PLG Self-Serve SaaS unit economics, benchmarked for 2026.

Product-led growth replaces the sales team with the product itself: users discover the tool, activate themselves, invite teammates, and eventually swipe a corporate card — all without talking to anyone. The median sales cycle is effectively one month, CAC lands between $300 and $1,800, and ARPA spans $30–$250. It looks like the cheapest motion in SaaS, and per-transaction it is. But PLG economics are a volume game governed by multiplication: visitor-to-signup, signup-to-activated, activated-to-paid, paid-to-expanded. Four conversion rates, each modest, multiply into a funnel where small leaks compound catastrophically.

The structural nuance founders miss is that "self-serve" still has a cost of goods. Freemium users consume hosting, bandwidth, and support; a free tier with 50,000 users and 3% conversion is 48,500 people you pay to serve in hopes the math works. Retention dynamics split in two: individual users churn fast (3.0% monthly median), but accounts that reach team-wide adoption expand through seats and exhibit the collaborative stickiness that built Notion and Figma. The cohort's economics therefore hinge on two rates — activation and expansion — far more than on any single CAC figure.

First, blended CAC must include the free tier's carrying cost. Infrastructure, abuse prevention, and support for tens of thousands of non-paying users is real spend; founders who report "CAC of $400" while ignoring $25k/month in free-tier hosting are measuring a fantasy. Always compute CAC per paying customer, fully loaded.

Second, the content and SEO engine. In PLG the content team is the sales team — salaries for writers, video producers, and SEO specialists replace AE salaries dollar for dollar. A six-person content org at $700k fully loaded, producing 200 paying customers a month, is a $290 CAC before a cent of ad spend.

Third, paid acquisition layered on top. Search and social ads work in PLG only when activation is already strong; bought traffic converts at the same activation rate as organic, so every point of activation improvement cuts paid CAC proportionally.

Fourth, the stealth sales assist. Many "product-led" companies quietly carry AEs who work product-qualified leads into team and enterprise plans. That headcount belongs in CAC — calling it "customer success" doesn't change the accounting.

An 8-to-11-month median payback reflects the funnel's reality: cheap acquisition, but modest ARPA and meaningful churn. Top-quartile teams recovering CAC in under 6 months combine elite activation (north of 25% signup-to-paid) with viral K-factors — shared artifacts, "powered by" badges, and team invites that acquire users at zero marginal cost. When expansion kicks in, seat growth shortens effective payback further.

The bottom quartile — payback beyond 14 months — almost always traces to one pathology: paid CAC that was never profitable, masked by organic volume in the blended number. Split paid and blended CAC and the truth appears: organic carries the business while ads quietly lose money. Read the 3.8x LTV:CAC strictly on blended CAC; on paid-only CAC the ratio is often half that. And measure payback from the conversion month, not signup — a user who sits free for five months before paying has a payback clock that starts at conversion, and cohorting any other way flatters the numbers.

The most expensive mistake in PLG is scaling paid spend against a broken activation rate. The funnel math is unforgiving: at a $40 cost per signup and 5% signup-to-paid conversion, true CAC per paying customer is $800 — but the dashboard reports $40, and the founder scales budget against the $40. Every additional ad dollar then amplifies the leak rather than the business: more signups, same 95% failure to activate, linearly more free-tier hosting cost. Teams burn $50k a month "growing" while the activated-user count barely moves, then conclude "paid doesn't work for PLG" when the actual diagnosis is that onboarding doesn't work. The fix is sequenced, not budgetary: instrument activation, lift signup-to-paid past 15–20% with reverse trials and checklist onboarding, and only then open the paid tap. In product-led growth, acquisition spend is a multiplier on activation quality — and any number multiplied by near-zero is near-zero.