SaaS Metrics That Actually Matter: A Hierarchy from North Star to Diagnostic

I've been in the SaaS analytics trenches long enough to recognize the pattern. Founders fire up their spreadsheets, dump 30 metrics into different tabs, set them to "automatic update mode," and then... never look at them again. Or worse, they stare at all 30 every Sunday night and have no idea which one actually matters.

The problem isn't that you're tracking too many things. It's that you're treating all metrics as if they're equal. They're not.

Some metrics are outputs—things that happen as a result of your business working. You can't directly control them. Other metrics are inputs—things your team can actually influence. And some metrics are diagnostics—they tell you why your output metrics moved, even if you can't directly change them.

When you understand the hierarchy, everything becomes clearer. You'll know which metric to obsess over (one), which metrics to watch as leading indicators (a few), and which ones to dig into only when something weird happens (the rest).

Let's build that hierarchy.

Your north star metric (and why you need exactly one)

A north star metric is the single measurement that best represents the value your product delivers to your customers. It's your business in one number.

The trap here is thinking your north star has to be revenue. For some SaaS companies, it should be. For others, it's something entirely different.

A collaboration tool's north star might be weekly active users because the core value is "people work together." An analytics tool's north star might be reports generated per week because the value is "teams get insights." A payment processing platform's north star might be transaction volume because that's where the real customer value lives.

Revenue is usually a lag indicator—it arrives after everything else. Your north star should be something that moves before your MRR does. It's the leading edge of your business.

Here's what matters: your north star should be something that:

1. Your users directly experience. If your north star metric went to zero tomorrow, users would notice immediately. If they wouldn't, it's not a north star—it's a secondary metric.

2. Moves before revenue. You should see changes in your north star before you see them in your bank account.

3. Connects to retention and expansion. Companies with strong north star metrics (users engaging daily, reports being generated, features being adopted) stick around longer and pay more. There's a measurable relationship.

4. You can actually influence. Your product team, marketing team, and customer success team should all be able to move this metric through their work.

Once you pick your north star, stop adding to it. Don't have a north star and a north star backup. Don't track three different versions of the same concept. One. That's the point.

The revenue metrics layer

Below your north star sits revenue. This is where most SaaS companies start paying attention—and for good reason. Revenue doesn't lie.

The core revenue metrics you need are straightforward:

Monthly Recurring Revenue (MRR) is your baseline. It's the predictable revenue you expect in a given month from active subscriptions. Calculate it by multiplying your average subscription price by your total paying customers, or sum all active subscription contracts for the month.

Annual Recurring Revenue (ARR) is MRR multiplied by 12. People obsess over ARR because it's the big number that investors care about. It's also the easiest way to compare your company to benchmarks and industry peers.

Expansion MRR (also called net new MRR from existing customers) is the revenue increase from existing customers. This is where the magic of SaaS lives. The best SaaS companies grow from expansion alone—new customer acquisition is bonus.

Contraction MRR (also called negative churn) is the flip side: revenue lost when existing customers downgrade, reduce seats, or cancel. You want this to be small and shrinking.

Here's what healthy growth actually looks like. According to OpenView Partners' SaaS benchmarks, median MRR growth rates vary wildly by ARR stage:

If you're at $1M ARR growing at 2% per month, you should be nervous. If you're at $10M growing at 4% per month, you're doing well. Context matters.

The dangerous move is bundling new customer revenue and expansion revenue together. "We grew MRR 8% this month" tells you nothing until you know whether it came from six new $5K customers or from upselling existing customers. The first scenario means you're good at acquisition. The second means you're good at retention and expansion. Those require completely different plays.

Split them. Always. The formula is:

Net MRR Growth = (New MRR from new customers) + (Expansion MRR from existing customers) – (Contraction MRR)

The retention metrics layer

Here's the uncomfortable truth: acquisition and expansion don't matter if your customers leave.

Retention comes in two flavors: logo churn and revenue churn. They're measuring different things, and both matter.

