The Right KPIs for Your Company's Stage (and the Ones You Can Safely Ignore)

A founder I talked to recently had a beautiful dashboard: 24 tiles covering MRR, CAC, LTV, NPS, churn, daily active users, burn multiple, pipeline coverage, and more. They had eleven paying customers.

Most of those tiles were noise. With eleven customers, "churn rate" swings 9 points every time one person cancels. LTV is a guess stacked on a guess. CAC averages a conference booth with two cold emails. Meanwhile the one thing they needed to know (do the people using this actually love it?) wasn't on the dashboard at all.

This is one of the most common analytics problems small companies have. It isn't a lack of data. It's tracking the right numbers for a different company: usually a later-stage one whose dashboard they saw in a blog post or investor deck. This guide walks through which KPIs fit each stage, why, and how to know when to switch.

First, what makes a KPI "right"?

A KPI (key performance indicator) is just a number you've agreed to watch closely because it tells you whether things are working. A good one passes three tests:

1. It answers the question your company is actually facing right now. Early on that question is "does anyone want this?" Later it becomes "can we grow this efficiently?" and then "can we grow it profitably at scale?"
2. You have enough data for it to mean something. A percentage calculated from eight events is a rumor, not a metric.
3. Someone can act on it this week. If the number moves and nobody would change what they do, it's a vanity metric.

Think of it like a doctor's checkup. A newborn gets weighed and measured; a 50-year-old gets cholesterol checked. Both are "health metrics," but swapping them would be pointless. Your company has life stages too.

Stage 1: Finding something people want (pre-product-market fit)

The question: Is there a group of people who get real, repeated value from this?

At this stage you have a handful of users and a product that changes weekly. Financial efficiency metrics are premature. What matters is evidence of pull.

KPIs that fit:

• Retention of early users. Of the people who signed up a month ago, how many are still using the product? A flattening retention curve, even at a low level, is the strongest signal you can get.
• Activation rate. What share of new signups reach the moment where they first get value: sending their first invoice, building their first report, inviting a teammate?
• Qualitative pull. A widely used gut-check is the "how would you feel if you could no longer use this product?" survey popularized by Sean Ellis. If a large share of active users say "very disappointed," you're onto something. Count it, but also read every answer.
• Number of active customers, as a raw count, not a percentage.

Safe to ignore for now: LTV, CAC, payback period, NPS benchmarks, burn multiple. You don't have stable enough data, and optimizing them now can push you to scale something that doesn't work yet.

Example: A two-person team building scheduling software for clinics stops looking at signup totals and starts tracking one thing: how many clinics booked at least 20 appointments through the tool in their second month. That number went from 2 to 9 over a quarter, and that told them more than any revenue figure.

Stage 2: Proving you can grow it (early traction)

The question: Now that some people love it, can we find more of them in a repeatable way?

You have dozens or a few hundred customers and some revenue. Now the numbers have enough volume to start meaning something, and the focus shifts from "is it good?" to "is growth repeatable?"

KPIs that fit:

• Revenue growth rate, usually monthly recurring revenue (MRR) growth month over month.
• Logo churn and revenue churn. Logo churn counts customers lost; revenue churn counts dollars lost. They can tell very different stories if your big customers behave differently from your small ones.
• Conversion through your main funnel: visitor to signup to activated to paying. You're looking for the step where most people drop off.
• Customer acquisition cost (CAC) by channel, kept rough. You want to know whether referrals, content, or outbound produce customers at wildly different costs, not to compute it to the penny.
• Sales cycle length if you sell to businesses.

Watch out for: blending channels. A single CAC number that averages a cheap referral channel with an expensive paid-ads channel hides the most useful decision you can make, which is where to spend the next dollar.

Stage 3: Scaling efficiently (growth)

The question: Can we pour money into growth and get a good return?

With hundreds or thousands of customers, a sales or marketing team, and probably outside investment, the financial efficiency metrics finally earn their place.

KPIs that fit:

• CAC payback period: how many months of gross margin it takes to earn back what you spent to acquire a customer. Shorter means you can reinvest faster.
• Net revenue retention (NRR): of the revenue you had from a group of customers a year ago, how much do you have today, counting upgrades, downgrades, and cancellations? Above 100% means existing customers grow on their own even if you sign nobody new.
• Gross margin: what's left after the direct cost of delivering the product.
• Burn multiple: how much cash you burn for each dollar of new recurring revenue you add. It's a blunt but popular way to ask "is our growth efficient?"
• LTV to CAC ratio: only now, when you have enough customer history for lifetime value to be more than a projection.

Example: A B2B SaaS company with 600 customers notices that its overall churn looks fine, but NRR in its small-business segment is well below its mid-market segment. That one split changes its pricing and who sales goes after.

Stage 4: Running a durable business (maturity)

The question: Is this a predictable, profitable machine?

At this point the focus shifts again: to predictability and profit. Common KPIs include operating margin, forecast accuracy (did the quarter land where you said it would?), revenue per employee, and segment-level profitability. Many teams also adopt a combined measure like the "Rule of 40" (growth rate plus profit margin) to balance growing fast against making money.

A quick reference

Stages aren't strict boundaries. Plenty of companies are "growth stage" in one product line and "pre-fit" in a new one, and that's fine. Pick KPIs per product or segment when that's the reality.

Common mistakes

Copying a later-stage dashboard. The most frequent mistake by far. An investor-friendly metrics page is designed for a company that has already answered your current question.

Tracking too many numbers. If everything is a KPI, nothing is. Three to five headline numbers per stage is plenty, with supporting detail one click away.

Percentages on tiny samples. "Churn went from 10% to 20%!" might mean one customer left instead of zero. Show the raw count next to any percentage while numbers are small.

Never retiring a KPI. Metrics that mattered at 20 customers can quietly keep eating meeting time at 2,000. Review your list every quarter or after any big milestone, like a funding round or a pricing change.

Defining things differently in different places. If sales counts an "active customer" one way and finance counts it another, the dashboard turns into an argument. Write a one-line definition for each KPI and keep it next to the number.

Gaming the metric. Once a number becomes a target, people find ways to move it that don't help the business. This is often called Goodhart's law. Pair each KPI with a "guardrail" number, e.g. watch activation rate and 30-day retention, so pushing one doesn't quietly hurt the other.

Putting it into practice

You don't need a data team to do this well. A simple loop works:

1. Write down the single biggest question your company faces this quarter.
2. Pick the 3–5 numbers that best answer it, with a plain-English definition for each.
3. Put them somewhere everyone sees weekly: a shared dashboard, not a spreadsheet someone rebuilds every Monday.
4. Revisit the list when the question changes.

For step 3, most teams either use a BI tool connected straight to their database or build a small internal page. Tools like Draxlr are one option for turning queries on an existing SQL database into a shared dashboard. Whichever route you pick, the hard part is choosing the numbers, not displaying them.

Key takeaways

• The right KPIs depend on the question your company is facing right now, not on what bigger companies track.
• Early on, measure pull: retention, activation, and what users tell you. Later, measure efficiency: payback, NRR, margin.
• Keep the list short, define each number clearly, and show raw counts next to percentages when samples are small.
• Retire KPIs deliberately as you move between stages.

What's on your team's dashboard right now, and is any of it there because "that's what you're supposed to track"? I'd love to hear which metrics you've dropped or added as your company has grown. Share in the comments.