When we made our first investments in early 2020, we were betting on founders and ideas that were, by definition, unproven. The companies we backed had no revenue, in some cases no product, and in every case a market that did not yet fully appreciate the value of what they were building. Five years on, three of those early investments have become market-defining companies whose stories offer lessons that are relevant to every founder building in enterprise technology today.
The stories of NexusAI, CloudShield, and DataBridge are not straightforward success narratives. Each company faced moments of genuine crisis - moments where a different choice, a different piece of advice, or a different response to adversity might have led to failure. Understanding how these companies navigated those moments is, we believe, more instructive than simply celebrating their outcomes.
NexusAI: The Pivot That Changed Everything
We invested in NexusAI in February 2020 based on our conviction in the founding team and their vision for applying AI to enterprise workflow automation. At the time of our investment, the company had a compelling technical demonstration and early interest from several enterprise prospects, but no paying customers.
The original product vision was ambitious: a general-purpose AI platform that enterprise customers could use to automate any repetitive workflow. The founding team believed - correctly, as it turned out - that the underlying technology was capable of addressing an enormous range of use cases. The problem was that "automate any workflow" is not a message that resonates with enterprise procurement teams. It is too abstract, too hard to evaluate, and too difficult to assign a specific ROI to.
After six months of struggling to close enterprise pilots that turned into paying customers, the NexusAI team made a critical decision: they would abandon the general-purpose platform story and go deep on a single specific use case - document processing for financial services firms. This was a painful decision. It meant ignoring dozens of prospects in other industries who had expressed interest. It meant narrowing the addressable market they were pitching to investors and potential hires. It meant betting that going narrow would ultimately enable them to go wide.
The results vindicated the decision. Within 90 days of pivoting to financial services document processing, NexusAI had signed three enterprise customers and was generating $180,000 in annual recurring revenue. More importantly, these customers were deeply engaged with the product, providing detailed feedback that dramatically accelerated product development. The specificity of the use case allowed NexusAI to build a product that was genuinely 10x better than anything else available for financial document processing, rather than slightly better than everything for every use case.
The lesson for founders is one that we repeat constantly: the fastest path to scale often runs through a narrow initial focus. The companies that try to serve everyone from day one typically serve no one particularly well. The companies that identify the smallest possible customer segment where they can be genuinely best-in-class, dominate that segment completely, and then use that success as a springboard to adjacent markets, consistently outperform.
Today, NexusAI has expanded well beyond financial services document processing. Its AI workflow platform serves 400+ enterprise customers across 22 countries in industries from healthcare to logistics to legal services. The core technology developed for financial document processing became the foundation for products addressing dozens of adjacent workflows. NexusAI's Series C round at a $1.2B valuation, completed in early 2021, validated both the specific product and the broader vision.
CloudShield: The Competitive Threat That Became a Forcing Function
CloudShield's story is a case study in how to respond to a competitive threat that initially appears existential. In mid-2021, approximately 18 months after our initial investment, a well-resourced competitor announced a product that appeared to directly compete with CloudShield's core zero-trust security offering, backed by a $40M Series A and a pedigree team from one of the leading enterprise security companies.
The CloudShield team's initial response was panic - understandable, given that the competitor's announcement came with significant press coverage and caused two enterprise prospects to pause their CloudShield evaluations. For a brief period, it appeared that the company might be unable to close the revenue needed to make its next milestone and raise a Series A of its own.
What happened next was a masterclass in competitive positioning. Rather than trying to compete head-to-head with the new entrant on the same feature set, CloudShield's founders spent two intense weeks interviewing their existing customers and their strongest prospects to understand exactly what differentiated their product in ways that genuinely mattered. The answer was surprising: what CloudShield's customers valued most was not its technical architecture, but its implementation methodology. CloudShield had developed a deployment process that allowed enterprise customers to implement zero-trust security in 30 days or less, compared to the industry norm of 6-12 months. This rapid deployment capability was something the new competitor - built on a more technically sophisticated but also more complex architecture - simply could not match.
