When should a startup raise a seed round?

Raise a seed round when you have enough of a working experiment to justify funding the next phase of it—not before, and not so late that cash pressure distorts your decisions. The seed stage is fundamentally about buying time to prove a hypothesis, so the question isn't 'can we get investors interested?' but 'is there something real enough here that outside capital will accelerate rather than paper over?' Get that answer right and the timing question mostly answers itself.

The core readiness test: do you have something worth proving?

Seed funding exists to turn a promising hypothesis into a demonstrated result—not to fund the search for a hypothesis in the first place. Before approaching investors, ask whether you can clearly articulate what the experiment is and what evidence would prove it worked. That doesn't necessarily mean revenue. It might be retention numbers, a working prototype with real users, or a measurable signal that people want what you're building. But it has to be something specific and falsifiable.

Founders often confuse 'we need money to start' with 'we are ready to raise.' These are different things. You can start most software companies for almost nothing and generate enough signal to make a coherent fundraising case within a few months. Raising before you have that signal doesn't just hurt your terms—it forces you into investor conversations where you have nothing to show, which trains you to pitch abstractions instead of progress. That's a bad habit that compounds.

The bar investors actually use at the seed stage is lower than founders tend to imagine in one dimension and higher in another. They don't need a profitable business. They do need founders who seem like the right people to run this specific experiment, with enough of a foundation that the outcome isn't purely a coin flip. If you've built nothing and talked to no customers, you're pitching yourself—and very few founders are compelling enough to win on that alone.

Raise before you're desperate, not before you're ready

There's a window between 'too early' and 'too late' that most founders don't think about carefully enough. Too early means you're raising on pure idea with no traction, which means poor terms, slow closes, and the psychological damage of a lot of nos before you've built anything. Too late means you're fundraising with a short runway, which gives investors leverage they don't deserve and forces you to say yes to deals you shouldn't.

The practical implication: begin building relationships with seed investors a couple months before you plan to formally raise. Not pitching—learning. Find out what they care about, what comparable companies they've funded, how they make decisions. When you do go into a proper raise, you want to be running a fast, parallel process rather than a slow sequential one. Paul Graham's writing on fundraising makes the point that talking to investors one at a time is deeply inefficient; momentum and competing interest are your best leverage, and both require that you start from a position of mild confidence rather than acute need.

Calendar timing matters more than founders admit. Investor attention clusters around certain periods—after Labor Day through Thanksgiving, and January through April are historically active windows. Raising in late November or late July doesn't disqualify you, but it slows everything down because partners are traveling or distracted. If your readiness and the calendar align, that's not a coincidence to manufacture—but it is a factor worth planning around.

How the seed round connects to your Series A—don't optimize in isolation

One of the most consequential mistakes seed-stage founders make is treating the seed round as a standalone event rather than the first move in a sequence. Graham's analysis of startup fundraising mechanics highlights a real constraint: if you sell too much of your company in the early rounds, you make a Series A structurally harder—not because investors don't like you, but because the cap table math stops working. Selling more than roughly 15% in a pre-seed or early seed context, then another 20-25% in your seed, starts to crowd out the equity pool that later-stage investors and future employees require.

This means your seed round sizing should be driven by what you need to reach a Series A-worthy milestone, not by how much you can raise or what a given investor wants to put in. The milestone you're aiming for at seed exit is usually clear: a growth trajectory and unit economics that suggest you're on the path to a real business, typically evidenced by revenue or rigorous proof of demand. If you raise too little, you run out of runway before you can demonstrate that. If you raise too much, you may spend it and still not hit the milestone—and you've diluted yourself unnecessarily in the process.

A disciplined way to think about seed sizing: work backward from the Series A. What does a Series A investor need to see? How long will it take to produce that evidence? What does it cost to operate for that period with a small buffer? That number is your seed target. It's almost always smaller than what you could raise if you tried, and almost always larger than what feels comfortable if you're trying to minimize dilution.

The operational discipline to maintain once you've raised

Raising a seed round creates a new set of failure modes that are easy to ignore in the excitement of having capital. The most common: treating the raise as the accomplishment rather than the starting gun. Once money is in the bank, the clock is running toward your next fundraise—and that raise will be evaluated on what the experiment produced, not on how compelling the original pitch was.

Graham's observation about companies that raise seed money and then drift for a year without urgency to become profitable (or reach the relevant milestone) is one of the most underappreciated risks in the startup lifecycle. The danger isn't that founders are lazy—it's that without a specific deadline forcing action, the urgency to achieve a milestone gets displaced by the busy work of running a company. Hiring, tooling, process, culture—all of it is real work, and none of it is the experiment you raised money to run.

Set a concrete milestone and a concrete deadline at the moment you close your seed round, before the money touches your bank account. What does success look like in 12 months? What does the trajectory need to look like at month 6 for you to believe you're on track? Build your budget around those constraints, not around what feels reasonable given what you raised. A seed round should feel slightly tight—if it feels comfortable, you're probably planning to spend money on things that won't move the needle.

“The more you raise, the more you spend, and spending a lot of money can be disastrous for an early stage startup.”

— Paul Graham, source

The one thing to do

Before approaching any investor, write down exactly what milestone the seed capital will fund and what evidence will prove it worked—if you can't answer that in one sentence, you're not ready to raise.

Frequently asked questions

Do you need revenue before raising a seed round?

Not necessarily, but you need meaningful evidence that the problem is real and your approach is plausible. That could be revenue, or it could be strong retention, clear user demand, or a technical proof of concept. Pure idea-stage raises are possible but rare and usually result in poor terms.

How much should a startup raise in a seed round?

Raise what you need to reach a Series A-worthy milestone—typically 12-18 months of lean operation. Work backward from the milestone, not forward from what investors will offer. Raising more than you need creates spending pressure and dilution without improving your odds.

Should you raise a seed round before building a product?

Only if you have an unusually strong track record or a technical challenge so credibly hard that investors believe only you can solve it. For most founders, spending a few months building a basic version first gives you far more leverage in the raise and better terms.

What happens if you raise your seed round too early?

You'll likely face worse terms, slower closes, and a higher risk of running out of runway before hitting a Series A milestone. More subtly, raising before you have signal can lock you into a narrative that turns out to be wrong—and pivoting is harder with investors already on your cap table.

Sources

More playbook answers · Growth Prophet home