Logo churn is the percentage of customers you lose in a period. If you had 100 customers at the start of the month and 95 at the end, you had 5% logo churn. Simple math, but it hides a lot of variation. You might have lost a small customer and a giant customer—those are very different situations.

Revenue churn (or dollar churn) is the percentage of MRR lost to cancellations. If you started with $100K MRR and ended with $98K, you had 2% revenue churn. This is more important than logo churn because it reflects what matters: your ability to keep making money.

Some companies have 10% logo churn but only 2% revenue churn because the customers they're losing are small. Some have 2% logo churn but 8% revenue churn because they're losing their biggest customers.

The benchmark here is crucial: according to Bessemer Venture Partners' State of the Cloud report, the median revenue churn for top-quartile SaaS companies is basically zero. The top performers have negative revenue churn—meaning expansion revenue exceeds contraction.

But here's the thing about churn benchmarks: they're almost useless for your business. What matters is your churn and whether it's improving. A 5% monthly churn rate is great if you're a $5K ACV SMB product and terrible if you're an enterprise platform.

Net Revenue Retention (NRR) is where the real magic is. NRR measures the percentage of revenue from existing customers in one period that's still there (or higher) in the next period.

The formula looks like this:

NRR = (Revenue from existing customers at end of period – revenue lost to churn + expansion revenue) / Revenue from existing customers at start of period

If you start the year with $1M ARR from existing customers and end with $1.15M (after accounting for churn and expansion), your NRR is 115%.

According to Bessemer's research, companies with NRR above 120% are in the top quartile. An NRR of 110% puts you in the top half. An NRR below 100% means you're in danger—expansion isn't covering your churn.

NRR is the single best predictor of long-term SaaS success. Companies with strong NRR don't need as much sales efficiency to grow exponentially. Companies with weak NRR are on a treadmill.

Why? Because NRR compounds. Let's say you have $1M ARR and 110% NRR. That means next year, that $1M becomes $1.1M. The year after, it becomes $1.21M. Even with zero new customer acquisition, your revenue grows. That's the power of a strong retention metric.

The acquisition metrics layer

Now we're at the engine that drives new growth. The acquisition metrics form their own hierarchy.

Customer Acquisition Cost (CAC) is the blunt force number: how much money do you spend (on sales and marketing) to acquire one customer?

The formula is straightforward:

CAC = (Sales + Marketing spend in period) / (New customers acquired in period)

If you spent $100K on sales and marketing last month and acquired 10 customers, your CAC is $10K.

CAC in isolation is useless. A $10K CAC might be incredible for an enterprise software company and terrible for a SMB product. Context is everything.

LTV:CAC ratio is the classic framework. LTV is Lifetime Value—the total profit you expect from one customer over their lifetime. The rule of thumb, cited across industry benchmarks, is that a healthy LTV:CAC ratio is 3:1 or higher. Some high-growth SaaS companies target 5:1 or more.

But here's the honest problem with LTV:CAC: calculating true LTV requires assumptions about churn, expansion, and costs that you'll probably get wrong. It's useful as a direction indicator ("we're moving the right way") but risky as a hard rule.

CAC payback period is operationally more useful. It answers a simpler question: how many months until we make back what we spent to acquire this customer?

The formula is:

CAC payback period = CAC / (Average monthly profit per customer)

If your CAC is $5K and each customer generates $500/month in gross profit, your payback period is 10 months.

According to OpenView Partners' benchmarks, the median CAC payback period for SaaS companies is between 6–12 months depending on the sales model. Self-serve products see faster payback (3–6 months). Enterprise products with long sales cycles see slower payback (12–18 months).

Early-stage founders usually optimize for CAC payback period because it answers the real question: "Can we afford this?" A 3-month payback is great. A 24-month payback is a problem, especially if you don't have the cash runway to wait.

The engagement metrics that predict churn

Retention and acquisition metrics tell you what happened. Engagement metrics tell you what's about to happen.

These are your leading indicators—the early warning system.

