IX. Reference

Metric Definitions

Every figure in this report is either reproduced from the KBCM/Sapphire survey or derived from figures that are. This page gives the arithmetic for both, each metric with a typeset formula and a live calculator, so any number in the report can be reconstructed rather than taken on trust. The nine KBCM benchmark definitions are transcribed from the survey itself; the core SaaS metrics use their standard industry definitions; the derived metrics show the exact formula and the certified inputs the report used.

How to Use This Page

Each card carries the metric’s definition, its formula (expand LaTeX source under any formula to copy it), and a calculator pre-filled with figures that reproduce the report’s numbers. Change any input to see the result update. Two of the KBCM formulas, as printed in the Survey Benchmark Definitions on page 5 of the 2025 KBCM Survey Report, contradict their own definitions; both are flagged and corrected below, because a reader who followed the printed version would not reproduce the report’s figures. Every input used in the derived metrics traces to the source registry on the Data Integrity page.

KBCM Benchmark Definitions

The nine benchmark definitions from the 2025 KBCM/Sapphire Private SaaS Survey, page 5. The two marked with a dagger (†) or double-dagger (‡) print a formula that contradicts its own definition; the corrected form is shown on the card and explained in the next section.

Gross Dollar Retention

GDR

Share of recurring revenue retained from existing customers over a period, excluding any expansion from upsells.

LaTeX source
\text{GDR}=\dfrac{\text{Starting ARR}-\text{Churned ARR}-\text{Downsell ARR}}{\text{Starting ARR}}\times 100\%
Gross Dollar Retention86.0%

Example. KBCM GDR sits near 86% and is flat across the certified window (2022–2024); it is not published on a comparable basis before 2022.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

Net Dollar Retention

NDR

Share of recurring revenue retained including expansion, reflecting both churn and growth within the customer base.

LaTeX source
\text{NDR}=\dfrac{\text{Starting ARR}-\text{Churned ARR}-\text{Downsell ARR}+\text{Upsell ARR}}{\text{Starting ARR}}\times 100\%
Net Dollar Retention101.0%

Example. Median NDR held near 102% pre-ZIRP (102.7% in 2018, 102% in 2019, 101.8% in 2020), peaked at 109% in 2021, then reverted to 101% by 2024; all certified.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

Gross Magic Number

Sales efficiency: new-logo plus upsell ARR generated per dollar of S&M spend in the period.

LaTeX source
\text{Gross Magic}=\dfrac{\text{New Logo ARR}+\text{Upsell ARR}}{\text{S\&M Expense}}
Gross Magic Number0.65

Example. Gross Magic ran ~0.65–0.70 across 2022–2024, rising toward 0.82 in 2025E.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

Net Magic Number

Sales efficiency on net-new ARR per dollar of S&M: new logo plus upsell, net of churn.

LaTeX source
\text{Net Magic}=\dfrac{\text{New Logo ARR}+\text{Upsell ARR}-\text{Churned ARR}}{\text{S\&M Expense}}

As printed in the 2025 KBCM Survey Report, page 5 (not used):

† Page 5 of the 2025 KBCM Survey Report subtracts Upsell, contradicting its own definition ("net new ARR including upsell"). Net-new ARR adds upsell. The corrected form is used here.

Net Magic Number0.50

Example. Net Magic held at exactly 0.50 for four straight years (2022–2025E).

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5 (corrected)Verified

Fully-Loaded CAC (KBCM)

S&M spent per dollar of new ARR won. This is KBCM's efficiency-ratio sense of the term, distinct from this report's broader all-in cost (see the note at the foot of the page).

LaTeX source
\text{FL-CAC}=\dfrac{\text{S\&M Expense}}{\text{New Logo ARR}+\text{Upsell ARR}}
Fully-Loaded CAC (KBCM)0.70

Example. $0.70 of S&M per $1 of new ARR won ⇒ a 0.70 ratio.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

New-Only CAC

A CAC variant isolating the cost of winning brand-new customers, excluding expansion and upsell.

LaTeX source
\text{New-Only CAC}=\dfrac{\text{New-Customer S\&M Expense}}{\text{New Logo ARR}}
New-Only CAC0.80

Example. Isolating new logos exposes the 37-month new-only payback the blended number hides.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

CAC Payback Period

Months for a new customer's gross profit to repay its acquisition cost.

