Unit economics

If you can't explain how one more customer makes you more profitable or less, you don't have a business model, you have a pile of transactions. Unit economics is the discipline of reducing the entire business down to what happens with one additional customer, one additional job, one additional unit sold. Everything else, fundraising, hiring, strategy, is downstream of this.

The core equation

For a subscription business:

LTV / CAC

where
  LTV = (ARPU × Gross Margin %) ÷ Monthly Churn %
  CAC = (Sales + Marketing spend) ÷ New Customers Acquired

Rule of thumb: LTV/CAC ≥ 3 to be venture-scale. ≥ 1 to not be actively losing money. Below 1 means you lose money on every customer and "make it up in volume", the oldest lie in business.

For a services business

The equation changes shape, but the principle is the same:

• Gross margin per engagement, revenue minus the direct cost of delivery
• Utilization rate, what % of paid hours are billable
• Effective bill rate, revenue per consultant hour
• Cost of acquisition per engagement

You want an answer to: "If I added one more client like my average client, what would it do to cash, margin, and capacity?"

CAC, the trap

Most teams underestimate CAC. The common mistakes:

• Excluding salaries. If your marketer earns $120K/year and generates 100 customers, your CAC is $1,200 higher than "ad spend divided by customers."
• Counting organic wrong. Inbound leads that closed themselves aren't free. Your content, SEO, and brand all cost money.
• Blended vs new-logo CAC. If you report expansion revenue as "new customers," your CAC looks artificially low.
• Payback period confusion. LTV/CAC ratio matters, but so does how long until you've recouped the CAC in gross profit. < 12 months is healthy; > 24 months creates a cash problem even with a good LTV ratio.

LTV, the trap

• Ignoring churn. A business with 2% monthly churn has a max LTV of 50 months of ARPU. Feels like a lot until you realize that's 4 years, and your product might not exist in 4 years.
• Using revenue, not gross profit. LTV should use gross profit, not revenue. Otherwise you're counting dollars you have to spend to deliver.
• Averaging across segments. Enterprise LTV and SMB LTV are wildly different. A blended number hides the fact that one segment is amazing and the other is losing money.

Segmented unit economics

Almost always, the aggregate LTV/CAC hides the real story. Segment by:

• Channel, paid search vs referral vs outbound
• Customer size. SMB vs mid-market vs enterprise
• Product line, core product vs add-ons
• Cohort, by quarter signed

The action is always the same: double down on the channel/segment/product where economics are strong; starve or reprice the ones where they're not.

When to care obsessively

Unit economics matter most when:

• You're raising money (investors will model this to the decimal)
• You're scaling paid acquisition (bad unit economics × more ad spend = faster bleed)
• You're making a pricing change
• You're evaluating a new market or segment

What good looks like

• You can quote LTV, CAC, and LTV/CAC by segment from memory
• Payback period is tracked monthly
• Your CAC calculation includes all fully-loaded costs
• When the marketing team pitches a new channel, you model it in unit economics terms before approving spend

Related: Pricing frameworks · Funnel math · Churn diagnostics