Source: https://www.revvly.ai/blog/product-led-growth-revenue-metrics

SaaS

9 min read

Aug 1, 2026

# Product-led growth metrics: the revenue numbers most guides skip

Most teams can tell you who activated. Fewer can tell you, with real numbers, what that activation was worth.

AR

Alain Readman

Co-founder, Revvly

Ask a product or growth leader at a subscription SaaS company what metrics they track for product-led growth, and you'll get a fast, confident answer: activation rate, product-qualified leads, free-to-paid conversion, maybe a viral coefficient if the product has a sharing loop. Ask the same person how much revenue their self-serve motion actually generated last quarter, split from what sales-assisted upgrades generated, and the answer gets slower. Most teams can tell you who activated. Fewer can tell you, with real numbers, what that activation was worth.

That gap exists for a structural reason. The guides that define "PLG metrics" are almost all written by product analytics vendors, tools that instrument clicks, sessions, and feature adoption, not billing events. They're excellent on the engagement side of the funnel and thin on the revenue side, because the revenue side lives in a different system than the one they sell.

## Quick answer

> Product-led growth metrics fall into two families: engagement metrics (activation rate, PQL rate, feature adoption, viral coefficient) and revenue metrics (self-serve conversion revenue, expansion MRR, net revenue retention by acquisition motion, usage-to-plan-tier fit). Most published PLG metric lists are engagement-heavy because they come from product analytics platforms that don't have billing data. The revenue-side metrics require connecting product usage directly to subscription and billing records, not just event logs.

## Why most PLG metrics lists are activation-heavy

Look at any widely cited PLG metrics guide and the pattern repeats. Mixpanel's own 2026 guide to product-led growth lists activation rate, product-qualified leads, free-to-paid conversion rate, viral coefficient, and retention by activation cohort as the "essential" metrics, with formulas provided for activation rate and viral coefficient.[1] Expansion revenue gets one paragraph: track it by cohort and by product area, no formula, no connection to billing data, no breakdown by how the expansion happened.

That's not a criticism of the guide. It's an accurate reflection of what a product analytics platform can actually see. Mixpanel, Amplitude, and similar tools instrument behavioral events: clicks, page views, feature usage, session length. They can tell you a user upgraded because they hit a usage cap. They generally can't tell you what that upgrade was worth in monthly recurring revenue, whether it renewed at the new tier the following month, or whether the account's net revenue contribution actually grew once you account for any concurrent contraction elsewhere in the account.

That distinction matters because the questions a product or growth leader actually needs answered are revenue questions: which packaging change moved the needle on expansion, which usage threshold reliably predicts an upgrade, whether the self-serve motion is compounding or just adding logos without adding revenue. Those questions can't be answered from event data alone.

## Five revenue-side PLG metrics worth tracking directly

### Self-serve to paid conversion rate, measured in revenue, not just accounts

The standard version of this metric counts converted accounts over total trial or freemium signups. That's useful for a headline number, but it treats a low-tier conversion the same as a high-tier one. The more useful version weights conversions by the plan value they convert into, so a spike in "conversion rate" driven entirely by low-tier signups doesn't get mistaken for revenue momentum.

The underlying benchmarks vary widely by trial design. First Page Sage's analysis of 86 SaaS companies from Q1 2022 through Q3 2025 found organic free-trial-to-paid conversion averaging 18.2% for opt-in trials versus 48.8% for opt-out trials that require a card upfront, a gap wide enough that trial design alone can swing your headline conversion number by more than 2.5x before any product change.[2] If you're comparing your own self-serve conversion rate against a published benchmark without knowing which trial model produced it, you're comparing the wrong thing.

