Startup KPIs: The Essential Metrics for Each Growth Stage (2026)

Many startups measure the wrong things at the wrong time, and end up making disastrous decisions as a result. KPIs (Key Performance Indicators) are your startup's instrument panel. They tell you whether you're accelerating in the right direction or heading into a wall at full speed. The metrics that matter at pre-seed are different from the ones that matter at Series A. A business angel looks at your user engagement. A Series A VC wants to see your LTV/CAC ratio and your unit economics. This guide maps out which KPIs to track at each funding stage, along with the benchmarks French investors use to evaluate your company.

What is a KPI for a startup?

A KPI is a number that triggers an action. A figure you glance at passively in a Google Sheet on Friday evening is just a number. If it goes up or down and you change nothing, it isn't steering anything.

For a startup, KPIs serve three purposes: validating your business assumptions, convincing investors you're on the right trajectory, and aligning your team around the same priorities. Everything else is noise.

KPI vs metric: what's the difference?

People mix these up all the time. The number of visitors to your website is a metric. The rate at which those visitors convert into active users is a KPI. The difference? You tie a KPI to a business goal and act on it.

In practice:

Vanity metrics are the classic trap. 50,000 LinkedIn followers. 10,000 app downloads. These numbers flatter the ego but say nothing about the health of your business. A seasoned investor ignores them right away.

Why KPIs change by stage

Because your startup is a different company at pre-seed than at Series A. At pre-seed, you're looking for a problem worth solving. At seed, you're validating that your solution works and that people will pay for it. At Series A, you're proving your model can scale.

Tracking ARR before you have any customers is like measuring the top speed of a car with no engine. Tracking only user engagement while raising €5M means showing up to an investor meeting without the numbers they expect.

The rule is simple: your KPIs should reflect the question you're currently trying to answer. And that question changes every 12 to 18 months.

Universal KPIs (every stage)

Some indicators span every stage. They won't always be your top priority, but any investor will expect you to know them, whatever your maturity. These five come up every time.

MRR / ARR (Monthly & Annual Recurring Revenue)

MRR (Monthly Recurring Revenue) is your recurring revenue per month. ARR (Annual Recurring Revenue) is MRR multiplied by 12. For a SaaS company, this is the pulse of the business. When it rises, you're growing. When it stalls, you have a problem. When it drops, you're in danger.

Formula:

• MRR = sum of all active subscriptions for the month
• ARR = MRR x 12

The most widely cited benchmark comes from Paul Graham: in Startup = Growth, he puts good weekly growth during Y Combinator at 5 to 7%, with 10% a week being exceptional and 1% a sign you haven't yet found your market. If you're below that range, find the cause before your next round.

One thing founders often forget: break down your MRR. New MRR (new customers), Expansion MRR (upsell, cross-sell), Contraction MRR (downgrades) and Churned MRR (lost customers) tell a much richer story than the headline number. Your MRR can grow 8% a month while you lose 30% of your existing customers, simply because new customers make up the difference. Your base is melting while the top-line number climbs.

CAC (Customer Acquisition Cost)

CAC measures how much you spend to acquire one customer. Without it, you're burning cash without understanding why.

Formula:

• CAC = (marketing spend + sales spend) / number of new customers acquired over the period

Watch out for calculation traps. Do you include your sales team's salaries? Your marketing tools? The time the founder spends running demos? An "official" CAC that leaves out 60% of real costs is accounting fiction. Investors know it.

The standard benchmark: a CAC payback period (the time it takes to recoup the cost of acquiring a customer) under 12 months is considered good for B2B SaaS. Beyond 18 months, you're financing the acquisition of customers who haven't paid for themselves yet.

To dig deeper into cost-efficient acquisition strategies, read our guide to growth marketing.

LTV (Lifetime Value) and the LTV/CAC ratio

LTV (Lifetime Value, or customer lifetime value) estimates the total revenue a customer generates over their entire relationship with your company. It's the counterpart to CAC: how much a customer brings in versus how much they cost you.

Simplified formula:

• LTV = ARPA (average revenue per account) x average customer lifetime
• Or for SaaS: LTV = ARPA / monthly churn rate

The LTV/CAC ratio may be the KPI VCs scrutinize most. The industry's golden rule: a minimum ratio of 3:1. Every euro spent on acquisition should bring in at least 3 euros of revenue over the customer's lifetime. High-growth startups aim for 3 to 5x.

Below 3:1, your business model doesn't hold up. Either you're losing money on every customer you acquire, or you aren't earning enough to fund your growth. Above 5:1, it may be a sign you're underinvesting in acquisition and leaving market share to your competitors.

Churn Rate

Churn is the percentage of customers you lose over a given period. It's the hole in the bucket. You can pour in as much water as you like (acquisition), but if the bucket leaks (churn), you'll never fill it.

