Every investment thesis eventually meets its test case — the company that either validates or complicates the framework you have been building. For BeMoreeDriven Capital's enterprise AI thesis, NeuralFlow is that company. When we co-led the $45M Series B alongside General Catalyst in February 2025, we were making a specific bet about where the enterprise software market is heading and why NeuralFlow is positioned to be one of its defining companies. This article is an attempt to explain that bet in detail.
We share investment reasoning publicly not because we are indifferent to competitive dynamics — we are not — but because we believe the venture capital market benefits from more transparency about how sophisticated investors think about AI-era enterprise software. Founders, co-investors, and the companies themselves deserve to understand the logic behind the capital allocations that shape their trajectories.
The Enterprise AI Landscape in Early 2024
To understand the NeuralFlow investment, you first need to understand the state of enterprise AI in the period when we began our diligence — roughly the first half of 2024. The market was saturated with AI point solutions: tools that automated a single workflow, answered questions about a document corpus, or generated content from a template. The narrative was AI-for-everything, and the investment activity reflected it. Seed rounds for AI workflow tools were closing at pre-revenue valuations that would have seemed implausible three years earlier.
But beneath the froth, something more interesting was happening. Enterprise IT and operations executives — the buyers who actually control budgets at Fortune 500 companies — were beginning to articulate a different set of requirements. They were not looking for more AI point solutions. They were drowning in them. What they needed was an orchestration layer: a platform that could coordinate multiple AI models, execute complex multi-step workflows, integrate with their existing enterprise systems, and maintain the auditability and governance requirements that their compliance and legal teams demanded.
This was the problem NeuralFlow was built to solve. And when we found them, they had already solved it for 22 Fortune 500 companies.
What NeuralFlow Actually Builds
NeuralFlow's core product is an autonomous workflow orchestration platform. The description sounds abstract, so let us be concrete. A typical NeuralFlow deployment at an enterprise customer might look like this: a global manufacturer needs to process purchase orders, validate against approved supplier lists, check inventory levels across 14 regional warehouses, generate a procurement recommendation, route for approval based on spend thresholds, and trigger payments in the ERP system — all without human intervention for the 85% of orders that fall within pre-approved parameters.
Before NeuralFlow, this workflow required a combination of manual review, rigid rule-based automation (brittle, expensive to maintain), and multiple point tools that did not communicate with each other. The average processing time was 4-6 business days. NeuralFlow processes the same workflow in under 4 minutes, with a human exception rate of 15% — the same 15% that actually require human judgment.
What makes this technically possible — and technically defensible — is NeuralFlow's integration layer. The platform maintains certified, maintained integrations with over 140 enterprise systems: SAP, Salesforce, Oracle, Workday, ServiceNow, Microsoft Dynamics, and dozens of industry-specific platforms. Building this integration layer took NeuralFlow's engineering team three years of focused work. It cannot be replicated in six months by a better-funded competitor with a faster model. The integrations are NeuralFlow's primary technical moat.
The Signal: From $6.2M to $22.4M ARR
When we first saw NeuralFlow's metrics, the number that stopped us was the ARR growth trajectory. The company had grown from $6.2M ARR at their Series A to $22.4M at the time we began Series B diligence — a 3.6x increase in approximately 18 months. That pace of ARR growth is unusual at the growth stage. But the number we found more compelling was the NRR: 138%.
An NRR of 138% means that NeuralFlow's existing customer base — excluding any new customer additions — grows its revenue with the company by 38% annually. This is the product of two dynamics: customers deploying NeuralFlow on additional workflow categories (the initial deployment is almost always a single workflow type; typical expansion is to 3-5 workflow categories within 24 months), and usage-based components of the pricing model that grow as the company's transaction volumes increase.
The implication is profound. A company with 138% NRR is growing from its existing base alone at a rate that many growth-stage companies struggle to achieve including new customer acquisition. Add new customer acquisition on top of that base — which NeuralFlow was achieving at roughly 40% of its total ARR growth — and you have a compounding revenue machine that is genuinely unusual.
We validated the NRR figure through customer reference calls with 14 of NeuralFlow's enterprise clients. The conversations confirmed what the numbers suggested: customers were not just renewing, they were actively expanding. The most common phrase we heard in reference conversations was some variation of: "We started with one workflow, and now we cannot imagine operating without NeuralFlow across our procurement, finance, and HR functions."
