What investors already know about your exit prospects - and you probably don't

Most founders, after closing their second institutional round, allow themselves a quiet moment of relief. Two rounds of validation. Two rounds of investor due diligence. Two rounds of someone with significant money on the line saying yes, this company is going somewhere. The instinct that follows is natural: we're on the path now. The hard part is behind us - the exit will come.

The latest data says otherwise. PitchBook's Q1 2026 European Venture Report reveals something most founders would find genuinely uncomfortable to read. A drill down on all European AI companies that have already raised at least two rounds of institutional finance, says 42% are predicted to have no exit at all.

Not failure in the dramatic sense. Just companies that drift sideways, become self-sustaining, get stuck, or quietly run out of momentum without ever reaching a clean ending. After two rounds. After all that effort.

And it’s not just in AI. Similar numbers repeat across other sectors. They should change how every founder thinks from Series A onwards.

This data reveals something else, too: that genuinely useful predictive analytics now exist for venture-backed businesses, but founders rarely see the predictions being made about their own companies.


The rise of predictive analytics in venture markets

Modern venture market platforms now combine vast historical datasets on private companies, covering financings, investors, exits, hiring, patents, news flow, and more, with machine-learning models that have matured through more than a decade of refinement. The output is no longer simple data retrieval. It is forward-looking probability: how likely is this company to be acquired, to reach IPO, to fail to exit, to raise its next round in the coming months, to outperform its peers?

Each of the leading institutional research platforms now offers some version of this capability. PitchBook has its VC Exit Predictor, which produces probabilities of IPO, M&A, and no-exit outcomes, alongside an Opportunity Score that ranks companies by their likely relative return on investment. CB Insights has Mosaic, with sub-scores covering momentum, market, money, and management, together with an Exit Probability framework. Crunchbase markets a suite of funding, growth, and exit predictions. Dealroom publishes Signal, focused on identifying companies most likely to be lining up for their next round. Major VC firms have built proprietary versions in-house.

Predictive analytics is no longer a novelty in the venture market. It is standard infrastructure, and it is shaping how investors evaluate every deal that lands on their desk.


Why the AI vertical has raised everyone's awareness

The latest European Venture Report covers all verticals within European venture, not just AI. But AI now accounts for 61.3% of European VC deal value, making it the largest single concentration of capital in the region and the cleanest sector through which to surface the power of predictive analytics. Here the patterns are most visible, and the implications most striking.

Of 4,769 European AI companies currently scoreable by PitchBook's VC Exit Predictor, 2,020 are predicted to never exit. 2,677 are predicted to be acquired. Just 72 are predicted to reach IPO.

The data says nearly half of these founders are not on the path they think they're on. But read what follows as a lesson in how venture exits work in 2026, not just a lesson about AI specifically.


M&A so dramatically dominates IPO that it must shape strategy

Take the IPO and M&A figures alone. Seventy-two predicted IPOs against 2,677 predicted acquisitions is a ratio of one to thirty-seven. This is not a marginal preference for M&A. It is the overwhelmingly dominant exit shape, and the magnitude reframes the question every founder should be asking. The right question is no longer will I be acquired? but who would acquire me, why would they want to, and when do I start building the relationships that make it possible?

That last part is widely misunderstood. Building acquirer relationships does not mean opening exit conversations from year three. It means starting genuine business relationships with companies that, several years from now, may turn out to be the most likely acquirers. Supplier agreements. Joint development work. Technical pilots. Integration partnerships.

Those relationships rarely begin as exit conversations. They begin as commercial conversations, and over time, if both sides find value, the strategic logic emerges naturally. The founders who think this way from the second institutional round onwards build optionality. The founders who don't, find themselves negotiating exits cold, with no relationship history and no leverage.


