B2B SaaS operating benchmarks: what good looks like in 2026

What is the Rule of 40 and does it matter below $10M?

The Rule of 40 adds your revenue growth rate to your profit margin; a combined score of 40% or better is the classic institutional bar. Almost nobody clears it anymore: the median private B2B SaaS company scored 15% in CY2024, down from 23% in CY2022 (Benchmarkit), and only the top quartile, at 35% and above in Benchmarkit's sample and roughly 40% and above in High Alpha's separately sampled cohort, reaches the traditional threshold. Below $10M it matters as a lens, not a gate: SEG's published buyer screening thresholds for the segment are GRR, gross margin and NRR, not a composite score. At that scale, profitability itself carries the price.

The three-year slide is worth reading closely, because its cause shows up elsewhere in this section. Growth compressed (section 01) faster than margins recovered, and the cost of growth rose: median CAC payback lengthened from 14 months in CY2023 to 18 months in CY2024 (Benchmarkit). At the twenty-fifth percentile the Rule of 40 score is minus 4%, meaning a full quarter of private SaaS companies are shrinking on a combined basis. Against that backdrop, a modest but efficient operator is not a consolation-prize asset; it is statistically rare.

The bootstrapped lens: slower, leaner, closer to the money

The data file behind this page keeps bootstrapped rows on purpose, because the funding split runs through every table. Bootstrapped companies grow slower (20% median versus 25% equity-backed, SaaS Capital 2026) and retain slightly less on the net line (103% NRR median at $3M to $20M ARR). But they buy their growth at a fraction of the cost: the median magic number is 2.85 for bootstrapped companies against 0.70 for VC-backed peers (Benchmarkit, CY2024), and 83% of bootstrapped companies are within two points of breakeven or profitable, versus 52% of equity-backed ones (SaaS Capital, 2026).

In a market where the median Rule of 40 score is 15%, that efficiency is the whole game. A bootstrapped company growing 20% at a 15% margin posts a 35% score, which sits at Benchmarkit's seventy-fifth percentile, purely as arithmetic. The positioning conclusion we act on in mandates: bootstrapped sellers rarely win the growth table and usually win the efficiency table, so the preparation work is making the efficiency story diligence-proof, with margins by cost line, CAC by channel, and revenue per employee ($130K median at $5M to $20M ARR, with best-in-class above $350K) laid out before a buyer asks.

Buyers rebuild your metrics; they do not accept them. A serious acquirer or their quality-of-earnings advisor starts from the raw billing ledger, not from your dashboard. They reconstruct MRR movements month by month, classify every dollar as new, expansion, contraction or churn, and recompute NRR, GRR and growth from first principles. The benchmarks on this page are the reference frame those rebuilt numbers get held against. If your reported metrics survive the rebuild, the conversation stays about price. If they move, every other number you reported inherits the doubt, which is a worse outcome than the metric itself being mediocre.

The definitional traps are where reported and rebuilt numbers diverge. The most common ones we see: logo churn quoted where a buyer expects revenue churn, which flatters SMB-heavy books when the churning accounts are small. Monthly churn quoted where a buyer works in annual terms, which understates the compounding, as the arithmetic in section 02 shows. In-month retention quoted where a buyer computes trailing-twelve-month cohorts, which lets one good month stand in for a year. NRR computed only on customers who renewed, which quietly excludes the churn it exists to capture. Gross margin with support or hosting costs sitting below the line. None of these are fraud; all are definitional choices a rebuild will reverse, at the least convenient moment of the process.

Survey medians differ from each other for knowable reasons. This page keeps publishers separate because their universes are not the same: surveys (SaaS Capital, Benchmarkit, High Alpha) measure self-reported annual figures from companies motivated to respond, while platform data (ChartMogul) is measured from billing systems and skews SMB and self-serve. That is why bootstrapped growth medians range from 20% to 34% across samples, and why survey retention medians and platform churn medians can both be true while pointing in different directions. When a buyer's benchmark disagrees with yours, the first question is never whose number is right; it is which population and which definition each number describes.

The practical move is to run the rebuild on yourself first. Before a process starts, compute your own metrics on buyer definitions: revenue and logo churn, both; trailing-twelve-month cohorts, not last month; NRR over the full starting base; margins with every cost of service above the line. Where the honest number is weaker than the reported one, disclose it with the explanation attached. A weak metric disclosed early is a discount conversation; a metric that changes definition mid-diligence is a trust conversation, and trust conversations reprice more than metrics do.