Formula:

• Monthly churn = (customers lost during the month / customers at the start of the month) x 100

B2B SaaS benchmarks in 2026:

• Mid-market / Enterprise: monthly churn below 1% (under 12% annually)
• SMB (small businesses): 5 to 7% annual churn for the best performers, which is an ambitious target
• Early-stage: annual churn of 10 to 15% in year one is common and shouldn't alarm you too much, as long as it falls quarter after quarter

Net revenue churn (which factors in expansion revenue) is even more telling. If your existing customers spend more over time (upsells, added seats), you can reach negative churn: your remaining customers bring in more than you lose from those who left.

Runway and burn rate

Runway is the number of months you have left before you can no longer make payroll. Burn rate is how fast you consume your cash each month.

Formulas:

• Monthly burn rate = total monthly expenses - monthly revenue
• Runway = available cash / monthly burn rate

A runway of 12 to 18 months gives you the room to negotiate calmly with investors. Below 6 months, you're negotiating from weakness, and VCs can sense it.

Practical tip: calculate your runway every week, not every month. Bad surprises arrive fast when salaries, servers and SaaS subscriptions keep piling up.

KPIs by growth stage

A 3-month-old startup and a 50-person scale-up shouldn't be measuring the same things. The metrics below follow each stage, along with what French investors prioritize.

Pre-seed / Ideation (0 to 6 months): through a business angel's eyes

You have no revenue. Maybe no product. What you do have is a hypothesis and (if you've done the work) some early validation signals. Business angels in France don't expect to see MRR at pre-seed. They're looking for something else.

Indicators to track:

1. Number of discovery interviews completed. Did you talk to 5 prospects or 50? The difference is fundamental. A founder who has run 50 problem interviews understands their market. A founder who has run 5 is speculating.

2. Addressable market size (TAM/SAM/SOM). Angels want to know whether the problem you're solving affects 500 people or 500,000. The lean canvas is a valuable tool for structuring this thinking.

3. Waitlist or beta sign-ups. 200 people signing up on a landing page with no product is a signal. 2,000 is a strong signal. The landing page conversion rate (visitors to sign-ups) says more than the raw number.

4. Qualitative engagement. Do your first users come back without being nudged? Do they send feedback unprompted? You won't find these signals on a dashboard, but you'll feel them.

5. Team composition. Strictly speaking this isn't a KPI, but angels back founders first and metrics second. Having a technical profile on the team changes everything. If you're looking for a technical co-founder, read our guide to finding a CTO.

What angels ignore at this stage: CAC, LTV, MRR, churn. You don't have enough data for these numbers to be statistically meaningful.

Seed (6 months to 2 years): proving traction

You have a product. A few first customers. Maybe a few thousand euros in MRR. The key word for this phase is "traction." Seed investors want to see that something is working and that it's accelerating.

Priority KPIs:

1. MRR and month-over-month MRR growth. Paul Graham's benchmark of 5 to 7% growth per week remains the reference point for seed investors. It's ambitious. A slower but steady trajectory is defensible. Flat growth makes investors start questioning your product-market fit.

2. Activation rate. What percentage of your sign-ups reach the "aha moment" (the point where they grasp your product's value)? For Slack, it was a team sending 2,000 messages. For Dropbox, it was storing a first file. Identify your aha moment and measure how many users reach it.

3. 30/60/90-day retention. This is the product-market fit test. If 40% of your users are still active after 90 days, you're onto something. If it's 10%, you have a product people try and then abandon.

4. LTV/CAC ratio (early estimates). Even with limited data, start calculating. A ratio around 3:1 is reassuring. Investors know your numbers will move, but they want to see that you're measuring and that the trend is heading the right way.

5. Net revenue retention. Are your existing customers spending more over time? An NRR above 100% means your customer base generates more revenue month after month, even without new customers. Above 120%, you're in the top tier.

The classic seed mistake: focusing solely on acquisition and ignoring retention. Acquiring 100 customers a month is pointless if you lose 80. Plug the leaks before you turn on the tap.

Series A (2 years and beyond): demonstrating scalability

At Series A, the rules change. VC funds (Partech, Elaia and Breega, to name a few French players) have moved past looking for traction. They're looking for scalability. The question shifts from "does it work?" to "can it get 10x bigger?"

What VCs require:

1. ARR and a path to the million. Series A funds expect ARR between €500K and €1.5M (or a clear path to get there within 12 months). Expected annual growth: at least 2 to 3x.

2. Validated unit economics. Your LTV/CAC ratio should sit solidly above 3:1, ideally between 3 and 5x. CAC payback should be under 12 months. These are now verified figures based on 6 to 12 months of data, well beyond estimates.

3. Rule of 40. This metric combines growth and profitability: annual growth rate + net profit margin. The result should exceed 40%. A startup growing 80% a year with a -30% margin scores 50%: it passes. A startup growing 20% with a -25% margin scores -5%: it has a problem.

