How to raise a Series A in late 2026: what investors told founders at Disrupt, with the numbers behind it

If you want to know how to raise a Series A in late 2026, start with this: three investors on a TechCrunch Disrupt panel this week described a market where top firms set a high bar, cumulative ARR charts read as a red flag, and rounds split into tranches are a sign of froth. Below is what they said, the published benchmarks, and how to prepare.

The panel, reported by TechCrunch on October 22, 2026, featured Shailendra Singh of Peak XV, Jahanvi Sardana of Index Ventures and Janelle Teng Wade of Bessemer Venture Partners. We pair their remarks with a July 2026 benchmark rubric from Burkland, a startup finance firm, and Crunchbase's Q3 2026 funding data. Upfynd is a software directory, not an investor, so nothing here is investment advice: it is what investors said and what the data shows.

Series A money is plentiful, but it is pooling in AI

The headline numbers look healthy. Crunchbase data puts global venture funding at $159 billion in Q3 2026, up 53% from Q3 2025 but down 25% from the $212 billion raised in Q2. Early-stage funding, a category that includes Series A and B rounds, totaled $40.6 billion, up 25% year over year.

The concentration is the catch. In the same report, AI startups raised $102 billion, or 64% of all global venture capital in the quarter. Crunchbase notes that its figures are as of October 2, 2026, and that reporting lags are most pronounced at the earliest stages, so the totals are preliminary.

That split is why founders hear two stories at once. A company in a sector investors are chasing can raise quickly at aggressive terms, while everyone else meets the bar Singh described.

How hard is it to raise a Series A in 2026?

Hard, by the account of at least one panelist. TechCrunch reports that Singh called the current market one of the hardest to raise in that he has seen, and quotes him directly: "Seed and venture, I've never seen it as dysfunctional as it is now." He added that "the burden is high to convince any top firm," and that the terms and amounts founders can get vary widely.

The strongest counter-argument is the data above: total funding is up sharply year over year, and investors are writing large checks. Both can be true. Singh also said the market is full of "false positives," described current optimism as unusual for two decades, and said "I think we will see a correction" in 2027. That is one investor's forecast, not a consensus, but it is a reason to raise on terms you could live with if the cycle turns.

Teng Wade said capital is flowing into defense, robotics and AI infrastructure.

What do Series A investors look for in 2026?

They look for honest numbers before they look for big ones. The single most specific warning from the panel concerned how founders present revenue.

How to show ARR to investors without a red flag

Singh told founders that a chart of cumulative revenue or cumulative ARR (annual recurring revenue) is "a massive red flag," according to TechCrunch's report. He called cumulative data "a big lie" and said founders need good time-series data instead.

The reason is mechanical. A cumulative line only goes up, so it hides churn, a flat quarter or a month where growth stalled. In our reading, the practical fix is a revenue series investors can recompute, period by period, with new, expansion and churned revenue shown separately. Singh did not prescribe a format, so treat that as preparation, not as his rule.

Tools for this are listed in our Accounting & Invoicing category, which held 155 products as of October 2026. ChartMogul, for example, says on its homepage that it turns billing events such as payments, refunds, upgrades and discounts into revenue metrics, with MRR movements and ARR views. Pilot says it handles bookkeeping, tax and advisory work for startups.

Intellectual honesty and the talent signal

Singh also said companies should be "intellectually honest" in how they present themselves during bull cycles, and that founders who attract very high-quality talent have an advantage. He urged founders not to rush, to watch for weak product-market fit and to set a very high bar for the problem they solve. He also described building a company as one of the "most excruciatingly difficult things" a person can do, and told founders to accept that going in.

Sardana's warning ran the other way, at founders who chase the market. "It feels like [founders are] deciding what to build based on what's getting funded, which is really scary," she said, per TechCrunch. Knowing how to pitch is necessary; letting the pitch decide the roadmap is the mistake.

Series A metrics and benchmarks for 2026, per Burkland

The panel gave no numeric bars, so for those we turn to Burkland's July 2026 guide. Burkland sells fractional CFO, accounting and fundraising-readiness services, so read its rubric as one service firm's guidance, not a study. Burkland itself calls these figures directional rather than pass or fail.

The table is Burkland's illustrative rubric for SaaS companies. Net revenue retention is the revenue you keep from last year's customers after expansion, downgrades and churn, so a figure above 100% means the existing base grows on its own. CAC payback is the months of gross profit it takes to earn back the cost of acquiring a customer. It also suggests software gross margins of 70% or more, 5-10 customer references, and flags one customer above 40% of ARR without a mitigation plan as a red flag.

How much ARR do you need for a Series A?

Burkland writes that many SaaS Series A rounds cluster around $1 million to $3 million in ARR. For AI companies in the 2026 cycle it sets a higher bar: roughly $3 million to $5 million in ARR reads as strong, 3x year-over-year growth is closer to the floor than the target, and gross margin after compute costs of 65% or more reads as strong.

Teng Wade made a related point on the panel: metrics still matter in AI, but they are less standardized when compute is expensive. She cited Bessemer's early evaluation of Anthropic, made when its margins were negative and it was spending heavily on compute.

Burn multiple for a Series A

Burn multiple is net burn divided by net new ARR, a formula popularized by David Sacks of Craft Ventures in 2020. Sacks called 2x "reasonable for an early-stage startup" and 5x "a terrible Burn Multiple." Burkland's 2026 rubric puts efficient early-stage SaaS under about 2x, with the example of $300,000 in monthly burn against $150,000 of net new ARR equaling 2.0x.

Skip the burn-multiple pitch if you are pre-revenue or your ARR is too small for the ratio to mean much; one large contract can swing it in either direction.

Tranched rounds and multi-price rounds are froth, not a template

A tranched round releases committed money in chunks, often tied to milestones, rather than all at close. Sardana said, "We're calling it the 'tranchopalooza,'" describing how common the structure has become.

TechCrunch also reports that it is now common to see a company close a round at up to three prices at once. Singh called that "a crazy sign of market froth." A nickname and a froth warning are not endorsements. If Singh's predicted correction does arrive, the money still waiting on a milestone is the money most exposed to it.

Teng Wade did offer one shift in the founder's favor. Her old advice was to be conservative with projections and asks; now, she said, "I would say just be audacious in your ask," then see how the market responds and tailor from there.

Series A dilution, round size and timeline

Burkland cites Crunchbase for a median Series A round of about $20 million in 2026. Its own guidance puts typical Series A dilution at about 15% to 25% of the company, including any option pool expansion.

How long does it take to raise a Series A?

Burkland estimates three to six months from first meetings to money received, and recommends starting preparation six or more months before the raise. It also says runway under six months weakens your leverage, while nine to twelve months or more is stronger.

When should you start raising a Series A?

Start when you can show a clean revenue time series that meets the bar for your sector, and while you still have the runway Burkland describes. Burkland's three-to-six-month estimate plus a six-month lead means an even longer window from first preparation to cash.

How to raise a Series A: the preparation checklist

1. Rebuild your revenue reporting as a time series that shows new, expansion and churned revenue, and retire any cumulative ARR chart.
2. Compute your burn multiple, net revenue retention and CAC payback, and compare them with Burkland's rubric for your sector.
3. Check customer concentration and line up 5-10 references.
4. Model dilution in the 15%-25% range Burkland describes, including option pool expansion.
5. Count runway from the first meeting plus three to six months of process.
6. Decide in advance which tranche milestones or price structures you would refuse.

For the paperwork side, our Legal & Compliance category lists 80 products and our Investing & Fundraising category lists 79, as of October 2026. If you are building one of those tools yourself, listing it on Upfynd is free.