Why We Led Alongside General Catalyst
Having a co-lead of General Catalyst's caliber in a round is itself a signal worth unpacking. General Catalyst is one of the most disciplined growth investors in the technology market. They do not participate in rounds for defensive or signal reasons. When they co-lead a Series B, they have done their own institutional diligence and arrived at independent conviction about the opportunity. The alignment of our view with theirs — reached through separate processes — was meaningful confirmation of our thesis.
The co-lead structure also reflects a conscious choice about what NeuralFlow needs at this stage of its development. The company is at the inflection point where it needs to build out its enterprise sales organization, expand internationally, and make significant R&D investments in next-generation workflow intelligence capabilities. A $45M round provides the capital runway to do all three without forcing a pace that could damage the culture or the product quality. Having two deeply engaged institutional investors on the board provides the management team with a broader set of operating and market access resources than any single investor could.
The Technical Moat: More Than Just Integrations
We have emphasized NeuralFlow's integration layer as a primary moat. But the full picture of technical defensibility is richer. The platform has three additional dimensions of technical advantage that compound over time.
First, NeuralFlow has built a proprietary workflow intelligence engine that learns from the exception patterns in its customers' deployments. Every time a human reviews a NeuralFlow exception — approving, rejecting, or routing it differently than the system recommended — that decision feeds back into the model. Over time, the exception rate decreases as the system learns the specific judgment patterns of each customer's organization. A competitor entering NeuralFlow's existing customer relationships would start with no training data and no understanding of those specific organizational patterns. This is a true data moat.
Second, NeuralFlow's governance and auditability layer is more sophisticated than any competitor we evaluated during diligence. Enterprise AI deployment in regulated industries requires complete audit trails — every AI decision must be explainable, logged, and accessible for regulatory examination. NeuralFlow's audit architecture was built to meet the requirements of financial services and healthcare regulators, which are the most demanding in the enterprise market. This compliance capability is not a feature — it is a prerequisite for selling to the regulated enterprises that represent NeuralFlow's core customer base.
Third, the platform's workflow composition interface allows business analysts — not just IT engineers — to build and modify automated workflows. The democratization of workflow automation to non-technical users dramatically expands the addressable surface area within each customer organization and accelerates deployment timelines. The average time from contract signature to first production workflow at NeuralFlow is six weeks. The industry average for comparable enterprise automation deployments is six to nine months.
The Road Ahead
The investment thesis for NeuralFlow's Series B is essentially a conviction that the company is at the beginning of a multi-decade market leadership position in enterprise workflow intelligence. The near-term milestones are straightforward: continue ARR growth above 40% annually, expand international operations into European and Asia-Pacific markets, grow the enterprise sales team from its current 45 to approximately 120 over 18 months, and extend the integration library to 200+ enterprise systems.
The medium-term opportunity is more transformative. As AI capabilities continue to advance, the boundary of what can be reliably automated will expand significantly. NeuralFlow's platform is designed to incorporate increasingly sophisticated AI reasoning capabilities as they become commercially available. The companies that will lead the autonomous enterprise era will not be foundation model builders — they will be the orchestration layer companies that translate AI capabilities into enterprise-grade operations at scale. That is where NeuralFlow is positioned.
The total addressable market for enterprise workflow automation and AI orchestration is estimated at $280B by 2030, according to IDC. We believe NeuralFlow is positioned to capture a disproportionate share of that market given its technical differentiation, its existing enterprise relationships, and the compounding advantage of its workflow intelligence data assets. It is rare to find a growth-stage company with this combination of proven commercial traction, genuine technical moat, and a market opportunity of this scale. We are proud to be partners in the journey.
"NeuralFlow did not just pitch us on the vision of the autonomous enterprise. They showed us 22 Fortune 500 deployments where it was already real. That is the kind of evidence that converts a thesis into a conviction." — James Alderton, Managing Partner, BeMoreeDriven Capital
For founders building in adjacent categories — enterprise AI orchestration, workflow intelligence, or enterprise system integration — NeuralFlow's trajectory illustrates the kind of proof points that growth investors find compelling: an NRR above 130%, an integration moat that took years to build, and a customer reference base that sells itself. If you are building something analogous, we would welcome the conversation.