What Top VCs Actually Evaluate in a B2B Startup

Founders pitch features. Disciplined investors evaluate survival. The gap between the two explains most rejected decks: a founder is answering 'why is my product good?' while the investor is quietly answering a harder question: 'what is most likely to kill this company, and has the team already beaten it?' Strip away the jargon and that question has only three parts.

Start with how startups actually die

The cleanest way to understand what investors weigh is to look at what kills the companies they pass on. CB Insights, analysing post-mortems of failed startups, found the causes cluster tightly. The top reason, by a wide margin, was no market need, at 42% of failures. After that came running out of cash, the wrong team, getting outcompeted, and pricing or cost problems. Founders often cite more than one, which is why these add past 100%.

The innovation trinity, lens by lens

Sort those failure modes and they fall into exactly three buckets, the same three that decide whether any idea is worth building. They are not a venture invention. Desirability, Feasibility and Viability is the framework IDEO and Stanford's d.school made standard for innovation: the overlap where a good idea actually lives, often called the innovation trinity. The order matters: each is necessary, and the first to fail is usually the one that should stop the conversation.

Desirability: do customers actually want this? The biggest killer by far. The investor is testing whether there is a real, urgent pain and a specific customer who feels it, not a clever product in search of a problem. 'No market need' and 'got outcompeted' both live here. For B2B, this is also where the access question sits: a wanted product that can't reach its buyer still fails the desirability test in practice.

Feasibility: can this team actually build and sell it? Here the investor weighs execution: does the team have the technical and go-to-market ability to deliver, and is there something that makes it hard for others to copy. 'Wrong team' is the failure mode. Founder-market fit (whether these specific people are unusually suited to this specific problem) is the heaviest single input. And feasibility is always read against stage: a disciplined investor doesn't hold a Concept-stage team to a Scaling-stage bar. The question is whether this team has cleared the one risk that defines the phase it is actually in: proving the idea, shipping a prototype, landing the first paying customers, or making the sales motion repeatable. Sequoia and a16z both frame it this way: judge the team by the milestone in front of them, not the one three rounds out. Mismatching the bar to the stage is how good early bets get passed over.

Viability: does the business sustain itself? The last gate is the economics: do the unit economics work, is there a path where revenue outruns cost, and is the company funded to reach it. 'Ran out of cash' and 'pricing problems' are the symptoms, but the root is almost always unit economics that never closed. This is the lens founders most often paper over with a hockey-stick chart, and the one a numerate investor stress-tests hardest.

The best investors apply the same three, each in their own language. Y Combinator compresses the whole of desirability into a single line: 'make something people want.' Andreessen Horowitz has argued for years that feasibility and distribution, not the product alone, decide who wins. And every numerate fund lives or dies on viability, the unit economics underneath the story. Different vocabularies, the same trinity, which is why an idea that clears all three is rare, and worth a great deal.

Why the lenses are scored separately

The discipline that separates a rigorous screen from a gut call is keeping the three apart. A charismatic founder can make a weak market feel exciting; a beautiful product can distract from broken economics. Scoring desirability, feasibility and viability independently, before letting them influence each other, stops one strong dimension from masking a fatal weakness in another. A deal that is a clear 'yes' on the product but an unexamined 'maybe' on unit economics is not a yes; it is an unfinished evaluation.

What this means for your deck

If an investor is really scoring these three, the most persuasive deck is the one that pre-empts them. Show the urgent pain and the specific customer (desirability), the evidence that this team in particular can win (feasibility), and unit economics you have actually examined rather than projected (viability). The fastest way to lose a numerate investor is to be brilliant on one lens and silent on another: silence on viability reads as a problem you are hoping they won't notice.

It is the screen EvoScale Capital runs on every submission: desirability, feasibility and viability, weighed independently, and paired with operators who have actually run the playbook rather than only funded it. Whoever is across the table, the question they are really asking is the same: of the three ways this company could die, which is closest, and have you already beaten it?