### Expansion MRR from in-product upgrades

Expansion MRR is the additional monthly recurring revenue existing customers generate through tier upgrades, seat additions, add-on purchases, or usage-based overages, isolated from new-customer MRR. Pavilion and Benchmarkit's 2025 B2B SaaS Benchmarks report, drawn from its industry-wide survey base, found that existing customers now generate 40% of new ARR industry-wide, rising above 50% for companies over $50M in ARR.[3] For a PLG motion specifically, where the product itself is supposed to surface upgrade moments without a sales conversation, this number is close to a report card: if in-product expansion isn't a growing share of new ARR, the self-serve upgrade paths aren't working as intended, whatever the activation dashboard says.

The operational trap is treating "someone upgraded" as the event to track instead of "how much MRR did that upgrade add, and did it stick past the first renewal." A [usage-based pricing model](https://www.revvly.ai/blog/usage-based-pricing) in particular needs this distinction, since overage charges can spike in a single month and vanish the next if usage was a one-time event rather than a durable shift in how the account uses the product.

### Net revenue retention, split by acquisition motion

NRR compressed to a median of 101% industry-wide in the Pavilion/Benchmarkit 2025 report, a signal that expansion is getting harder to come by across the board.[3] A single blended NRR number for a company running both self-serve and sales-assisted motions hides which motion is actually driving that number. If self-serve NRR is declining while sales-assisted NRR holds steady, blending them into one figure can mask a real problem in the PLG funnel for months.

This is a case where the metric definition itself doesn't change, NRR is still expansion minus contraction minus churn over a starting revenue base, but the segmentation does. Calculating it separately for accounts that never touched sales versus accounts that did requires tagging revenue by acquisition motion at the billing record level, not inferring it from a CRM field that may not be kept current.

### Usage-to-plan-tier fit rate

This is the percentage of accounts running meaningfully above or below what their current plan tier is built for; think accounts consistently above 80% of a usage allowance (upgrade-ready, and at risk of a poor experience if they hit a hard cap) or consistently below 20% (over-provisioned, and a contraction or churn risk at renewal). It's not a standard, universally defined metric the way MRR is, but growth teams running usage-based or tiered pricing increasingly build some version of it, because it's the leading indicator behind both expansion MRR and churn, weeks before either shows up in a billing report.

Building this requires usage data and plan/tier data living in the same place. If usage events sit in a product analytics tool and plan assignments sit in a billing processor, someone is stitching the two together manually, usually in a spreadsheet, usually not in real time. It's also the same signal that feeds early [churn risk detection](https://www.revvly.ai/blog/churn-product-signals): an account well below its usage allowance for several billing cycles running is often the first sign of a renewal at risk, weeks before a cancellation request shows up anywhere else.

### Time-to-expansion

The interval between an account's activation date and its first expansion event (upgrade, add-on, or usage overage that converts to a plan change). A shortening time-to-expansion across cohorts is one of the more reliable signals that a specific packaging or in-product prompt is working, more reliable than watching aggregate expansion MRR, which mixes cohorts of very different ages and can look flat even while newer cohorts are expanding faster.

## Why these numbers usually live in a spreadsheet instead of a dashboard

None of the five metrics above are exotic. What's hard is that each one requires joining data that typically lives in three different systems: product usage events (product analytics tool), plan and billing history (a billing processor), and account ownership or motion tagging (CRM). A product analytics platform can show usage. A billing processor can show what was charged. Neither one, on its own, can show whether a usage spike caused a plan change that generated durable expansion revenue.

The common workaround is a recurring manual export: pull usage data from the product analytics tool, pull billing data from the processor, join them in a spreadsheet, and rebuild the join every time someone wants an updated number. That process works until the company adds a second pricing model, a regional price point, or a bundle, at which point the spreadsheet's assumptions stop matching reality and nobody notices until the numbers stop reconciling with what finance reports separately. This is the same [fragmented tool stack problem](https://www.revvly.ai/blog/revenue-stack) that shows up across billing, revenue reporting, and churn analysis; PLG revenue metrics are just the version of it that hits product and growth teams first, because they're usually the ones building the spreadsheet.