LaTeX source
\text{Payback}=\dfrac{\text{CAC}}{\text{ARPA}\times \text{Gross Margin}\div 12}

As printed in the 2025 KBCM Survey Report, page 5 (not used):

‡ Page 5 of the 2025 KBCM Survey Report prints CAC × (Gross Margin ÷ 12), which is not a duration and rises as margin improves, inverting the relationship. Payback is CAC divided by monthly gross profit; the corrected form reproduces the reported 24- and 37-month figures.

CAC Payback Period40.0 months

Example. Fully-loaded payback held ~24 months; new-only worsened to 37 months by 2024.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5 (corrected)Verified

Rule of 40

A SaaS health benchmark: revenue growth rate plus EBITDA margin should sum to at least 40%.

LaTeX source
\text{Rule of 40}=\text{Revenue Growth \%}+\text{EBITDA Margin \%}
Rule of 4036%

Example. Median R40 recovered from −14% to +6% (2022→2024), yet only 5% of companies clear 40.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

AE Productivity

New plus expansion ARR generated per quota-carrying account executive.

LaTeX source
\text{AE Productivity}=\dfrac{\text{New Logo ARR}+\text{Expansion ARR}}{\#\ \text{Quota-Carrying AEs}}

KBCM cautions that recent "improvements" stem from smaller teams after layoffs, not stronger reps: the denominator shrank faster than the numerator.

AE Productivity$1.00M / AE

Example. $6M of new + expansion ARR across 6 AEs ⇒ $1.00M per quota-carrying AE.

Source: KBCM/Sapphire 2025 Private SaaS Survey, p5Verified

Two Formula Corrections

Two of the nine formulas, as printed in the Survey Benchmark Definitions on page 5 of the 2025 KBCM Survey Report, do not match their own definitions. This report uses the corrected forms, which are what produce its figures.

† Net Magic Number. Page 5 prints (New Logo ARR − Upsell ARR − Churned ARR) ÷ S&M Expense, but the definition beside it on page 5 describes net new ARR “including new logo, upsell, and churned ARR.” Net new ARR adds upsell; it does not subtract it. Subtracting it would push Net Magic Number below Gross Magic Number by twice the upsell contribution, which is not what the metric means. This report uses the definition-consistent form:

‡ CAC Payback Period. Page 5 prints CAC × (Gross Margin ÷ 12). Payback is a duration, a cost divided by the monthly gross profit that repays it, so the result should fall as gross margin improves, because higher margin repays the cost faster. As printed, the result instead rises with margin, which inverts the relationship. This report uses CAC divided by monthly gross profit:

which yields the reported 24-month (fully-loaded) and 37-month (new-only) figures.

Core SaaS Metrics

The recurring-revenue base, the churn-and-retention family, and the unit-economics ratios that every SaaS operator tracks, distinct from the KBCM survey benchmarks above, and stated in their standard industry definitions. Each formula is the conventional definition operators use, so any figure here can be reproduced without reference to the survey.

Annual Recurring Revenue

ARR

The annualized value of recurring subscription revenue: the normalized run-rate a SaaS business plans and is valued against.

LaTeX source
\text{ARR}=\text{MRR}\times 12
Annual Recurring Revenue$12,000,000

Example. $1.0M MRR ⇒ $12.0M ARR.

Source: Standard SaaS definition

Monthly Recurring Revenue

MRR

Total predictable subscription revenue billed each month across all active accounts.

LaTeX source
\text{MRR}=\text{Active Accounts}\times \text{ARPA}_{\text{monthly}}
Monthly Recurring Revenue$1,000,000

Example. 1,000 accounts × $1,000 ARPA ⇒ $1.0M MRR.

Source: Standard SaaS definition

Average Revenue Per Account

ARPA

Average recurring revenue per account, the pricing-power denominator behind CAC payback and LTV. (ARPU is the per-user variant.)

LaTeX source
\text{ARPA}=\dfrac{\text{MRR}}{\text{Active Accounts}}
Average Revenue Per Account$1,000

Example. $1.0M MRR ÷ 1,000 accounts ⇒ $1,000 ARPA.

Source: Standard SaaS definition

Gross MRR Churn Rate

Revenue lost to cancellations and downgrades as a share of starting MRR. Expansion is ignored, so it can never go negative.

