The Solo Unicorn Club ran its pitch day in New York City four days ago. Eight products. Five awards. One question from the audience that stuck with the panelists more than any product demo: "When one person with AI tools can build what used to require a team, what do solo founders actually compete on?"
The answer was fast, and unanimous: taste, trust, and distribution. What nobody said was how you develop the first one of those when you're building alone.
Execution Is Table Stakes Now
The case for AI-native solo building has been made well enough to fill a pitch room. Medvi at $401M revenue, two people. Tools that let one developer spin up infrastructure that would have needed a dedicated DevOps hire in 2022. AI writing code, copy, and customer support scripts. The surface-level story is persuasive: you can build things that used to require funding to staff.
But "you can build it" isn't a moat. If one person can build something in a weekend, so can the next person who gets the same idea. The execution layer has compressed so dramatically that it no longer separates winners from also-rans. What separates them is the decisions about what to build, how to position it, and whether anyone finds out it exists.
Taste. Trust. Distribution.
Taste Is Accumulated Data, Not Instinct
When people talk about taste as a competitive advantage, they usually describe it as something some founders have and others don't — an innate ability to know what's right, what's too much, what will resonate. That's not quite how it works.
Taste is a pattern-matching skill that develops through feedback loops. Founders who have good product taste have typically built enough things, shipped enough features, heard from enough customers, and watched enough outcomes to know — quickly, often subconsciously — which direction is right. They're not guessing. They're extrapolating from accumulated evidence.
The problem for solo founders is that feedback loops are slow and thin. There's no PM doing weekly user interviews. No data analyst tracking cohort behavior. No support team whose tickets surface patterns in what users are confused about. You build something, you ship it, and then you wait — usually for signals that are noisy, delayed, or absent entirely.
And while you're waiting, you're making the next hundred decisions. About what to build next. About which bug to prioritize. About whether the pricing is right. About which channel to put time into this week. Each decision gets made on gut feel, because the feedback loop from the last decision hasn't closed yet.
Over time, if you survive long enough, some pattern-matching develops. But it's slow. The founders at the NYC pitch day who had the sharpest intuitions about their markets weren't just smart — they'd been moving faster through those feedback loops than their peers.
The Information Problem Nobody Talks About
Most solo founders are more operationally blind than they realize.
They know what they shipped. They don't know whether the users who signed up on the day they launched a new feature stuck around or churned faster than baseline. They know they spent "a lot of time" on a feature. They don't know whether it was 4 hours or 18, or whether that time came out of marketing or customer conversations.
This is the feedback loop problem in its most granular form. You can't develop taste about your own product decisions without information about which decisions produced which outcomes. And most solo founders are running on memory, which systematically flatters the decisions that felt good and discards the ones that felt bad.
There's a version of this that looks like strategy. A founder will say they're prioritizing user activation over new features. Then you look at where their actual time went last week, and they were in their code editor for 22 hours and had one customer conversation. The priority was a belief. The behavior was something else.
What Fast Feedback Loops Look Like
The founders who develop taste quickly share something: they're seeing outcomes faster, and they're being honest about what the outcomes are.
That sounds obvious. In practice it requires two habits most solo founders don't have.
The first is tracking what's actually getting your time versus what you think is getting your time. Not at a macro level — at the level of where you were, what you were working on, and when you switched. The gap between stated priorities and actual time allocation is usually the first thing that falls apart when building alone. Nobody's watching. The urgent crowds out the important at a rate you don't notice until you look at the data.
The second is measuring outcomes at the decision level, not the product level. Knowing that your app has 300 users isn't as useful as knowing that users who activated the correlation view in the first week retained at 2x the rate of users who didn't — and that that correlation view took you 14 hours across four days in early May. That's the unit of feedback that builds taste. Was that 14 hours well-spent? You can't answer that without both sides of the equation.
Trust and Distribution Feed on Taste
The other two things the panelists named — trust and distribution — don't exist in isolation from taste.
Trust is what users and customers extend to a solo founder who keeps making decisions that seem right. The product improves in ways users didn't ask for but immediately value. The pricing changes make sense. The roadmap feels coherent. Trust accumulates when the decisions are good, and it's nearly impossible to fake over time. It's taste made visible to your users.
Distribution is the one thing AI genuinely hasn't solved. Reaching people is still human work — relationships, channels, content, presence. But knowing which distribution channels are worth the time requires the same feedback loop discipline. If you spent 12 hours writing content last month and it drove two signups, that's a data point. If you spent 3 hours in niche community conversations and it drove twelve, that's a different data point. Most solo founders don't track distribution time at this resolution, so they keep running a strategy shaped by intuition rather than evidence.
The Meta-Product Problem
We built xeve partly as our answer to this. Not as a productivity tool in the motivational sense — not streaks, not gamified commit counts, not "you coded for 4 hours today, nice job." As a measurement system.
The question we're trying to answer for ourselves is: where does the work actually go, what is the relationship between where the work goes and what outcomes follow, and are we getting better at that mapping over time?
That's the taste-building flywheel. You need the data before you can see the pattern. You need to see the pattern enough times before you can extrapolate. And you need to extrapolate accurately enough, often enough, to have what the panelists in New York were calling taste.
The solo unicorn pitch day was a proxy for something real: there is now a population of builders competing on taste, trust, and distribution rather than on whether they can staff an engineering team. The building part is solved. The feedback loop part is mostly not.
That's the gap we're working on.