Daily Active Users (DAU) and Monthly Active Users (MAU) are the foundation. DAU is the number of unique users who do something meaningful in your product on a given day. MAU is the same thing stretched to 30 days.

The ratio DAU/MAU tells you about stickiness. If your DAU/MAU ratio is 50%, that means half your monthly active users show up on any given day. A ratio of 30% means it's skewing weekly. A ratio of 10% means your users come back once or twice a month.

For collaboration tools, DAU/MAU is critical. For reporting tools, it's less critical because users might only need to check dashboards weekly. Know what's normal for your category.

Feature adoption rate is the percentage of your customers (or users) who have used a specific feature. If you just shipped an upsell feature and adoption is stuck at 15%, you've got a problem. If adoption jumps to 75% in two weeks, you shipped something people want.

Time to Value (TTV) is how long it takes a new user to get their first tangible benefit from your product. If new users take 30 days to extract value, you'll have high churn. If they extract value in 30 minutes, they'll stick around.

Here's the magic: these three metrics predict churn before it happens. Users with low DAU, poor feature adoption, and slow TTV are on the way out. You can intervene—better onboarding, better product education, support outreach—before they become a churn statistic.

These metrics are why activation is so critical in SaaS. A new customer who reaches a meaningful activation moment in their first week is dramatically more likely to renew than one who doesn't. Track when that activation happens and make it your obsession.

The product-led growth (PLG) metrics

If your business model is product-led (users sign up for free and convert to paid), you need a different lens on these same metrics.

Product Qualified Lead (PQL) is the PLG version of a "sales qualified lead." It's a free user who has demonstrated enough product value and/or feature usage that they're highly likely to convert to paid. The definition varies wildly:

For a code editor, a PQL might be a free user who's created 3+ projects. For an analytics tool, a PQL might be someone who's built 2 reports and shared them. For a note-taking app, a PQL might be a user with 50+ notes.

The critical thing: define this in your product, actually measure it, and use it to trigger conversion moments. "You've hit your limit" is fine. "You've done enough that you're ready to upgrade" is better.

Free-to-paid conversion rate is the percentage of free users who convert to a paid plan. Industry-wide, this ranges from 2–10% depending on the business model, price point, and positioning. If your free-to-paid conversion is 0.5%, something's broken in your PLG funnel. If it's 15%, you're doing something right.

Viral coefficient measures how many new users each existing user brings in. For a B2C SaaS product (think Slack, Figma), a viral coefficient above 1.0 is magic—it means each user brings in more than one new user, and you have a growth flywheel. For B2B SaaS, coefficients are typically lower (0.2–0.5 is normal) because adoption is more intentional.

PLG isn't for every SaaS company. It works brilliantly for products with low barrier to trial, short time to value, and strong network effects. It's harder for enterprise software where the sales process is required. Know which model you're playing.

How to measure SaaS metrics with GA4

Here's where theory meets practice.

Google Analytics 4 is a product analytics tool, but with some smart instrumentation, it can give you visibility into the metrics above. The key is treating GA4 as part of your analytics stack, not your entire stack.

Activation tracking starts with defining what "activation" means for your product. In GA4, you'd create a custom event when a user hits that moment. For an analytics tool, activation might be firing when a user completes their first report. For a collaboration tool, it might be when two users interact on the same document.

Set up this event in your code (using the GA4 Measurement Protocol or Google Tag Manager), then use GA4 segments to see how activation rates change over time. Compare "users who activated within 7 days" vs. "users who took 30+ days" and measure which group has better retention.

Funnel analysis in GA4 can track the trial-to-paid journey. Create a funnel that shows: Sign Up → Activation → First Key Action → Trial Conversion → Paid Activation. Each step shows you where users drop off. If 60% sign up but only 15% convert to paid, something's wrong in that middle section.

Custom dimensions let you layer subscription tier, customer segment, or acquisition channel into every single event. This is powerful. You can ask: "Do enterprise users have different activation rates than SMB users?" "Do users acquired through organic search have better retention than users from ads?" GA4 lets you answer these questions if you've set it up correctly.

