Work-Bench Snapshot: Augmenting Streaming and Batch Processing Workflows
The Work-Bench Snapshot Series explores the top people, blogs, videos, and more, shaping the enterprise on a particular topic we’re looking at from an investment standpoint.
This post was originally published on The Data Source, my monthly newsletter covering the top innovation in data infrastructure, engineering and developer-first tooling. Subscribe here!
This is the first installment in an ongoing series exploring open source business strategies through the eyes of those who've shaped them. The series features interviews with founders who have successfully navigated the complex balance between open source adoption and commercial sustainability. Each piece provides unfiltered insights into the strategic decisions, pivots, and lessons learned while building businesses around open source software. These firsthand experiences highlight the nuanced realities of open source as both a community endeavor and a business model. This inaugural feature spotlights Paul Dix of InfluxData, whose journey with InfluxDB exemplifies many of the challenges and opportunities inherent in open-core strategies.
The strategic choice between open source and closed source remains a defining factor for startups, particularly in infrastructure software. Paul Dix, CTO and co-founder of InfluxData, the company behind InfluxDB, provides nuanced insights from his firsthand experience navigating these strategic decisions, emphasizing both the opportunities and challenges inherent in open-source models.
Initially, Paul's venture, Errplane, focused on SaaS applications for real-time server monitoring. However, when applying to Y Combinator in 2013, Paul included an open-source time-series database as a secondary idea, drawing from his previous experiences building similar solutions for fintech and server monitoring projects. These experiences underscored for him the importance of openness and transparency to developer trust and product adoption.
When Errplane's primary product struggled to gain traction, Paul and his team revisited their secondary idea. He recognized open source as critical for attracting developer attention, remarking, "Developers are only going to pay attention if we open source it." Upon announcing InfluxDB in 2013, the project quickly gained popularity but faced significant monetization hurdles. Paul vividly remembers the early frustration: "We offered support for months and ended up closing exactly one contract. Clearly, it wasn't scalable."
A pivotal industry event occurred in late 2015 when Amazon launched a highly successful hosted Elastic service. Seeing Amazon monetize Elastic more effectively than Elastic itself was a stark wake-up call for Paul. It underscored the market reality that cloud providers could quickly and efficiently leverage open-source projects for significant commercial gain. This realization prompted a strategic shift for InfluxDB—restricting advanced clustering and high-availability capabilities to a closed-source, commercial fork. Despite significant backlash, particularly vocal on platforms like Hacker News, Paul remained firm in his decision, openly engaging with critics and transparently communicating their motivation: "We had to build a sustainable business."
Paul highlights the internal and external challenges of managing both open and closed-source offerings simultaneously, noting, "Our closed-source product competes directly with our open-source offering." Ideally, Paul believes, open-source should complement commercial offerings rather than directly compete. He views open source primarily as a powerful marketing and distribution channel rather than a direct sales pipeline.
Additionally, Paul reflects on the broader strategy of commoditizing dependencies through open source, highlighting InfluxDB's involvement in Apache Arrow and Apache DataFusion. He explains, "We contribute significantly to Apache DataFusion because it serves as a critical dependency, allowing us to leverage a robust query engine without bearing the full cost of development ourselves." This strategy of commoditizing dependencies helps them strategically focus resources where it matters most for their product.
Reflecting on these lessons, Paul advises founders to clearly delineate open and closed-source components from inception, explicitly structuring teams and resources accordingly. He recommends allocating a modest share—approximately 10-20% of engineering resources—to open-source initiatives, ensuring the commercial product consistently outpaces its open-source counterpart in feature richness and innovation.
Today, InfluxDB successfully balances a vibrant open-source community with a profitable enterprise offering, illustrating the viability of a strategically managed open-core model. Paul's journey underscores the importance of clarity, strategic flexibility, and intentional organizational design, providing critical guidance for startups seeking sustainable paths through the complex landscape of open-source software.
If you’re a founder navigating similar open source monetization challenges, I'd love to hear your experiences. The strategies that worked for InfluxDB might not be the perfect fit for your specific context, but there's immense value in comparing approaches across different domains and market segments. Whether you're just starting to evaluate an open source model or already deep in the trenches of balancing community engagement with commercial viability, reach out at priyanka@work-bench.com.