Over the past three years at Swarm Capital, my team has reviewed more than 800 seed-stage SaaS companies. We have invested in eight of them. We have turned down 792 companies that, on paper, looked like reasonable bets: strong founding teams, interesting markets, plausible product roadmaps. The gap between what we passed on and what we invested in is not obvious from the outside. Explaining it clearly requires me to share the analytical framework we have developed — and to acknowledge where that framework has been wrong.
What follows is the distillation of three years of pattern recognition across more than 200 SaaS companies that we tracked from seed stage through their next funding round or beyond. Some of the patterns we found confirm conventional wisdom. Others are genuinely counterintuitive. All of them have affected how we make investment decisions today.
The Conventional Wisdom That Is Mostly Right
Let us start with what the data confirms about conventional wisdom, because not everything that is well-known is wrong.
Net Revenue Retention is the single most predictive metric for long-term SaaS success. Companies with NRR above 120% from their first cohort of paying customers consistently outperform on every dimension: revenue growth, capital efficiency, talent acquisition, and eventual exit multiple. NRR above 120% means that your existing customers are growing their usage faster than you are losing customers — a fundamentally different business model than one where you are running to stay still. Companies with strong NRR in their first year of revenue almost never fall below 100% NRR later; companies that start below 100% NRR in year one rarely recover without a fundamental product pivot.
The implication for seed-stage founders: optimizing for NRR from the first enterprise customer is more important than optimizing for growth rate. A 10-customer cohort with 130% NRR is a vastly more valuable signal than 50 customers with 85% NRR. Slow down, go deep with your first customers, make them extraordinarily successful, and the growth rate will follow.
Category creation beats category disruption. SaaS companies that define a new category consistently generate better outcomes than those that are competing for share in an existing, well-defined market. The market dynamics are simply more favorable: category creators get to define the evaluation criteria, education costs are borne by the market broadly rather than by any single vendor, and the first credible player in a new category often retains pricing power for years.
The counterintuitive implication: the best seed-stage SaaS companies often do not look impressive in the early going precisely because they are creating a category. There is no existing budget line for the problem they solve. Customers need to be convinced they have the problem before they can be convinced to buy the solution. The sales cycles are longer and harder than mature-category sales. But the companies that survive this early friction often emerge as monopolists of the category they created.
Five Factors That Most Investors Undervalue
1. The Shape of the First Ten Customer Cohort
Most investors look at the number of customers and the aggregate revenue. We look at the shape of the first ten customer cohort: specifically, how similar or different those customers are to each other, and how the company talks about them.
Companies that have found product-market fit tend to describe their first ten customers in very similar terms — "we are solving [specific problem] for [specific type of company] and all ten of our customers are [type A]." Companies that have not found PMF tend to describe a heterogeneous mix: "We have a healthcare company, a fintech company, a retail company, and a logistics company — the product works for all of them." The heterogeneous cohort is not evidence of a large market; it is evidence that the company has not yet figured out for whom its product works best and why.
The best SaaS companies in our portfolio had a homogeneous initial customer cohort and a clear hypothesis about why those customers shared the problem they were solving. The companies that struggled had heterogeneous initial cohorts and either could not articulate or were not honest about why the customers were different from each other.
2. The Urgency Structure of the Pain
Not all customer pain is created equal. Enterprise software buyers tolerate a lot of pain for a long time — inefficient processes, suboptimal tools, manual workarounds — before they pull the trigger on a new vendor relationship. The SaaS companies that break through are those that are solving pain that buyers feel urgently and immediately, not pain that they acknowledge intellectually but deprioritize in practice.
We evaluate the urgency structure of customer pain by asking specific questions in reference calls with a company's customers. "When did you first realize you needed this product?" and "What would have happened if you had not purchased this product?" are the two questions that tell us the most. If customers cannot articulate a specific moment of urgency and a clear counterfactual cost of not purchasing, the buyer urgency is probably lower than the founder believes.
The companies in our portfolio with the best early growth had customers who described their pre-purchase pain in visceral terms: "We were losing customers because we could not detect churn early enough," "Our compliance team was working 80-hour weeks and still finding issues in audits," "We had to turn down three enterprise deals because our data governance posture was not audit-ready." Specific, costly, urgent pain creates the enterprise buyer motivation that drives short sales cycles and strong initial expansion.
3. The Go-to-Market Motion That Is Already Working
Seed-stage SaaS companies almost always have a story about how they will sell their product. The story is usually "we will hire a VP of Sales and build an outbound motion" or "we will grow through product-led growth" or "we will rely on channel partnerships." These are narratives. What we look for is evidence of a go-to-market motion that is already working — even if it is working in a non-scalable way.
The companies that succeed fastest after their seed round invariably have at least one acquisition channel that works. It might be founder-led sales to personal networks. It might be a specific vertical community where the founders are respected voices. It might be an inbound channel driven by content or community that the founders have been building for years. Whatever the channel, the existence of a working acquisition motion — even a small one — is a strong predictor of what the company will be able to scale after the seed round.
Conversely, the companies that struggle post-seed are disproportionately those where the founders believe that hiring a VP of Sales will solve a go-to-market problem that is actually a product-market fit problem in disguise. No VP of Sales, however talented, can sell a product that customers do not urgently need. The role of the seed round is to validate product-market fit, not to hire into it.