## A checklist before you trust your PLG revenue numbers

- Is self-serve conversion rate weighted by revenue, or just counting logos?
- Can you isolate expansion MRR by the specific in-product trigger that caused it (usage overage, manual upgrade, add-on purchase)?
- Is NRR calculated separately for self-serve and sales-assisted accounts, or blended into one number?
- Is usage-to-plan-tier fit checked against the account's actual current plan, not a static usage threshold that ignores which tier they're on?
- Does time-to-expansion reset per cohort, or is it a single trailing average that hides recent changes?
- If two people on the team pulled these numbers independently this week, would they get the same answer?

If the honest answer to that last question is no, the metrics aren't wrong so much as unreliable, which in practice is the same problem.

## Turning revenue-side PLG metrics into decisions

The point of tracking these numbers isn't the dashboard, it's the decisions downstream of it. A rising usage-to-plan-tier fit rate at the top of a tier is a direct input into a packaging decision: add a higher tier, introduce a usage-based add-on, or adjust where the tier boundary sits. A shortening time-to-expansion after a specific onboarding change is evidence that change is worth keeping. A declining self-serve NRR next to a stable sales-assisted NRR points a retention team at exactly which motion needs a workflow, not a company-wide retention initiative.

This is where a platform that unifies subscription billing, modular pricing, and revenue reporting has a structural advantage over stitching a product analytics tool to a billing processor by hand. Revvly's modular pricing architecture lets a product or growth team add a usage-based tier or an add-on module without an engineering sprint, and because usage, plan, and billing data live in the same system, expansion MRR, NRR by segment, and usage-to-plan-tier fit come out of the same real-time dashboard rather than a rebuilt spreadsheet join. Revvly's AI-powered churn prediction also uses the same usage-versus-plan signal described above to flag at-risk accounts before a renewal, which is the same underlying data doing double duty for both expansion and retention decisions. Plans start at [$29 per month](https://www.revvly.ai/pricing) for Revvly's entry tier; module and usage-based pricing scale from there depending on which parts of the platform a company needs.

## Frequently asked questions

**What are product-led growth metrics?** Product-led growth metrics measure how effectively a product itself drives acquisition, activation, retention, and expansion without a sales-led motion. They split into engagement metrics (activation rate, product-qualified leads, feature adoption) and revenue metrics (expansion MRR, net revenue retention, self-serve conversion revenue), and most published lists lean heavily toward the engagement side.

**What is expansion MRR and how is it different from new MRR?** Expansion MRR is additional monthly recurring revenue generated by existing customers through upgrades, seat additions, or usage overages. New MRR comes from newly acquired customers. Tracking them separately matters because a company can look like it's growing on a blended MRR chart while new customer acquisition is actually slowing and expansion is doing all the work, or the reverse.

**How do you calculate self-serve to paid conversion rate?** Divide the number of accounts that convert from a free trial or freemium tier to a paid plan by the total number of accounts that started in that free tier, then multiply by 100. A more useful version weights each conversion by the plan value it converts into, so the metric reflects revenue momentum, not just account count.

**What is net revenue retention (NRR) and why does it need to be split by acquisition motion?** NRR measures revenue retained from existing customers, including expansion, minus contraction and churn, over a set period, expressed as a percentage of the starting revenue base. Splitting it by acquisition motion (self-serve versus sales-assisted) matters because blending both into one number can hide a real decline in one motion behind stability in the other.

**Can a product analytics tool alone track PLG revenue metrics?** Not reliably. Product analytics tools capture behavioral events well but generally don't hold billing history, plan tier data, or revenue amounts, so metrics like expansion MRR or NRR require joining that event data with a separate billing or revenue system.

**What causes PLG revenue metrics to look healthy in a dashboard but wrong in practice?** The most common cause is a manual join between a product analytics export and a billing export that goes stale as pricing models get more complex, for example after adding a usage-based tier or a regional price point that the spreadsheet's original logic didn't account for.

