Investors Say Series A Has No Formula. Here Is What To Do Instead.

Three investors walked onto a stage at TechCrunch Disrupt this week to explain how to raise a Series A. Then they spent most of their time explaining why there isn't one way to do it.

That's useful, weirdly. If you've been hunting for the magic ARR number that unlocks a round, the people who write the checks just told you it doesn't exist. What they said instead is worth reading closely, because it describes a market that's loud, uneven, and easy to misread when you're raising for the first time.

Shailendra Singh of Peak XV, Jahanvi Sardana of Index Ventures and Jenelle Teng Wade of Bessemer all said there's no fixed definition of a Series A. TechCrunch's Dominic-Madori Davis reported that no single metric decides it, and no single playbook gets you there.

Singh went further. He said he has never seen seed and venture as "dysfunctional" as they are now, and that raising is among the hardest he has seen. Founders carry a high burden to convince top firms they can break out, while terms vary wildly from one company to the next.

So the honest summary is this: the bar moved, and nobody posted the new height.

Singh pointed out that rounds with up to three prices at once are now common. He called that a "crazy sign of market froth." Sardana had her own label for the way rounds get split into pieces: "tranchopalooza."

Here's what that looks like in practice. One headline valuation gets announced, but the money arrives in tranches, and each tranche can carry different terms depending on milestones. The number in the press release and the number in your bank account can tell two different stories.

If you're a first-time founder, read the term sheet like a contract for a used car. Ask what's included, what triggers the next payment, and what happens if you miss a target. A big headline figure with fine print attached isn't the same as a big round.

Singh had a blunt warning about one chart in particular. Cumulative revenue or ARR charts, he said, are a "massive red flag." The market is full of "false positives," meaning revenue that looks real on a slide and then evaporates when customers churn or pilots end.

A line that only goes up because you never subtract anything is easy to draw. It's also easy for an experienced investor to spot.

Show instead:

• Monthly revenue, not a running total
• Net retention by cohort, so people can see who stayed
• How many customers came from the same kind of motion
• Gross margin, with compute costs included if you sell AI

Notice what's missing from that list. Nothing says "hit $X and call us." The investors didn't give one, and the TechCrunch report doesn't contain a single ARR threshold, so anyone selling you a magic figure is guessing.

Somewhat, and the reason is cost. Teng Wade said metrics matter less rigidly in AI because these companies are expensive to run. Bessemer looked at Anthropic in its early days, when margins were negative and compute spending was heavy. The bet was that if a company became a top-three player, it could pay off even with money-losing margins.

That logic has limits. It works for a handful of companies at the very top of a category. For everybody else, negative margins are just negative margins.

The scale of the money in the market tells you why this thinking exists. TechCrunch reported on Thursday that OpenAI is working on a raise of up to $30 billion at a $1.25 trillion valuation, which would make it the most valuable private company ever. Town, a two-year-old startup building AI assistants for businesses, raised a $108 million Series B at a unicorn valuation on October 19, after a $55 million Series A in June. Those stories shape what investors compare you against, even when the comparison is unfair.

Teng Wade said capital is going into defense, robotics and AI infrastructure. She also said foundation model companies are moving toward vertical integration, though neo-clouds (companies renting AI compute) and neo-labs (AI research outfits) still have room.

If you're building in one of those areas, you've got tailwind. If you're not, don't panic. A hot sector gets more attention, but it also gets more competition, and investors in a crowded category will demand a clearer reason you'll win.

Singh added that founders who can attract very high-quality talent have an advantage. That's a quiet point that deserves attention. Great early hires are a signal investors can verify by checking a few LinkedIn profiles.

You plan the things you control. Whatever the market does, you can still write down your assumptions, your customer evidence, your costs, and what each dollar of the round will buy.

A useful exercise: build three versions of your next eighteen months. One assumes you raise what you hoped. One assumes you raise half. One assumes you raise nothing and have to get to profit or default alive on your own. If all three versions hold together, you can negotiate without fear.

This is where a structured planning workspace like Foundra earns its place, because keeping your financial model, competitive notes and go-to-market assumptions in one place makes it much easier to see which numbers change when the round does. A spreadsheet works too. The point is to have the three scenarios written down before you take a single meeting.

Here Teng Wade changed her own advice. She used to tell founders to be conservative in their projections and asks. She said that no longer holds. Now she advises being "audacious in your ask" and then tailoring based on how the market responds.

That's a real shift, and it fits a market where prices vary so much. If you ask for less than the market would give, you can't easily go back up. If you ask for more and get pushback, you can adjust.

But audacious doesn't mean fantasy. It means asking for the high end of what your evidence supports, not inventing evidence to support the ask. If a partner pokes at your numbers, you want to be able to answer every question without flinching.

Sardana said it's "really scary" when founders choose products based on what's getting funded. She told founders to learn to pitch but not let fundraising dictate what they build.

Think about how that plays out. A founder sees defense and robotics soaking up money, pivots a software idea toward those fields, and raises a seed. Eighteen months later they have no customers who care and no team that understands the problem.

Raise money because your product works and you need fuel. Don't build a product because it's the type of thing that raises money. Singh made a similar point from the other direction: set a very high bar for the problem you're solving, and watch for weak product-market fit before you take on the pressure of a big round.

That pressure comes with a clock. As Tomasz Tunguz said at a separate Disrupt session, the expectations for execution are "absolutely tremendous" for startups at every stage.

Singh said this is the most optimistic he has seen venture in two decades, which in his experience is usually a sign of trouble. He expects a correction in 2027.

You can't time that, and neither can he. But you can prepare for it. Keep your burn lower than the round would allow. Hold off on hires you could delay by a quarter. Build relationships with investors before you need them, because the ones who've followed your progress for a year will answer your email when the market gets cold.

If a correction doesn't arrive, you've simply spent less and built a stronger company. That's a decent outcome for being cautious.

There's no Series A formula, so stop looking for one. Read term sheets for tranches and multiple prices. Skip cumulative ARR charts and show monthly numbers and cohorts. Write down three funding scenarios. Ask boldly but stay inside your evidence. Don't pick a product because it's fundable.

And as Singh said, don't rush. Building a company is among the most excruciatingly difficult things you can do, and a round doesn't make it easier.

What revenue do you need for a Series A in 2026? The investors at Disrupt said there's no fixed definition or single metric. TechCrunch's report includes no ARR threshold, so treat any specific number you hear as an opinion.

Why did Singh call cumulative ARR charts a red flag? He said the market is full of false positives. A running total hides churn and one-time revenue, so he'd rather see monthly figures and customer retention.

What is a "tranchopalooza"? It's Sardana's term for splitting a round into tranches. The headline valuation may look large, while the cash arrives in pieces with conditions attached.

Should first-time founders ask for more money than they need? Teng Wade now suggests being "audacious in your ask" and tailoring it based on market response. Stay within what your evidence supports.

Is a venture correction coming? Singh expects one in 2027, based on how optimistic the market has become. It's a forecast, not a fact, but it's a fair reason to keep burn modest.

How many scenarios should I model before fundraising? At least three: raise what you hoped, raise half, and raise nothing. If each one holds together, you're negotiating from a stronger position.

1. Investors break down how founders can raise a Series A today, TechCrunch, October 22, 2026
2. OpenAI working on new $30B raise at $1.25T valuation ahead of IPO, TechCrunch, October 22, 2026
3. AI assistant startup Town nabs unicorn valuation four months after launch, TechCrunch, October 19, 2026
4. The secret to fast startup growth isn't early sales, top names in tech say, TechCrunch, October 22, 2026