- Home
- PLG Metrics
PLG Metrics
The KPIs that actually predict product-led growth — activation, TTV, PQL conversion, expansion MRR and the experimentation metrics most teams miss.
Most PLG dashboards measure the things that are easy to instrument rather than the things that predict revenue. Signups, MAU, page views, NPS. None of them tell you whether the growth model works.
These are the metrics I actually run programs against, in the order they matter, with the benchmark ranges I've seen hold up across SaaS and high-intent FinTech funnels.
Tier 1 — Acquisition metrics
Organic signups as a percentage of total. The cleanest single health signal for a PLG model. If 70%+ of signups arrive without paid spend, the product and its content are doing the selling. If organic is under 30% and falling, you have a paid acquisition business wearing PLG clothes, and your CAC will only go up.
Signup-to-activation rate. The most important single metric in PLG, full stop. Every downstream number — retention, conversion, expansion, referral — is a multiple of it. Improving it 10 points does more for revenue than any pricing change you can ship in the same quarter.
Channel-level activation rate. Paid, organic and referral traffic activate at wildly different rates — commonly 3x spread between paid and referral. Blended activation hides this completely. Segment it or you will keep optimizing onboarding for a targeting problem.
Practical rule:never report activation as one number. Report it by channel, cohort week, and plan.
Tier 2 — Activation metrics
Time to value (TTV). Median minutes or hours from signup to the first meaningful action. Measure the median and the 90th percentile — the tail is where churn hides. TTV is the metric onboarding work should move; activation rate is the metric that proves it worked.
Activation rate definition. What counts as activated varies by product, and you have to define it empirically before you can measure it. Compare 90-day retained cohorts against churned cohorts and find the first-week event with the largest separation between them. That event is activation. Do not vote on it in a meeting.
Feature adoption depth versus breadth. Breadth (how many features touched) is a vanity measure. Depth — repeat use of the one or two core actions — predicts retention. A user who does the core action four times in week one is worth more than one who tried eight features once.
Tier 3 — Retention and expansion
D1 / D7 / D30 retention. Run these as cohort curves, not averages. The shape matters more than the level: a curve that flattens at 20% is a real business; one that keeps sloping down has no floor and no flywheel.
PQL definition and conversion rate. A product qualified lead is an account that has hit usage behavior correlated with purchase — not someone who downloaded a PDF. Define the behavior, score against it, and track PQL→paid conversion. This is the metric that makes sales and product speak the same language.
Expansion MRR from self-serve. Split expansion into self-serve versus sales-assisted. Self-serve expansion is the compounding part of the model; sales-assisted expansion is a service business. Healthy PLG shows net revenue retention above 100% with the majority of expansion happening without a human.
Tier 4 — Experimentation metrics
These are the ones almost nobody reports to leadership, and they are the best leading indicator of whether the other three tiers will improve.
Test velocity. Experiments shipped per month. Learning rate beats win rate: a team running six tests a month with a 25% win rate will out-improve a team shipping one carefully argued redesign a quarter.
Win rate. Percentage of experiments that beat control at your significance threshold. Healthy programs land between 20% and 35%. A win rate above 60% usually means underpowered tests or peeking, not genius.
Confidence calibration. When your team says it's 80% confident, does the test win 80% of the time? Log the pre-test prediction on every experiment and check it quarterly. Miscalibrated teams burn traffic on tests they already knew the answer to — or ship redesigns they had no business being confident about.
FinTech-specific metrics
The standard SaaS metric set needs translation when there is no free tier.
- Application completion rate is the activation proxy. Signup means nothing; a submitted application is the value moment.
- Verification step conversion — KYC, identity, income and document upload — is the critical activation gate. In most lending funnels this single step accounts for the largest drop in the entire journey. Instrument it per sub-step, not as one event.
- Return borrower rate is the expansion equivalent. Refinance, second product, repeat purchase. It replaces seat expansion as the compounding revenue term.
Benchmark ranges
| Metric | Early stage | Growth stage | Mature PLG |
|---|---|---|---|
| Signup-to-activation | 20–30% | 35–50% | 50%+ |
| D30 retention | 15–25% | 25–40% | 40%+ |
| PQL conversion | 5–10% | 10–20% | 20%+ |
| Test velocity | 1–2 / month | 4–8 / month | 10+ / month |
Treat these as orientation, not targets. A vertical SaaS tool with a two-week evaluation cycle and a consumer lending funnel with a single-session decision will sit in different places for structural reasons.
What investors actually look at
Net revenue retention, the shape of the cohort retention curve, organic share of acquisition, CAC payback, and — increasingly — activation rate, because it is the earliest credible signal that the model compounds. A deck with growing signups and flat activation reads as a paid-growth story.
Where to go next
- PLG strategy: how to build the operating system these metrics measure
- Growth experimentation: how to move the activation number
- The PLG flywheel: why activation multiplies everything downstream
- PLG software: the analytics and experimentation stack behind these numbers
Key takeaway:instrument activation properly first. Every other metric on this page is downstream of it.
Tactical Insights On Product-Led Growth
Get weekly PLG insights — experimentation frameworks, activation teardowns, and FinTech growth plays from someone who's run it at scale. No filler.
No spam. Unsubscribe anytime.