At Gravis Robotics Capital, we evaluate hundreds of robotics and automation companies each year. The process of assessing these companies — understanding which ones represent genuine investment opportunities and which ones, despite being technically impressive or addressing large markets, are not yet ready for institutional capital or not a fit for our specific thesis — is one of the core disciplines of our work as investors.

We are sharing our evaluation framework not to give founders a checklist to optimize their pitches against, but because we believe transparency about how we think helps the founders who are genuinely building extraordinary companies present themselves more effectively, and helps founders who are not yet ready to know specifically what they need to work on before they approach institutional seed investors. This is a framework for honest self-assessment as much as it is a description of our investment process.

Dimension One: Technical Differentiation and Defensibility

The first dimension we evaluate is technical: what has this team built, and is it genuinely differentiated from what exists or what could be built by a well-funded competitor? For robotics companies, this question has hardware and software dimensions that interact in complex ways.

At the seed stage, we do not require that a company's technical differentiation be fully proven in commercial conditions. What we do require is a clear and credible hypothesis about where the technical moat lies, evidence that the founding team has the depth of expertise to realize that hypothesis, and a technology architecture that is plausibly defensible as the company scales. The hypothesis must be specific: "we have developed a novel grasp planning algorithm that achieves 15% higher success rates than the state of the art on our target object class" is a specific, testable claim. "Our approach to manipulation is fundamentally different" is not.

We look hard at patent landscapes, not because we believe patents alone create defensibility in fast-moving technology markets, but because the process of building a patent portfolio reveals whether a team has thought carefully about what is truly novel in their approach and has identified claims that are specific enough to be enforced. Teams that have done serious patent work understand their technical differentiation more clearly than those that have not.

Data moats are an increasingly important form of technical defensibility in AI-driven robotics. Companies that are building proprietary datasets through their commercial deployments — datasets of grasping attempts and outcomes, inspection images with labeled defects, robot trajectories with annotated task contexts — are creating competitive assets that improve their systems and are difficult for competitors to replicate. We specifically ask founders how their system gets better with deployment, and what proprietary data assets they expect to accumulate.

Dimension Two: Market Sizing and Entry Strategy

The second dimension is market: is this a large enough opportunity to build a venture-scale company, and does this team have a credible strategy to access it? Both parts of the question matter. Many technically impressive robotics companies are building for markets that are too small or too consolidated to support venture-scale returns. And many founders have clear views on the eventual size of their market but are vague or unrealistic about the specific path from zero to meaningful revenue.

On market sizing, we prefer bottoms-up calculations grounded in specific customer economics over top-down market research reports. A founder who can tell us "there are 3,200 mid-size food processing facilities in North America that fit our target profile, each with $80,000 in annual value creation potential from our system, implying a $256M annual revenue opportunity in North America alone" has done better market analysis than one who cites a report projecting the global food processing automation market at $12 billion by 2030.

On market entry strategy, the key question is whether the team has a specific, credible path to their first 10 to 20 customers. We ask founders to name the specific facilities or companies they are targeting, explain why they believe those customers are most likely to convert, and describe the specific relationships or domain expertise advantages they have in accessing those customers. Generic go-to-market plans that rely on trade shows, broad outbound marketing, and channel partnerships that have not yet been established are red flags at the seed stage.

Dimension Three: Team Composition and Capability

At the seed stage, team is the dimension that correlates most strongly with our conviction. The technology and market will evolve — often in ways that require the founding team to make significant pivots in product design, market positioning, or business model. The team's capability to navigate those pivots, to learn from failure without losing morale or cohesion, and to recruit and retain the specialized talent that a robotics company needs is the most reliable predictor of success that we can assess at the seed stage.

Our team assessment begins with technical depth. Does the founding team have genuine expertise in the core technical disciplines required to build their specific product? For a robotic manipulation company, we want to see expertise in mechanical design, controls engineering, and computer vision at minimum. For an outdoor autonomous vehicle company, we want to see expertise in sensor fusion, localization, and software systems engineering. We are skeptical of teams where all the technical expertise is concentrated in one founder — the complexity of building a physical AI product typically exceeds any one person's span of mastery.

Commercial capability matters equally to technical capability, and it is often underrepresented on founding teams. The ability to identify and cultivate enterprise customer relationships, to negotiate complex pilot and commercial agreements, and to build the trust with operations leaders that is necessary for a robotics company to succeed in enterprise markets is not a skill that can be hired in after the fact. We look for evidence that founding teams have thought seriously about their commercial capabilities and have plans to address any gaps, whether through additional co-founder recruitment, early commercial hires, or operating advisor networks.

Dimension Four: Capital Efficiency and Milestone Planning

Hardware robotics companies need capital, and at the seed stage, the relevant question is whether the founding team has a realistic plan for using seed capital to reach milestones that will justify their next raise on favorable terms. This requires a detailed understanding of what will be built with seed capital, what milestones that work product will enable, and how long it will take.

We are concerned by seed-stage plans that call for large hardware build programs before customer validation, that assume commercial revenue will arrive faster than enterprise sales cycles realistically allow, or that require more capital than the seed round provides without a clear strategy for bridge financing if timelines extend. These are all common patterns in early robotics company financial planning, and they often reflect optimistic assumptions that serial hardware founders know to discount significantly.

The milestone plan also tells us something about the team's understanding of their own technology risk. A team that articulates specific, measurable technical milestones — "achieve 92% pick success rate on our target SKU catalog in a controlled environment by month six, then 90% in a live customer facility by month twelve" — demonstrates the kind of rigorous engineering thinking that we want to see. A team that describes milestones in vague terms like "product-market fit" or "technical readiness" is not thinking precisely enough about what success looks like.

Dimension Five: Competitive Landscape Awareness

The final dimension of our evaluation is competitive awareness. Does this team have a clear and honest view of what already exists in their target market, who is likely to enter, and what their specific advantages are relative to both current and future competition? Robotics founders sometimes make the mistake of dismissing competition, either by claiming there is none (which is almost never true in large markets) or by assuming that their technical approach is so superior that competition is irrelevant. Neither posture reflects the reality of competitive dynamics in commercial markets.

What we look for is an honest competitive analysis that acknowledges the strengths of existing solutions, identifies specific ways in which the company's approach is differentiated and how those differences translate into customer value, and has a view on how the competitive landscape will evolve as the market matures. This analysis should include both current direct competitors and the risk of competition from large incumbents or well-funded new entrants who might enter the market if it proves large enough.

Key Takeaways

  • Technical differentiation must be specific and defensible — general claims about novel approaches without specific, testable claims are insufficient at seed stage.
  • Bottoms-up market sizing grounded in specific customer economics is more credible than top-down market research projections.
  • Team capability in both technical and commercial dimensions is the strongest predictor of success and the primary focus of our early-stage conviction.
  • Seed capital plans must include realistic milestone planning based on actual hardware development and enterprise sales timelines.
  • Honest competitive awareness that acknowledges existing solutions and articulates specific differentiation is a signal of founder maturity and market understanding.

Conclusion

The companies that meet our bar across all five dimensions — technical differentiation, market sizing, team capability, capital efficiency, and competitive awareness — are relatively rare. But they exist, and finding them is one of the most rewarding aspects of our work at Gravis Robotics Capital. If you are building a robotics or automation company and believe you are building something genuinely exceptional, we want to hear from you. Our $115M Seed Round is specifically designed to back companies like the ones described in this framework — and we have the capital, expertise, and networks to be the most valuable partner for those companies from day one. Reach out to us to start the conversation, and explore our portfolio to understand the kinds of companies we back.