← Back to Insights

Seed-Stage Investing in Enterprise Software: A Framework for Pattern Recognition

Seed-stage investing in enterprise software is fundamentally a pattern recognition exercise. You cannot build a discounted cash flow model on a company with no revenue. You cannot apply comparable company analysis to a product that has not yet shipped. What you can do — if you have done it enough times — is recognize the patterns that have historically separated enterprise software companies that eventually became category leaders from those that stalled, pivoted, or quietly wound down.

After years of investing in enterprise software at the earliest stages, we have developed a framework for this pattern recognition that shapes every investment decision ROI AI Capital makes. We are sharing it here not because we think it is the only valid approach, but because we believe transparency about how we think helps founders understand whether our fund is the right partner for them at this stage of their journey.

The Team Dimension: Looking for Earned Insight

In enterprise software investing, the most important founding team attribute is not intelligence, work ethic, or charisma — though all three matter. It is what we call earned insight: a deep, specific, operationally grounded understanding of the problem they are solving that could only have been acquired through direct experience.

Enterprise software problems are not obvious from the outside. The pain points that CHROs feel around workforce planning are not visible to someone who has never worked inside an HR organization trying to model headcount scenarios under multiple strategic scenarios. The frustrations of enterprise CIOs trying to integrate new point solutions into existing security and identity infrastructure are not apparent to founders who have only built consumer products.

Earned insight is recognizable in how a founder describes their problem. When a founder says "in my experience running an HR operations team at a 10,000-person company, the thing we struggled most with was X, and here is exactly why every existing solution failed us" — that is earned insight. When a founder says "research shows that HR leaders want better analytics tools" — that is not.

We also look for founding teams that include a credible commercial archetype alongside the technical founder. Enterprise software sales is a specific skill set. The path from a compelling demo to a signed multi-year contract at a Fortune 500 company involves procurement cycles, security reviews, legal negotiations, champion management, and executive alignment that most technical founders have not navigated before. Teams that include someone who has actually done enterprise sales — who knows what "legal red lines" means, who has dealt with a 90-day security review — have a significant early advantage.

The Market Dimension: Seeking Underpriced Disruption

Our market evaluation framework looks for what we call underpriced disruption: situations where a large, established software category is in the process of being disrupted by a new technology architecture, but where most investors have not yet recognized the full magnitude of the opportunity.

The clearest current example of this pattern is the AI disruption of legacy enterprise software. Core HR systems, ERP platforms, talent management suites — these are massive categories with enormous installed bases and long replacement cycles. But many of them were built on architectures designed in the 2000s that are fundamentally ill-suited to the kind of AI-native intelligence that enterprise buyers are now demanding. The replacement cycle is beginning, and it is beginning at the edges: new point solutions that deliver AI-native value in specific workflows, gradually expanding their footprints as they prove their worth.

We look for markets where:

• The incumbent solutions are technically outdated but commercially entrenched — creating a gap between what buyers need and what they can currently buy
• The regulatory or competitive environment is creating new urgency — data privacy requirements, pay transparency laws, and skills scarcity are all driving enterprise buyers toward new solutions
• The TAM is large enough to support multiple successful businesses — we are looking for markets where a $100M ARR business represents less than 5 percent market share
• The technology shift is durable, not a cycle — AI-native enterprise software is not a feature upgrade, it is a platform shift

The Product Dimension: Architecture as Competitive Moat

In consumer software, the most durable competitive moats are often network effects — the product becomes more valuable as more people use it. In enterprise software, the most durable moats are more often architectural: the product is built in ways that create deep integration into customer workflow, making replacement costly and painful.

At seed stage, we evaluate product architecture through several lenses:

Integration depth: Does the product integrate into the systems, data sources, and workflows that enterprise customers already use, or does it require a clean-room implementation? Products that integrate deeply with existing infrastructure — SSO providers, HRIS systems, communication tools, data warehouses — are harder to displace than standalone tools.

Data flywheel: Does the product improve as more data is added to it? AI-powered products that become smarter and more accurate as customers use them have a structural advantage over competitors that do not benefit from customer data accumulation.

Workflow centrality: Does the product sit in the center of a workflow that users access daily, or does it serve a peripheral function that is easy to route around? Products that are in the critical path of daily work — where people open the tool first thing in the morning and last thing at night — create stronger switching costs than periodic-use tools.

The Commercial Dimension: Early Revenue Quality

Even at seed stage, the quality of early commercial traction tells us a great deal about whether a company is on a path to building a sustainable enterprise business. We look at several specific indicators:

Who is buying, not just how much: A handful of design partners who are paying meaningful amounts — not token proof-of-concept fees — tells us that the product is solving a real problem with enough urgency that buyers are willing to allocate budget. The specific titles and seniority of early buyers also matters: a CHRO signing off on an early contract is a different signal than a department head making an under-the-radar purchase.

The nature of the conversation: Are enterprise buyers engaging with the product as a curiosity or as a solution to a specific, budgeted need? The best early enterprise sales conversations are ones where the buyer has a specific problem, has been trying to solve it with inadequate alternatives, and sees the new product as genuinely better — not just new.

Expansion signals: Even in very early deployments, expansion signals — users expanding their usage beyond the initial scope, requesting additional features, asking about team seats — are powerful indicators of product-market fit. They suggest that the product is delivering value that justifies deeper investment from the customer.

Our Investment Decision Process

When we evaluate a company against this framework, we are not looking for perfection on every dimension. We are looking for exceptional strength on at least two dimensions and no fatal weaknesses on the others. A team with extraordinary earned insight in a market that is clearly undergoing technology disruption can earn conviction despite an early-stage product that still needs work. A beautiful product that solves a clearly real problem can earn conviction even when the founding team's commercial experience is thin, if the team demonstrates awareness of that gap and has a credible plan to address it.

What we cannot work around: founders who have not actually experienced the problem they are solving in any meaningful way; markets where the technology disruption is already at a point of competitive clarity and the best-funded competitors have obvious advantages; and products that require extensive customization for every customer in ways that make gross margin improvement fundamentally difficult.

If you are a founder building in enterprise software or HR technology, we would welcome the conversation. Reach out to us directly, and tell us about your earned insight.