The forgotten layer in startup ecosystems: investor capacity

Last month, we discussed the need to develop the funding sector in parallel with entrepreneurs, and asked what it would look like to build venture capital communities with the same intentionality that many regions now apply to talent and founder development. In this month’s newsletter, we dive a little deeper into that theme, arguing that policymakers should be more focused on capacity building than capital deployment, and may benefit from viewing investors more like startups themselves.

Seed and early stage funding is often the limiting factor within an ecosystem. It is invariably local in nature, meaning that the investment capacity should ideally be built within the ecosystem.

One issue is that other asset classes, like property, may be more attractive for investors. Tax incentives, including loss reliefs, can potentially help make venture investing more attractive in comparison – but this is only part of the picture.

The bigger challenge is that effective early-stage investing is a skill in its own right: just as tech companies need judgement to identify product-market fit and the appropriate strategy for their firm, early-stage investors must form their own judgements about the startups in which they invest – often in the absence of any real product or customer traction. These skills take time and experience to acquire. And just as entrepreneurs rely on networks for talent and industry insights, investors need networks through which they can source and evaluate dealflow, which again takes time and effort to cultivate.

In the AI-Native era, that skill set is changing. Investors are increasingly evaluating companies whose products can be built faster, with smaller teams, and at far lower cost than in previous technology waves. That can widen the top of the funnel, but it also makes speed more critical and discernment harder: when technical barriers become easier to overcome, investors must place greater weight on founder learning speed, distribution capability, trust, and whether early revenue reflects durable demand rather than short-lived experimentation. Early traction might be a misleading signal if there are few barriers to users moving between models.

This shift is already visible in how quickly companies are reaching early funding stages. As we highlighted in January, across leading ecosystems, AI-Native startups are reaching seed stage significantly faster than other tech companies — compressing timelines and increasing pressure on investors to make earlier, higher-stakes decisions. This suggests that speed of decision-making gives AI investors an advantage and clearly benefits startups, too. A core question for investors and ecosystem leaders is how this might be encouraged.

Additionally, angel groups and early-stage investment firms are quite fragile entities: research by the Angel Resource Institute shows that early attrition rates are high. EBAN estimates that typical Angel networks require 5–7 years to reach maturity, but many early-stage networks dissolve before reaching that point because they fail to become financially self-sustaining. Many early-stage VC firms fail after their first fund, with around half of VC firms never raising a second fund and disappearing within about 10 years.

This fragility and creative destruction is reminiscent of what we see with startups themselves. Indeed, thinking of early-stage investors as startups in their own right may be a helpful framing for policymakers: rather than considering them as monolithic entities whose sole role is to deploy capital, ecosystem leaders should recognize that investors are themselves often on a learning curve and may benefit from assistance.

Like startups, emerging investors must discover a workable model under uncertainty: how to source differentiated opportunities, what pattern-recognition to trust, where they can add value beyond capital, and how to adapt when the market itself changes. In AI, that adaptation may be particularly demanding, because the criteria for backing great companies are shifting in real time.

In a similar way that peer interaction is highly beneficial for founders, we see significant benefits in facilitating the sharing of knowledge and contacts between early-stage investors. Angel networks, for example, can bring otherwise-isolated investors together to make better collective decisions than they would make individually – as well as increasing the size of their deal-sourcing network and the combined firepower of their checks. Catalyzing such networks – or supporting the capacity-building of young networks that already exist – can be a highly-effective intervention for policymakers.

In an AI-Native economy, focusing support only on local startups will not be enough. As the cost of company creation falls and early traction becomes easier to manufacture, the harder task will be nurturing investors who can tell the difference between speed and substance, between revenue and real demand, and between companies that are merely timely and those that may truly endure.