Business Metrics
Intermediate 15 min read Business Metrics

Unit Economics

One profitable customer model, many customers later

Unit economics asks: does one customer eventually pay back what they cost to acquire? LTV, CAC and payback period are the three numbers that make growth sustainable or fatal.

unit economicsLTVCACpayback periodcontribution margingross marginLTV:CAC
Prerequisites: Revenue Metrics
Before we start

What you will learn

  • Compute CAC and LTV from data
  • Interpret LTV:CAC and payback period
  • Explain the trap of masking CAC with growth
Introduction

What is this?

Unit economics measures the profitability of one unit (one customer, one order) over its life. CAC is the acquisition cost; LTV is the lifetime value net of the cost to serve; payback period is how long it takes one customer's contribution to repay CAC. The ratio LTV:CAC roughly says whether a customer ultimately pays for themselves — generally above 3 for healthy SaaS.

Motivation

Why do you need it?

  • Growth without unit economics is borrowing from the future with a script that never repays.
  • CAC can look fine while payback lengthens — danger hidden in an average.
  • Improvement in LTV:CAC is the strongest signal that a business model works.
  • Interviewers use these prompts to test commercial literacy, not just SQL chops.
Think about it

Mental model

Imagine

Opening a lemonade stand

Think of

Unit economics

Each glass costs cups, lemons, sugar and your time (cost to serve); a regular customer buys 40 glasses a summer (LTV). If your marketing flyer costs a dollar to reach each new lemonade lover and they only ever buy 50 cents of profit, the stand grows to bankruptcy.

Mechanics

How it works

  • CAC = total acquisition spend (marketing + sales) / number of new paying customers in the period.
  • Gross margin % removes cost of goods before lifetime calculations.
  • LTV = (average revenue per customer per month x gross margin) x average customer lifetime in months.
  • Lifetime derives from retention: 1/(churn rate) months is the classic model.
  • Payback period = CAC / (monthly contribution margin).
  • LTV:CAC = the ratio of lifetime contribution to acquisition cost.

The economics of one customer over time.

Acquire
Month 1
Revenue months
Payback
Customer ends
Hands-on

Examples

Payback from a churn sheet

Payback from a churn sheet
-- Blended snapshot for the quarter
SELECT
  SUM(ua_spend) / NULLIF(SUM(new_paying), 0)  AS cac,
  -- monthly contribution per customer
  AVG(customer_contribution_margin) / NULLIF(churn_rate, 0)  AS ltv_month_model,
  SUM(ua_spend) / NULLIF(SUM(new_paying), 0)
    / NULLIF(AVG(customer_contribution_margin), 0) AS payback_months
FROM quarterly_totals;

CAC comes from spend per new paying customer; LTV collapses to contribution over churn in the textbook model; payback divides CAC by monthly contribution so the business sees recovery time, not just a ratio.

Real world

In production

  • A growth team sets UA budgets against a payback < 12 months gate and pauses channels that break it.
  • A CFO reviews LTV:CAC per channel monthly, reallocating spend from low-ratio cohorts.
  • A product team improves onboarding to lift activation and LTV rather than boosting CAC to smooth retention.
Guidelines

Best practices

  • Compute CAC per channel and cohort, not just blended.
  • Use gross margin in LTV; lifetime revenue without margin inflates LTV.
  • Define churn over the same cohort window as the lifetime formula.
  • Publish the assumption sheet (churn rate, margin, payback) alongside the numbers.
Watch out

Common mistakes

  • Dividing UA spend by all users when only paying customers repay CAC.
  • Modeling LTV with constant churn when cohorts mature.
  • Using blended CAC to hide disastrous channels.
  • Treating LTV:CAC as a law when definition has changed the numbers.
Root cause analysis

Common issues

Error
LTV:CAC improves but cash worsens
Root cause

Payback period lengthened while the ratio improved via margin shifts

Fix

Track payback and cash timing alongside the ratio.

Error
Constant churn assumption fails for older cohorts
Root cause

Churn decays with tenure in reality

Fix

Use per-tenant survival curves instead of a single rate.

Error
CAC doubles when a campaign goes silent
Root cause

Untracked touchpoints were absorbed into other lines

Fix

Attribute spend at the channel/experiment level before reporting CAC.

When things break

Debugging tips

  • Rebuild one cohort's LTV by hand from the revenue table to catch model drift.
  • Compare CAC by channel to spot the channel that absorbs spend.
  • Plot payback per cohort: improving payback implies repair, declining implies danger.
  • Test LTV sensitivity to churn = 1%. Check margin: LTV is margin-scaled, not revenue-scaled.
Interviews

Interview checkpoint

Test yourself

Quick quiz

0/2
1

What does payback period measure in months?

2

Revenue-based LTV is a fine substitute for margin-based LTV.

Go deeper

Official documentation

Unit economics (a16z framework)
docs
SaaS metrics guide (OpenView)
docs