**How long does it take to set up revenue-side PLG tracking?** It depends on how many systems currently hold the underlying data. If usage, plan, and billing data already live in one platform, real-time dashboards for expansion MRR and NRR by motion can be live in days. If the data is spread across a product analytics tool, a billing processor, and a CRM, the first step is usually a one-time reconciliation to establish a reliable baseline before automating the join.

## Sources

1. Mixpanel, "Product-led growth in 2026: A complete guide (and the metrics that actually matter)," published June 9, 2026. [https://mixpanel.com/blog/product-led-growth/](https://mixpanel.com/blog/product-led-growth/)
2. First Page Sage, "SaaS Free Trial Conversion Rate Benchmarks," data aggregated Q1 2022 through Q3 2025 across 86 SaaS companies. [https://firstpagesage.com/seo-blog/saas-free-trial-conversion-rate-benchmarks/](https://firstpagesage.com/seo-blog/saas-free-trial-conversion-rate-benchmarks/)
3. Pavilion and Benchmarkit, "2025 B2B SaaS Benchmarks: CAC, NRR & Growth Rate Metrics." [https://www.joinpavilion.com/resource/b2b-saas-performance-benchmarks](https://www.joinpavilion.com/resource/b2b-saas-performance-benchmarks)

## Frequently asked questions

### What are product-led growth metrics?

Product-led growth metrics measure how effectively a product itself drives acquisition, activation, retention, and expansion without a sales-led motion. They split into engagement metrics (activation rate, product-qualified leads, feature adoption) and revenue metrics (expansion MRR, net revenue retention, self-serve conversion revenue), and most published lists lean heavily toward the engagement side.

### What is expansion MRR and how is it different from new MRR?

Expansion MRR is additional monthly recurring revenue generated by existing customers through upgrades, seat additions, or usage overages. New MRR comes from newly acquired customers. Tracking them separately matters because a company can look like it's growing on a blended MRR chart while new customer acquisition is actually slowing and expansion is doing all the work, or the reverse.

### How do you calculate self-serve to paid conversion rate?

Divide the number of accounts that convert from a free trial or freemium tier to a paid plan by the total number of accounts that started in that free tier, then multiply by 100. A more useful version weights each conversion by the plan value it converts into, so the metric reflects revenue momentum, not just account count.

### What is net revenue retention (NRR) and why does it need to be split by acquisition motion?

NRR measures revenue retained from existing customers, including expansion, minus contraction and churn, over a set period, expressed as a percentage of the starting revenue base. Splitting it by acquisition motion (self-serve versus sales-assisted) matters because blending both into one number can hide a real decline in one motion behind stability in the other.

### Can a product analytics tool alone track PLG revenue metrics?

Not reliably. Product analytics tools capture behavioral events well but generally don't hold billing history, plan tier data, or revenue amounts, so metrics like expansion MRR or NRR require joining that event data with a separate billing or revenue system.

### What causes PLG revenue metrics to look healthy in a dashboard but wrong in practice?

The most common cause is a manual join between a product analytics export and a billing export that goes stale as pricing models get more complex, for example after adding a usage-based tier or a regional price point that the spreadsheet's original logic didn't account for.

### How long does it take to set up revenue-side PLG tracking?

It depends on how many systems currently hold the underlying data. If usage, plan, and billing data already live in one platform, real-time dashboards for expansion MRR and NRR by motion can be live in days. If the data is spread across a product analytics tool, a billing processor, and a CRM, the first step is usually a one-time reconciliation to establish a reliable baseline before automating the join.

AR

Alain Readman

Co-founder, Revvly

Alain spent a decade building and scaling subscription businesses before founding Revvly. He writes about the infrastructure gap between what SaaS companies report and what they actually earn.

Revvly

Orchestrate the flow of revenue.

© 2026 AcceleratorApp Limited. All rights reserved.