LaTeX source
\text{Gross MRR Churn}=\dfrac{\text{Churned MRR}+\text{Contraction MRR}}{\text{Starting MRR}}\times 100\%
Gross MRR Churn Rate3.0%

Example. $30k lost on $1.0M starting MRR ⇒ 3.0% gross churn.

Source: Standard SaaS definition

Net MRR Churn Rate

Gross churn net of expansion. A negative value is "net-negative churn": expansion outruns losses, the growth engine behind NDR > 100%.

LaTeX source
\text{Net MRR Churn}=\dfrac{\text{Churned MRR}+\text{Contraction MRR}-\text{Expansion MRR}}{\text{Starting MRR}}\times 100\%
Net MRR Churn Rate-1.5%

Example. Expansion of $45k against $30k lost ⇒ −1.5% net churn (healthy).

Source: Standard SaaS definition

Logo (Customer) Churn Rate

The count-based churn rate: what fraction of customers left, regardless of their size. Diverges from revenue churn when churn skews small or large.

LaTeX source
\text{Logo Churn}=\dfrac{\text{Customers Lost}}{\text{Starting Customers}}\times 100\%
Logo (Customer) Churn Rate2.8%

Example. 14 of 500 customers lost ⇒ 2.8% logo churn.

Source: Standard SaaS definition

Customer Lifetime Value

LTV

The gross-profit a customer contributes over its expected lifetime. Gross-margin-adjusted, since revenue LTV overstates value.

LaTeX source
\text{LTV}=\text{ARPA}\times\dfrac{1}{\text{Logo Churn Rate}}\times \text{Gross Margin \%}
Customer Lifetime Value$37,500

Example. $1,000 ARPA × 75% GM ÷ 2% churn ⇒ $37,500 LTV.

Source: Standard SaaS definition

LTV : CAC Ratio

Return on each acquisition dollar. Rule of thumb: ≥ 3× is healthy; < 1× destroys value; very high can signal underinvestment in growth.

LaTeX source
\text{LTV:CAC}=\dfrac{\text{LTV}}{\text{CAC}}
LTV : CAC Ratio3.0×

Example. $37,500 LTV ÷ $12,500 CAC ⇒ 3.0×, the classic healthy threshold.

Source: Standard SaaS definition

Expansion Rate

Upsell and cross-sell revenue from existing customers as a share of starting MRR; the lever that pushes NDR above GDR.

LaTeX source
\text{Expansion Rate}=\dfrac{\text{Expansion MRR}}{\text{Starting MRR}}\times 100\%
Expansion Rate4.5%

Example. $45k expansion on $1.0M MRR ⇒ 4.5% expansion.

Source: Standard SaaS definition

ARR Growth Rate

Period-over-period growth of the recurring run-rate, the numerator of the Rule of 40.

LaTeX source
\text{ARR Growth}=\dfrac{\text{Ending ARR}-\text{Beginning ARR}}{\text{Beginning ARR}}\times 100\%
ARR Growth Rate30.0%

Example. $10M → $13M ARR ⇒ 30% growth.

Source: Standard SaaS definition

Burn Multiple

Dollars burned per dollar of net-new ARR added. Bands: < 1 great, 1–1.5 good, 1.5–2 suspect, > 2 concerning.

LaTeX source
\text{Burn Multiple}=\dfrac{\text{Net Burn}}{\text{Net New ARR}}

The reciprocal of the Bessemer Efficiency Score (net-new ARR per dollar of net burn). Burn Multiple (net burn per net-new-ARR dollar) is the more commonly cited operator form.

Burn Multiple1.25×

Example. $10M burned for $8M net-new ARR ⇒ 1.25×, "good" territory.

Source: Standard SaaS definition

SaaS Quick Ratio

Revenue added per dollar of revenue lost. ≥ 4 signals efficient, durable growth; near 1 means new revenue barely offsets churn.

LaTeX source
\text{Quick Ratio}=\dfrac{\text{New MRR}+\text{Expansion MRR}}{\text{Churned MRR}+\text{Contraction MRR}}
SaaS Quick Ratio3.17×

Example. $95k added ÷ $30k lost ⇒ 3.17× (approaching the ≥4 efficiency bar).

Source: Standard SaaS definition

Customer Acquisition Cost (blended)

CAC

The classic count-based CAC: sales-and-marketing spend per new customer. (KBCM instead reports an ARR-ratio CAC; see the KBCM tier above.)

