How much cash are we burning for each new dollar of annual recurring revenue?
№ 013 · Burn multiple
Definition
MetricNet cash burned per dollar of net new ARR
Unitx (net burn / net new ARR), trailing quarter or year
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 2.5-3.0x | 1.5-2.0x for venture-backed growth-stage; medians peaked in 2022 and have improved since but have not returned to historical norms | Under 1.0x |
Venture-backed SaaS, quarterly or trailing-12-month measurement, 2022-2025; bands per the Sacks framework as commonly applied by growth investors, trend data from ICONIQ portfolio reporting · Unlike NRR or payback there is no large-sample published quartile table; the 1x / 2x bands are an investor convention (Craft Ventures framework) corroborated by ICONIQ trend commentary rather than survey percentiles. Bootstrapped and profitable companies have burn multiples at or below zero and should ignore this card. Stage matters: pre-$10M ARR companies routinely run above 2x while proving the motion.
Category split omitted: No two independent Tier 1-2 sources publish burn-multiple distributions by stage or segment; only framework bands and portfolio trend commentary exist.
When it looks bad
The trailing line rises while ARR growth is flat or slowing: burning more to add the same or less, the signature of a company pushing product into a market that is not pulling.
Four quarters ago: $3M burned, $3M net new ARR (1.0x). Latest quarter: $4.5M burned, $1.5M net new ARR (3.0x). Same team, same product, tripled cost of growth.
What to do about it
- Decompose the numerator before cutting anything: burn = S&M inefficiency + R&D bets + G&A drag; if the magic number (card 015) is healthy but burn multiple is bad, the problem is outside go-to-market and S&M cuts will make it worse.
- Fix the denominator through retention: net new ARR nets out churn, so a 3-point GRR improvement flows straight into the burn multiple without touching spend; dunning and renewal-desk work (cards 005, 006) are the fastest levers.
- Use the quarterly burn multiple as the gate for hiring plans: no net headcount adds in any quarter where the multiple exceeded plan, a mechanical rule that prevents the classic burn ratchet.
- Benchmark against your own trajectory, not just the 1x bar: ICONIQ treats a declining burn multiple as the signal of strengthening pull; a company going 3.0x to 2.2x to 1.6x is a better story than one flat at 1.4x.
Sources
- ICONIQ Growth Burn multiples peaked in 2022, gradually improving, not yet back to historical norms; used as a proxy for product-market-fit strength iconiqcapital.com ↗
- ICONIQ Growth Burn multiples among the efficiency metrics showing tangible stabilization in 2025 iconiq.com ↗
- Bessemer Venture Partners Adjacent capital-efficiency framework: CCS 0.25-0.5x good (~40% IRR), 0.5-1x better (~80% IRR), 1x+ best (~120% IRR); public cloud median 1.3x at ~$100M revenue bvp.com ↗
- Klipfolio Burn multiple target under 1x, 1-2x generally acceptable, above that a warning sign; Bessemer efficiency score best-in-class above 1.5x klipfolio.com ↗
- SaaS Capital About half of equity-backed companies operate at a loss vs 15% of bootstrapped; equity-backed spend 107%+ of ARR, the funding side of the burn multiple saas-capital.com ↗