4. The Technical Architecture Decision Made at Seed Stage
This one genuinely surprised us. The technical architecture decisions that SaaS companies make at the seed stage have an outsized impact on their ability to scale — not just technically, but commercially. Companies that build for enterprise readiness from the beginning — multi-tenancy, role-based access controls, SOC 2 compliance posture, API-first architecture, and audit logging — consistently close enterprise deals faster than companies that build these capabilities as an afterthought.
Enterprise buyers evaluate new SaaS vendors through an increasingly rigorous security and compliance lens. The security review process that might have taken two weeks in 2019 now routinely takes three to six months at large enterprises and requires documentation that many early-stage companies simply cannot produce because they built their product without enterprise security in mind. Companies that skipped this work at the seed stage pay for it in long sales cycles, lost deals, and expensive re-architecture later.
The counterintuitive implication: spending 20-30% of your initial engineering resources on enterprise-grade infrastructure is not over-engineering — it is sales acceleration. The seed-stage companies in our portfolio that made this investment are closing enterprise deals at 2-3x the rate of comparably positioned companies in our deal pipeline that did not.
5. The Founder's Theory of Defensibility
The fifth factor is perhaps the most important and the hardest to evaluate: the founder's theory of why their company will be defensible at scale. Specifically, what will make it hard for a well-capitalized competitor to take their market in five years?
Most seed-stage founders have a plausible answer to the surface-level version of this question: "we have a better product," "we are first movers," "our team has unique domain expertise." These are not theories of defensibility — they are descriptions of current advantages that will erode over time. A real theory of defensibility explains the mechanism by which the company's advantage compounds rather than diminishes as the market develops.
The best founders we have backed have defensibility theories that are specific, mechanistic, and tied to data: "Our model improves with every transaction processed, and after 100,000 transactions, our accuracy will be far ahead of any model trained on generic data." "Our product creates workflow integrations that take an average of nine months to build internally, so switching costs increase rather than decrease as customers use more of our platform." "We are building the canonical dataset for this domain, and the first mover advantage in data network effects is durable."
These theories matter not just for investor confidence but for company-building decisions. Founders who have thought clearly about their defensibility theory make better product decisions, better hiring decisions, and better go-to-market decisions — because every decision can be evaluated against the question of whether it strengthens or weakens the defensibility thesis.
The Patterns We Got Wrong
Intellectual honesty requires acknowledging where our framework has been wrong, or at least incomplete.
We have underweighted the importance of timing. Some of the best SaaS companies succeed not primarily because of the superiority of their product or team, but because they are in the right place at exactly the right moment — when a new regulatory requirement creates sudden demand, when a major technology shift creates a new integration layer, when a macro event permanently changes enterprise buying behavior. We have been too slow to invest in companies riding genuine timing tailwinds because we were looking for evidence of deep defensibility that was unnecessary given the timing advantage.
We have overweighted early revenue as a quality signal. Early enterprise revenue is often a signal of sales skill rather than product-market fit. A talented founder who can close deals through force of personality and network can build a pipeline of early customers that looks like PMF but actually reflects a customer set that is not representative of the scalable target market. We have had investments where the first five customers looked like strong validation but turned out to be anomalously close to the founders and not representative of the broader market. We now do more work to understand why customers bought and whether the buying motivation is generalizable.
We have occasionally been wrong about market timing. There are companies we passed on because the market seemed too early that have subsequently proven to be right. The lesson is that founders who are building in markets that are one or two years away from their inflection point need more analytical attention, not less. The seed stage is precisely when you should be backing companies in markets that are not yet obvious, because by the time the market is obvious, the best seed opportunities are gone.
What This Means for Founders Raising Seed Rounds
If you are a SaaS founder preparing to raise your seed round, the frameworks above have direct practical implications for how you should position your company and what evidence you should be gathering before approaching investors.
Know the shape of your first cohort. If you have initial customers, be able to describe clearly what they share in common and why they bought your product. The investor's question about your first customers is not really a question about revenue — it is a question about your ability to identify and systematically reach the customers for whom you solve the most important problem.
Build the enterprise foundation early, even if your initial customers are not enterprise. The incremental cost of building with enterprise-grade security and architecture at the seed stage is far smaller than the cost of retrofitting it later. If you believe enterprise will eventually be your primary market, treat yourself as an enterprise company from day one of product development.
Develop a specific theory of defensibility. Not "we will be defensible because we have a great team" — but a specific mechanistic explanation for why your advantage will compound rather than erode over time. This exercise will sharpen your product strategy, your hiring priorities, and your go-to-market approach, in addition to making your investor conversations more substantive.
Show evidence of urgency, not just validation. Customer validation is table stakes. What distinguishes the companies that investors get excited about are customers who are buying urgently, expanding rapidly, and articulating specific, costly counterfactuals for what would have happened if they had not purchased. Find two or three of these customers before your first investor meeting. Their stories will do more work for your fundraise than any financial model.
The SaaS market is large, growing, and genuinely good for well-positioned companies. But the bar for what constitutes a fundable seed-stage opportunity has risen substantially over the past three years. The companies that succeed are those that approach the seed stage with the same rigor that they will eventually apply to their enterprise sales motion: clear target customer, specific pain, evidence of urgency, and a product that is genuinely better — not just slightly better — than what customers have been using before.
