Burn multiple, net burn divided by net new ARR, has replaced burn rate as the number investors actually price into your next round, and most founders are still calculating it wrong.
A startup burning $500,000 a month and a startup burning $500,000 a month look identical on a cash flow statement. They are not identical companies. One of them might be adding $400,000 in new annual recurring revenue for that spend every month. The other might be adding $80,000. That gap is what burn multiple startup investors now use to tell real capital efficiency apart from a big scary number on a bank statement. David Sacks, the Craft Ventures general partner who popularized the metric in a 2020 Medium post, built it because burn rate alone tells you nothing about what the money actually bought. Burn multiple tells you exactly that, and it has become the first number a lot of VCs pull up on a board deck before they look at anything else.
The formula is simple enough that most founders assume they already know it: net burn divided by net new ARR, measured over the same period, usually a quarter. Burn $3 million in a quarter and add $2 million in net new ARR, and your burn multiple is 1.5. You spent a dollar fifty for every dollar of new recurring revenue you booked.
Here is where it goes wrong. Founders plug in total revenue growth instead of net new ARR, which pulls in one-time services fees and non-recurring contracts that inflate the denominator. Others use gross burn instead of net burn, or measure burn from one quarter against ARR growth from a different one because their finance calendar and their board deck calendar don't line up. Any one of those mistakes can swing the number by half a point, and a half point is the difference between a clean term sheet and a hard conversation about runway.
There's a second, quieter mistake: netting churn out of the wrong side of the equation. Net new ARR already accounts for churn and downgrades, since it's this quarter's ending ARR minus last quarter's. Some founders then subtract churn again from the numerator when they calculate burn, double counting the same lost revenue and flattering their multiple in the process. Run the numbers straight, once, and the metric does its job. Run them twice and you're just negotiating with yourself.
Gross burn is every dollar that leaves the business: payroll, rent, software, marketing, everything. Net burn subtracts whatever revenue came in during that same period. Sacks's formula wants net burn, because gross burn on its own punishes companies that are already collecting real revenue while they scale.
Say you spent $4 million in a quarter and collected $1.5 million in revenue. Your gross burn is $4 million, but your net burn, the number that belongs in the formula, is $2.5 million. Skip that subtraction and you'll overstate how inefficient you actually are, sometimes by a wide margin. The distinction matters more the later stage a company gets. A Series B company generating meaningful revenue looks dramatically worse on gross burn than it does on the net figure investors are actually underwriting.
Bessemer Venture Partners tracks a related idea in its Cloud Index through the Rule of 40, which blends growth rate and profit margin into a single score. Burn multiple does a narrower job. It doesn't require profitability data at all, which is exactly why it caught on with earlier-stage companies that Bessemer's own metric was never built for.
What a good burn multiple looks like at each stage
Sacks's original scale, and the version most VCs still cite, treats anything under 1 as excellent capital efficiency, 1 to 1.5 as good, 1.5 to 2 as suspect, 2 to 3 as concerning, and anything above 3 as a problem regardless of how fast the top line is growing.
But the bar moves with stage, and grading a seed-stage burn multiple on a Series B curve misreads the number entirely. Pre-seed and seed companies with under $1 million in ARR routinely run multiples of 2 to 3, because they're still finding product-market fit and every dollar of ARR is hard-won. Companies in the $1 million to $3 million ARR range, chasing repeatable sales motion rather than proven product-market fit, should be tightening toward 1.3 to 1.6. By Series A, with $3 million to $8 million in ARR, 1 to 1.5 is typical, and something close to 1.2 alongside clean net revenue retention is the real target. Series B companies with $8 million to $15 million in ARR are expected to run closer to 0.8 to 1.2, because by then the growth motion is supposed to be proven, not experimental.
A seed startup running a 2.5 isn't in trouble. A Series B company running a 2.5 is burning investor patience along with cash.
Klaviyo is the case worth studying. The email marketing platform reported roughly $585 million in revenue for 2022 in its S-1 filing, coverage TechCrunch detailed ahead of the company's 2023 IPO, after raising a comparatively modest amount of venture funding relative to companies of similar scale. That ratio, meaningful revenue built without proportional cash burn, is exactly what a low burn multiple is supposed to signal to the market, and it's a large part of why the IPO drew as much attention as it did.
Contrast that with WeWork, whose 2019 IPO collapsed in part because journalists and investors dug into a business burning enormous sums to add revenue that didn't come close to justifying the spend, reporting The Wall Street Journal covered extensively at the time. WeWork never used the term burn multiple. Its numbers are still the textbook illustration of what a bad one looks like: capital going out faster than any rational multiple of ARR was coming in.
The metric faded from headlines for a stretch after Sacks first wrote it up, then came roaring back through 2022 and 2023 as interest rates rose and venture capital got tighter, a shift TechCrunch and other outlets tracked as investors stopped rewarding growth at any cost. Founders who could point to a sub-1 burn multiple suddenly had real leverage in rooms where, eighteen months earlier, growth rate alone would have carried the meeting.
Frankly, the reason burn multiple caught on this fast is that it's harder to game than burn rate. A founder can slow burn rate by freezing hiring for a quarter and call it discipline. Burn multiple punishes that move immediately if the freeze also stalls net new ARR, because both sides of the ratio move together. It forces you to look at spend and growth at the same time, which is the whole point of the metric.
Calculate it every quarter, not just in the weeks before you fundraise. Track it against the benchmark for your own stage, not some average pulled from a VC's blog post. And when the number moves in the wrong direction, don't reach for the cost-cutting lever first. Figure out whether net new ARR stalled before spend did, because that's usually where the real problem started.
One more thing worth saying plainly: burn multiple is a diagnostic, not a strategy. It tells you how efficiently you turned cash into recurring revenue last quarter. It doesn't tell you whether you're building the right product, whether your market is big enough, or whether the ARR you added will still be there next year. A company can post a beautiful 0.9 while quietly signing customers who churn within six months, and the multiple won't catch that until the churn shows up in a later quarter's net new ARR. Use it alongside net revenue retention and CAC payback, not instead of them, and it will tell you something true. Use it alone, and you're just watching one dial on a dashboard that has several.
Also read: How to Read a Company's Financial Statements Before Investing in SaaS • How to Split Equity Between Co-Founders Without Blowing Up the Company • How to Build a Startup Financial Model Investors Will Actually Believe