Limitations to know: GA4 is web analytics. It sees sessions and events. Your billing system is separate. GA4 doesn't know which customers are paying vs. free unless you explicitly connect that data. You need to sync your billing data (Stripe, Zuora, whatever) with GA4 using custom dimensions or a data warehouse. Without that connection, GA4 will tell you "10,000 users visited our pricing page" without telling you "and 500 of them paid."

This is why many SaaS companies use GA4 and a dedicated product analytics platform and a data warehouse. GA4 handles web analytics and event tracking. Product analytics handles feature usage and user cohorts. Data warehouse brings it together.

Where Emilytics fits for SaaS teams

Here's the practical reality: GA4 is a tool. It's powerful, but it requires you to ask the right questions.

Most SaaS founders and product teams spend half their time building complex Explorations in GA4 to answer a simple question: "What's our activation rate this month?" They click through menus, adjust filters, set up segments, and 20 minutes later they have an answer.

Then next month, something changes. They need to rebuild the exploration. The filter logic wasn't documented. A new teammate doesn't know how it was built. You lose continuity.

This is where connecting GA4 with a natural language analytics layer changes the game. Imagine asking "What's our activation rate this month broken down by acquisition channel?" and getting an instant answer. Or "Show me weekly active user trends by customer segment." Or "Which features correlate with better retention?"

You get to ask questions in plain language instead of building complex queries. The system learns your KPI definitions—your north star metric, your activation threshold, your ideal customer profile—and surfaces answers instantly.

For teams managing multiple metrics hierarchies (maybe you care about weekly active users and NRR and CAC payback period), this becomes critical. You can't manually track 30 things. You need a system that understands what matters.

FAQ

Q: What if I'm an early-stage company with only 50 paying customers? Do these metrics matter yet?

A: Absolutely. You won't have the volume to smooth out noise, so monthly churn will look volatile. But you need to care about it immediately. Understanding whether you have 2% or 10% monthly churn at your early stage determines whether you're building a sustainable business or not. The benchmarks don't apply (your company is too small), but the behavior—tracking it, acting on it—does.

Q: Should I optimize for growth or retention first?

A: Retention first. If your retention is broken, growth just means more water leaking out of a broken bucket. Getting to a place where you have positive unit economics and reasonable retention (20% monthly churn or better for SaaS) is table stakes. Then you can invest in growth. The math: if you have 5% monthly churn and a 3-month CAC payback period, every customer generates 40 months of revenue on average. That's profitable. If you have 15% monthly churn, you're lucky to get 8 months. You're on a treadmill.

Q: How often should I be looking at these metrics?

A: North star metric: weekly. Revenue metrics: monthly (after it settles—you might have invoicing delays). Retention and churn: monthly (they're too noisy to look at weekly). Acquisition metrics: weekly if you're actively optimizing acquisition; monthly otherwise. Engagement metrics: weekly. Think of it like a dashboard hierarchy: glance at weekly metrics every few days, deep-dive on monthly metrics when the month closes, and quarterly metrics when you're doing strategic planning.

Q: Can I just use Stripe to measure everything?

A: No, and this is a common trap. Stripe tells you about revenue and subscriptions, but it doesn't tell you about activation, engagement, feature adoption, or retention. Stripe is the source of truth for MRR and ARR. But to understand why your MRR moved, you need GA4 and product analytics. They're complementary systems.

I work with SaaS teams who are drowning in data but starving for answers. The companies that win aren't the ones tracking the most metrics. They're the ones who've built clarity around their metrics hierarchy, who know which metrics are outputs and which are inputs, and who've set up systems to monitor them continuously.

That's what we do at Emilytics: we connect your data sources (GA4, product events, billing systems, revenue data), we help you define your north star and the metrics that predict it, and then we let you ask questions in natural language instead of building manual reports.

Your SaaS metrics should tell a coherent story. This guide is the framework for making that happen.