CloudShield reoriented its entire go-to-market around this differentiation: "Zero-trust security in 30 days, guaranteed." It was a specific, verifiable claim that enterprise security buyers could easily evaluate, and it addressed one of their primary objections to adopting zero-trust architecture. Within six months of making this positioning shift, CloudShield had signed 40 new enterprise customers and had dramatically outpaced the well-funded competitor in market traction.
The story did not end there. CloudShield's rapid deployment methodology turned out to be not just a marketing message but a genuine technical moat. The engineering insights developed in the process of reducing deployment time to 30 days created architectural advantages that allowed CloudShield to deliver better performance at lower cost than competitors. These advantages compounded over time, ultimately contributing to the product superiority that drove the company's NASDAQ IPO in March 2024 at a $2.8B market cap.
DataBridge: Building the Business Model That Matched the Product
DataBridge's story illustrates a challenge that is specific to infrastructure and platform businesses: finding the business model that correctly captures the value the product creates. When we invested in DataBridge in March 2020, the company had a technically impressive product - a real-time data integration layer that connected enterprise data silos and made unified analytics genuinely possible. What it did not have was a clear answer to the question of how to monetize that value.
DataBridge's initial pricing model was based on the number of data connections, charging customers per integration. This seemed logical, as connections were the primary driver of infrastructure cost for DataBridge. The problem was that it created perverse incentives: customers had strong reasons to minimize the number of integrations they set up, which limited both DataBridge's revenue per customer and - more importantly - the value customers extracted from the product. A customer who connected only their CRM and their ERP got much less value than a customer who connected all 20 of their data sources, but under the connections-based pricing model, connecting more sources meant higher cost.
After analyzing churn patterns and expansion revenue data, DataBridge's team identified the metric that best predicted customer lifetime value: the number of analytics queries run per month. Customers who used DataBridge heavily - who ran thousands of queries per month across their unified data layer - derived enormous value and almost never churned. Customers who set up a few integrations and ran queries infrequently were at much higher risk of churning or reducing usage.
DataBridge switched to a usage-based pricing model tied to query volume, with generous allowances that ensured most customers saw their monthly bills decline in the first year of the transition. The result was initially painful for revenue - total ARR dropped 15% as some customers moved to lower-tier plans. But customer engagement increased dramatically as the pricing friction that had discouraged heavy use was removed. Within 12 months, DataBridge had not only recovered the lost ARR but had grown to 3x the pre-transition ARR, driven by expansion within existing accounts.
The lesson is one that applies across B2B software: the optimal pricing model aligns customer incentives with product usage in a way that creates a natural expansion engine. When customers pay more because they use the product more - and they use the product more because it delivers more value - you create a virtuous cycle where revenue growth and product value creation are self-reinforcing. Getting this alignment right is not simple, but it is one of the highest-leverage decisions a B2B software founder can make.
Common Threads: What These Stories Tell Us
Looking across these three portfolio stories, several themes emerge that we believe are broadly applicable to founders building enterprise technology companies.
First, the best companies are built by founders who are genuinely willing to change their minds based on evidence. NexusAI pivoted its product positioning, CloudShield reconceived its competitive advantage, DataBridge redesigned its business model. None of these changes were easy. All of them required the founders to acknowledge that their initial approach was not working and make difficult, costly changes. This intellectual flexibility, combined with the conviction to fully commit to the new direction once decided, is one of the most important attributes we look for in founders.
Second, competitive threats are often forcing functions that push companies to understand their true differentiation more clearly. CloudShield's story is an extreme example, but we have seen this pattern repeatedly: the existence of credible competition forces founders to answer the question "why should a customer choose us?" with specificity and rigor. Companies that never face meaningful competition sometimes lack this clarity, and it eventually catches up with them.
Third, the most valuable learning in early-stage B2B software comes from customers who are deeply engaged with the product. Shallow pilots and brief evaluations teach you relatively little. Customers who use the product extensively, who have strong opinions about what it does well and poorly, and who are willing to invest their own time in making it better - these are the customers who will make your company. Identifying and cultivating these relationships should be a primary focus for every early-stage B2B founder.