Three decisions that look different once you accept the M&A path

Once a founder genuinely accepts that M&A is the most likely exit, three decisions take on different weight. The first is board composition. A non-executive director with deep industry experience, ideally someone who has worked inside or alongside the likely acquirer category, contributes something a typical early-stage VC partner cannot. The first of these non-exec seats, often introduced post Series A when early scaling has been established, is a golden opportunity to introduce industry gravitas, acquirer-side perspective, and accelerate relationships that take years to build organically.

The second is partnership choices. Pilots, joint developments, and technical integrations with potential acquirers do double duty: they generate revenue or technical validation today, and they quietly build the familiarity that makes future acquisition conversations easier. The companies that get acquired most cleanly are usually the ones the buyer has already worked with for years.

The third is capital structure. Cap tables that include strategic CVC investors from competing acquirers, complex preference stacks, or unusual exit-related terms (drag-along thresholds, special voting rights, tag-along complications) all reduce optionality at exit. None of these decisions are easily reversed once made. They must be designed in (or out) deliberately from the early institutional rounds, when realistic exit shapes start to come into view.


When founders should start thinking about exit

Conventional founder advice says don't think about exit too early, and at pre-Seed and Seed, that advice still broadly holds. The only question that matters at the earliest stages is whether the product solves a real problem for a real customer. But the conventional wisdom has a subtler corollary that is rarely articulated: after the first institutional round, exit thinking should start to develop quietly in the back of the founder's mind. Not as strategy. As a watching brief. Who are the categories of acquirer in this space? What kinds of companies do they tend to buy? What kinds fail to find buyers?

By the second institutional round, and certainly by Series A, founders should have a working point of view on potential exit shapes. This also happens to be precisely when the predictive analytics platforms can begin to score a company, because two qualifying institutional rounds is often the minimum input these models need to generate a meaningful prediction. And that is precisely when sophisticated investors begin expecting founders to have a thoughtful, if early-stage, view of where the company is heading.


Predictive analytics reveal factors, not formulas

But a word of caution about predictive tools. Founders using them for the first time often misread their role. The instinct is to look for a formula: if I do A, then B, then C, my exit probability rises. That is not how the models work. They don't reward specific actions. They read patterns across broad areas of company performance that have correlated historically with successful exits.

In the PitchBook example, they are grouped under three headings: company profile, financing activity, and active investors. Understanding this matters because it explains why the formula-chasing instinct fails. The model is reading the shape of the company across these dimensions, not scoring individual inputs that founders can optimise in isolation. Trying to game a single metric within any one area achieves nothing if the broader pattern doesn't hold together.

The right way to use these tools is in a diagnostic, not prescriptive, fashion. A predicted no-exit outcome does not mean a company is doomed. It means the historical patterns associated with successful exits aren't strongly present today, and the value lies in understanding which of the three areas a company is weakest in and treating that weakness as a strategic question worth confronting.

The wrong way to use them is to fixate on individual scoring inputs and try to game them. That approach addresses symptoms rather than causes and tends to leave founders worse off than if they had ignored the tools entirely. Used well, predictive analytics give founders a structured way to see their company as investors see it, and to ask harder questions about the strategic decisions in front of them.


The takeaway

Two rounds of institutional capital is not the milestone of safety most founders assume it to be. The historical patterns associated with successful exits produce three outcomes for companies that have crossed that threshold: a small number reach IPO, a much larger number get acquired, and nearly half don't exit at all. The discipline this calls for depends on which of those three outcomes the data is pointing towards in your case.

If your prediction leans towards M&A, there is a well-trodden path to follow. Build a working point of view on the realistic exit shape. Identify the categories of acquirer who would plausibly buy the company, and start the kinds of business relationships that, over years, can mature into something more. Choose that non-executive board seat with that perspective in mind.

If your prediction leans towards no-exit, the work is fundamentally different and considerably harder. It means the historical patterns associated with successful exits are not strongly present in your company today, and no amount of relationship-building with prospective acquirers will change that. The harder question is why. Which of the three factor areas is signalling weakness, and what does the fix require?

That conversation is the one most founders aren't having, and it is the conversation the data says they should be having as soon as the second round of institutional capital lands.

 
Let's talk.