confluent’s Series E financing round, which brought in $250 million at a $4.5 billion valuation, was announced in April 2020.
Neha Narkhede, who left Confluent in December 2019 and will continue as non-employee director of the company going forward, sold $77.8 million worth of stock to new and existing investors in September, according to the S-1. Some of that was counted as compensation, it stated.
Jay Kreps, Confluent’s CEO, and his wife sold $39.5 million in convertible founder stock, some of which counted as compensation on top of his annual salary, which is $350,000. Jun Rao, meanwhile, sold nearly $30 million worth of stock. Rao doesn’t appear to have a formal title at Confluent. He is not listed as a director on the S-1, only as a co-founder and 5% beneficial owner. His LinkedIn status lists him as a Confluent co-founder, which hasn’t changed in a while.
No indication on the timing of Confluent’s initial public offering was given. Whether it will capture the excitement of the market, as did Snowflake’s blockbuster IPO last year, has yet to be seen. In light of Cloudera’s recent $5.3 billion private party buyout and the continued evolution of data platforms, it will be very interesting to watch how CFLT proceeds.
You can access Confluent’s S-1 form from the SEC’s website at this link.