How much burn per dollar of growth?
№ 223 · Burn multiple
Definition
MetricBurn multiple, net burn divided by net new ARR
Unitx, dollars of net burn per dollar of net new ARR
Net burn divided by net new ARR added in the same period, shown as a bar or line. Below 1x is efficient; above 2x signals over-investment. Pull cash burn and net new ARR from finance each quarter. Example: burned $1.4M in Q3, added $700K ARR = 2.0x burn multiple.
- ARR
- Annual recurring revenue. MRR at period end multiplied by 12, or the annualised value of active contracts.
Benchmarks
| Bottom 30% | Median | Top 30% |
|---|---|---|
| above 2.5x | 1.4x to 1.6x | under 1.0x |
private B2B SaaS, $1M to $100M ARR, FY2024 and FY2025, Craft Ventures definition of net burn divided by net new ARR measured over the same period · Definitions diverge enough to change the number by a factor of three, so the denominator has to be stated before any comparison. Benchmarkit computes cash burned over net new ARR and finds the metric declining with scale, with under 1.0x the working target by the $25M to $50M band. SVB measures dollars burned per dollar of new revenue at Series A and reports 5.0x for AI companies against 3.6x for non-AI, far higher because the stage is earlier and the denominator is revenue rather than ARR. Bessemer inverts the same ratio into an efficiency score, so a Bessemer figure above 1 means the opposite of a burn multiple above 1. Seed-stage numbers are not comparable to growth-stage numbers under any definition, and a quarterly multiple on a lumpy enterprise book is close to meaningless.
| Category | Bottom 30% | Median | Top 30% |
|---|---|---|---|
| Series A, all sectors, revenue denominator | 5.0x and above, the median for AI-native companies at this stage | 3.6x for non-AI companies | under 2.5x |
| $25M to $50M ARR private B2B SaaS, net new ARR denominator | above 2.0x | about 1.4x | under 1.0x |
| Above $100M ARR private B2B SaaS | above 1.5x | at or below 1.0x | negative, cash generative |
When it looks bad
The bars sit flat or rise as ARR grows, so the multiple on the right of the chart is as tall as the one on the left, which means scale is producing no efficiency and the next round of spend will not either.
Burn of $2.4M in a quarter against $800k of net new ARR gives 3.0x at $30M ARR, a level that is defensible at seed and is a diligence problem at that scale.
What to do about it
- Recompute the multiple with expansion ARR separated from new logo ARR. Benchmarkit puts the expansion CAC ratio at about $1.00 against $2.00 for new customer ARR, so a company with a poor blended multiple often has a working expansion engine and a broken new logo engine, and the fix is allocation rather than a cost cut.
- Cap the multiple in the operating plan instead of capping burn. Release incremental spend above plan only when the trailing two quarters print under target. This works because burn multiple is now the number the diligence uses, with 83 percent of Series C and later investors treating it as critical, so the plan and the raise run off the same metric.
- Fix gross margin before touching opex when the problem is the denominator. Benchmarkit's median blended gross margin of 77 percent against 81 percent on subscription revenue shows services mix above 15 to 20 percent of revenue as the usual cause, and repricing or productising implementation improves net new ARR quality without a headcount decision.
- Publish the multiple on a trailing twelve-month basis alongside the quarterly one. A single quarter with a large renewal or a churned enterprise logo can swing the quarterly figure by more than 1.0x, which invites the wrong conversation at exactly the wrong moment.
Sources
- Benchmarkit Defines the burn multiple as cash burned in a period divided by net new ARR, reports that it falls as companies scale, and sets the working goal at below 1.0x by the $25M to $50M ARR band, turning negative above that when a company generates more new ARR than cash burned. benchmarkit.ai ↗
- Silicon Valley Bank Reports that the median Series A AI company burns $5 to gain $1 of new revenue, against $3.60 for non-AI companies at the same stage. svb.com ↗
- SaaStr (reporting SVB) Reports Series A burn multiples of 5.0x for AI companies against 3.6x for non-AI, and notes half as many startups graduating to the next round within three years compared with 2020. saastr.com ↗
- SaaS Mag Synthesises stage-level burn multiple benchmarks with Series A medians near 1.2x, growth stage at $25M to $50M ARR near 1.4x with top performers below 1.0x, and companies above $100M ARR at or below 1.0x, and reports that 56 percent of seed investors and 83 percent of Series C and later investors now treat burn multiple as a critical evaluation metric. saasmag.com ↗
- High Alpha Finds AI-native companies growing roughly three times faster than traditional B2B SaaS peers at every ARR band with about 5 points lower gross margin, producing better Rule of 40 scores driven by growth rather than margin. highalpha.com ↗
- SaaS Capital Reports equity-backed private SaaS companies spending 101 percent of ARR across all departments against 96 percent for bootstrapped peers, which is the spending side of the same ratio. saas-capital.com ↗
- Lighter Capital Reports the median sales and marketing multiple at about 3x for 2025, meaning $150k of S&M spend generated about $450k of annual revenue, down from about 6x in 2024. lightercapital.com ↗
- KeyBanc Capital Markets and Sapphire Ventures Reports that private SaaS EBITDA margins have improved every year since 2022 and are expected to cross into positive territory in 2026, which is the point at which burn multiples turn negative for the median company. prnewswire.com ↗