Guide

Imagine a B2B SaaS company exiting the year at $5.0M ARR. It grew during the year, so recognized revenue is $4.2M — average ARR over the year was lower than ending ARR.

Assumptions:

  • Gross margin 78% → COGS = 22% of revenue
  • S&M = $3.2M
  • R&D = $2.0M
  • G&A = $0.9M
  • D&A = $0.1M
  • Interest income on venture cash = $0.08M
  • Tax = $0 (losses)
Revenue                     $4,200,000
COGS (22%)                    −924,000
Gross profit                 3,276,000     GM = 78.0%

S&M                         −3,200,000
R&D                         −2,000,000
G&A                           −900,000
D&A                           −100,000
Operating income            −2,924,000

Interest income                +80,000
Net income                  −2,844,000

EBITDA ≈ operating income + D&A = −2,924,000 + 100,000 = −$2,824,000.

LineAmountNote
Revenue$4.2MGAAP; ARR exit $5M
COGS$0.92MHosting, support slice, payment fees
Gross profit$3.28M78% GM — SaaS-healthy
S&M$3.2M≈76% of revenue — growth mode
R&D$2.0M≈48% of revenue — still product-heavy
G&A$0.9M≈21% of revenue — watch this as you scale
Op. income−$2.92MNot yet leveraged
Net income−$2.84MExpected at this stage if growth is real

Operating leverage

If S&M stayed flat next year while revenue grew to $8M recognized, operating loss would shrink fast. That is operating leverage. If S&M grew in lockstep forever, it would not.

Trade-off

Invest in S&M vs print a prettier P&L. Cutting S&M improves near-term operating income and burn but stalls pipeline. Over-spending S&M before payback is proven buys vanity ARR. Prove CAC payback on a segment, then scale S&M against that segment.

At seed, the P&L may show $200k revenue and $1.5M opex — almost all R&D. That is normal if the goal is product learning. At Series B, expect revenue in the low millions to teens, S&M becoming the largest or second-largest opex line, and still deeply negative operating income if growth is the mandate.

A team-level sketch

You do not need NetSuite to sketch an engineering-aware contribution:

Revenue influenced (ARR tied to your surface)
− COGS you control (infra, model APIs, support)
= Gross profit contribution
− Your team opex (fully loaded headcount + tools)

Platform team of 8 at $20k fully loaded → $160k/month opex. They own infra that is $70k/month of company COGS on $500k/month revenue. A year-long rightsizing program cuts infra to $50k/month:

COGS savings = $20k/month = $240k/year
Team cost    = $160k × 12 = $1.92M/year

The program does not "pay for the whole team" — and it should not have to — but $240k/year is a concrete P&L win you can put next to feature work that might add $50k ARR.

Common mistakes

  • Reading $5M ARR as $5M revenue. Growing companies recognize less than ending ARR.
  • Treating a multi-million operating loss as failure at this stage if growth and payback are real.
  • Scoring only ARR shipped and under-investing in margin work.

Check yourself

  1. Why is recognized revenue $4.2M on $5M ending ARR? The company grew. Average ARR during the year was lower than the exit run-rate.
  2. S&M stays $3.2M and recognized revenue goes to $8M. What should happen to operating loss? It shrinks — that is operating leverage — unless other opex explodes.

Key takeaways

  • $5M ending ARR and $4.2M revenue can both be true.
  • 78% GM and a ~$2.9M operating loss can be "on plan" in growth mode.
  • Operating leverage appears when opex does not rise as fast as revenue.
  • Score margin wins next to ARR wins.

Further learning

Sal Khan’s intro to the income statement:

Introduction to the income statement | Stocks and bonds | Finance & Capital Markets | Khan AcademyKhan Academy

Back to the map: the P&L.