LaTeX source
\text{CAC}=\dfrac{\text{S\&M Expense}}{\text{New Customers Acquired}}
Customer Acquisition Cost (blended)$12,500

Example. $500k S&M ÷ 40 new customers ⇒ $12,500 CAC.

Source: Standard SaaS definition

Report-Derived Metrics

These are the report’s own calculations from KBCM figures. Each uses inputs that are certified against the source registry, so the worked results can be reconstructed directly.

Expansion-to-Churn Coverage

Net-expansion points earned per point of gross churn: how much cushion expansion provides over the revenue base churn is eroding.

LaTeX source
\text{Coverage}=\dfrac{\text{NDR}-\text{GDR}}{100-\text{GDR}}
Expansion-to-Churn Coverage1.07×

Example. 2021: (109−86)/14 = 1.6×. 2024: (101−86)/14 = 1.07×, and the cushion is thinning.

Source: Report-derived from certified NDR/GDR; endpoints certified, middle years interpolatedCalculated

Churn Tax

The slice of sales-and-marketing efficiency consumed replacing churned revenue rather than growing: the gap between gross and net Magic Number.

LaTeX source
\text{Churn Tax}=\text{Gross Magic}-\text{Net Magic}
Churn Tax0.23

Example. 0.73 gross − 0.50 net = 0.23 of every S&M dollar lost to replacing churn (~24%+).

Source: Report-derived from certified Magic Number seriesCalculated

Churn Before Payback

The share of a new cohort expected to churn before it repays its acquisition cost, where a long payback and everyday churn collide.

LaTeX source
\text{Share}=\left[1-\left(1-\text{Annual Churn}\right)^{\text{Payback}\div 12}\right]\times 100\%
Churn Before Payback37%

Example. 14% annual churn × 37-month payback ⇒ ~37% of a cohort churns before it pays back.

Source: Report-derived from certified churn + CAC-payback figuresCalculated

Outputs and Their Feeder Metrics

Outputs and their feeder metrics. Each row shows a scorecard output and the feeder metrics that move it.
Output (what the scorecard prints)Its feeder metrics (what actually moves it)
GDRGross churn rate, downsell rate, logo churn, contract-length mix, services attach
NDREverything that feeds GDR, plus expansion rate and expansion’s coverage of churn
Net magic numberGross magic number, less the churn and downsell consumed from each S&M dollar
Rule of 40ARR growth (itself fed by the mix above) and EBITDA margin (fed by the OpEx lines)

The standing example of why the distinction matters: NDR printing 101% while churn destroys 14 points and expansion replaces 15. The output reads as stability; the feeders describe a system with almost no slack. Wherever this report says “read the feeders,” it means judge the business by the right-hand column, because the left-hand column is built to net the story away.

On “Fully-Loaded CAC”

The survey defines “Fully-Loaded CAC” as S&M (sales and marketing) ÷ (New Logo + Upsell ARR), an efficiency ratio: the S&M spent per dollar of new ARR won. This report uses the phrase “fully-loaded CAC” in a different, broader sense: the all-in cost of acquisition once R&D, onboarding, customer-success, and overhead are added to S&M, which runs about 2.8× the reported, S&M-only CAC. The two share a name but measure different things. Where this report says “fully-loaded CAC,” it means the all-in cost, not the survey’s efficiency ratio.

Frequently asked questions

What is the difference between GDR and NDR?

Gross dollar retention counts only losses, so it can never exceed 100%. Net dollar retention adds expansion from existing customers, so it can. Both net downsell away, which is why neither shows it.

What does the Net Magic Number measure?

The Net Magic Number divides new logo ARR plus upsell ARR minus churned ARR by sales and marketing expense in the same period. Stuck at 0.50 for four straight years, it means fifty cents of net new ARR per dollar spent.

What is fully-loaded CAC and why does it differ?

Fully-loaded CAC adds R&D, onboarding, customer success, and overhead to sales and marketing, capturing the all-in cost of winning and keeping a customer. In this data it runs about 2.8 times the reported, S&M-only CAC.

How is expansion-to-churn coverage calculated in this report?

Subtract gross dollar retention from net dollar retention, then divide by 100 minus gross dollar retention. At 101% NDR and 86% GDR that gives 1.07 times, down from 1.6 times in 2021. The denominator counts churn plus downsell.

Last reviewed: July 2026

Previous
The Benchmark's Blind Spots