4. Gross margin. For SaaS, investors expect gross margin above 70%. Below that, your model lacks the characteristics of scalable SaaS. Your cost to serve (support, infrastructure, onboarding) is too high relative to revenue.

5. Sales efficiency. How much revenue does each euro invested in sales and marketing generate? The Magic Number (covered in detail below) is the reference indicator.

SaaS-specific KPIs

If you're building a SaaS product, three additional indicators deserve your attention. These engagement and activation metrics are also the backbone of a product-led growth strategy, where the product itself drives acquisition.

NPS (Net Promoter Score)

NPS measures how likely your customers are to recommend you. It's a single question: "On a scale of 0 to 10, how likely are you to recommend [product] to a colleague?" Scores of 9-10 are promoters. 7-8 are passives. 0-6 are detractors.

Formula: NPS = % promoters - % detractors

An NPS above 50 is considered excellent in B2B SaaS. Above 70, you're in the category of products users love. Below 0, you have a major satisfaction problem.

NPS is imperfect (it doesn't tell you why people do or don't recommend you), but it has one advantage: it's a reliable proxy for word of mouth, the cheapest acquisition channel there is.

DAU/MAU ratio (engagement)

The DAU/MAU ratio (Daily Active Users / Monthly Active Users) measures how often people use your product. The higher the ratio, the more your product is part of your users' daily routine.

Benchmarks:

• 20% and above: good for a B2B tool (users open it roughly 1 day in 5)
• 50% and above: excellent, your product is a daily tool
• Below 10%: your users don't come back often enough, which signals high churn ahead

For B2B tools that aren't designed for daily use (CRM, invoicing tools, etc.), a lower ratio is acceptable. What matters is that the ratio matches your product's natural usage frequency.

Magic Number (sales efficiency)

The Magic Number measures how efficiently your sales and marketing spend turns into growth. It tells you how well your growth engine performs.

Formula:

• Magic Number = (current quarter MRR - previous quarter MRR) x 4 / previous quarter sales & marketing spend

How to read it:

• Above 0.7: your growth engine is efficient, invest more
• Between 0.5 and 0.7: decent, optimize before scaling
• Below 0.5: you're burning cash faster than you're generating revenue, so rethink your acquisition strategy

The Magic Number is especially useful at Series A, when investors are assessing your ability to turn capital into growth. A high Magic Number says: "Give me more budget and I'll bring you more revenue." That's exactly what a VC wants to hear.

How to choose your KPIs (the OMTM method)

You now know about twenty KPIs. Good news: you don't need to track all of them. Bad news: you'll probably try anyway.

The classic mistake: tracking too many metrics

A dashboard with 47 metrics is a dashboard nobody looks at. When everything is a priority, nothing is. Everyone heads off in a different direction. The CEO watches MRR, the CTO watches load time, and the Head of Sales watches the pipeline. Nobody is looking at the same thing. The result: hour-long meetings where everyone defends their own numbers and nobody makes a decision.

The rule of thumb: at each stage, pick 3 to 5 KPIs at most. And among those 3 to 5, designate the one that comes first. This approach has a name: the OMTM method.

One Metric That Matters (OMTM): definition and application

The OMTM (One Metric That Matters) concept comes from Alistair Croll and Benjamin Yoskovitz in their book Lean Analytics. The idea is simple: at any given moment, one metric sums up the health and trajectory of your startup better than all the others. That's the one the whole team needs to know, understand and work to improve.

You still track others, but you align your decisions around that one.

How to choose your OMTM:

1. Identify your biggest growth bottleneck. Struggling to acquire users? Your OMTM is CAC or conversion rate. Users leaving after 2 weeks? Your OMTM is 30-day retention. Can't monetize? Track freemium-to-paid conversion.

2. Make sure you can act on it. "The total SaaS market in France" can't be steered. "Our onboarding conversion rate" can: you can change the onboarding and measure the impact.

3. Make sure the whole team can influence it. If only the founder understands the metric, it stays a personal spreadsheet.

4. Change it when the bottleneck changes. Your OMTM isn't set in stone. When retention goes from 30% to 60%, it's time to switch to acquisition or monetization.

Table: priority KPI by business type (marketplace, SaaS, D2C)

Depending on your business model, the relevant KPIs change completely. Recommended OMTMs by business type and stage:

Treat this table as a starting point. Your OMTM depends on your specific context: your stage, your model, your main acquisition channel and the growth bottleneck you're trying to remove.

How to track your KPIs day to day

Knowing what to measure is useless if you don't know how to measure it. Discipline matters more than the tool, though the right tool saves you time.

Free tools (Google Sheets, Notion)

For an early-stage startup (pre-seed, early seed), a well-structured Google Sheet is enough.

Google Sheets:

• Create one tab per month with your 3 to 5 KPIs
• Add a trend line (built-in sparklines)
• Share the file with the whole team
• Update it every Monday morning, rather than "whenever I have time"

Notion:

• Notion databases let you build a simple dashboard with filtered views
• The upside: it integrates with your other docs (meeting notes, roadmap, OKRs)
• The limitation: no advanced data visualization, no complex calculations

The bottom line: the tool doesn't matter. What matters is reviewing your KPIs every week, as a team, and turning them into decisions. A startup that spends 15 minutes every Monday analyzing its 3 KPIs consistently beats one with a paid dashboard nobody opens.

Paid tools (Mixpanel, Baremetrics, ChartMogul)

Once your data volume outgrows what Google Sheets can handle (typically beyond 50 to 100 paying customers), specialized tools take over.

Mixpanel: product analytics. User event tracking, conversion funnels, cohort retention. The free plan (up to 20 million events per month) is more than enough at seed stage. It's the go-to tool for understanding how users interact with your product.

Baremetrics: built for SaaS. Connects directly to Stripe and shows your MRR, churn, LTV and ARR in real time. The upside: zero setup, with data flowing in automatically from your payment tool.

ChartMogul: a Baremetrics competitor with a more analytical interface. Advanced segmentation by plan, cohort and geography. Especially useful when you have several pricing tiers and want to understand which segment performs best.

Amplitude: an alternative to Mixpanel, often preferred by product teams for cohort analysis and user journeys.

The advice: don't switch to a paid tool too early. As long as you can keep your tracking in a spreadsheet with 30 minutes of work per week, that's enough. Move to a paid tool when your data volume makes manual tracking time-consuming or unreliable.

Simple dashboard template

Here's the structure of a minimalist seed-stage dashboard:

Three critical columns:

1. Trend: is it moving in the right direction?
2. Target: what are you measuring yourself against?
3. Status: are you on track?

Make this table impossible to miss: put it on the first slide of your weekly meeting, or even make it the CEO's desktop wallpaper if needed. What isn't visible doesn't get measured.

FAQ

How many KPIs should an early-stage startup track?

Between 3 and 5, no more. At pre-seed, focus on validation signals: qualitative engagement, beta sign-ups, number of interviews completed. At seed, shift to traction metrics: MRR, retention, activation. The most common mistake is measuring too much and doing nothing with it. A founder who knows 3 numbers by heart and moves them every week progresses faster than one with a 40-metric dashboard who doesn't know which one to optimize.

Which KPIs should you present to a VC?

It depends on your stage. At seed: show your MRR, monthly MRR growth, 30/60/90-day retention and your early unit economics (CAC, LTV/CAC). At Series A: add ARR, net revenue retention, the Rule of 40, gross margin and the Magic Number. In every case, show trends over 6 to 12 months rather than a snapshot. An investor wants to see a trajectory. And be transparent about your weak spots: high churn that you've identified and are working to reduce looks far better than churn swept under the rug.

How do you calculate CAC correctly?

The basic formula: (marketing spend + sales spend) / number of new customers over the period. Everything hinges on what you include in "marketing and sales spend." The honest version includes marketing team salaries, sales team salaries, tools (CRM, ads, automation), ad budgets and the founder's time spent selling. The dishonest version includes only ad budgets. Guess which one investors prefer. Also calculate your CAC by channel (organic, paid, outbound, referral) to understand where your most profitable customers come from. A blended CAC of €200 can hide an organic CAC of €30 and a paid CAC of €500. Those are two very different realities.

MRR vs ARR: what's the difference?

MRR (Monthly Recurring Revenue) is your monthly recurring revenue. ARR (Annual Recurring Revenue) is MRR multiplied by 12. MRR is more useful day to day for tracking your month-by-month trajectory. ARR is the figure you share with investors because it gives an annualized view of your business's size. An MRR of €42,000 sounds less impressive than an ARR of €504,000, even though it's exactly the same thing. One caveat: if you have prepaid annual contracts, your ARR and your MRR x 12 can diverge. In that case, calculate ARR from active contracts rather than from the month's MRR. Always state your calculation method when you share these numbers.

swanbase supports early-stage founders for the long haul, in Paris and remotely. It costs €0 and is application-based: swanbase takes equity.

Recommended articles

Product market fit: definition, measurement, and how to reach it

A clear definition of product market fit, the exact measurement protocol (the Sean Ellis test), the Superhuman example, the limits of the 40% threshold, and concrete steps to reach PMF.

Growth marketing for startups: definition, AARRR and first experiments

What growth marketing means for an early-stage startup, how it compares to traditional marketing and growth hacking, the six-stage AARRR framework, the metrics to track and sourced examples from Dropbox to Respire.

Lean Canvas: The Practical Guide to Structuring Your Startup

The Lean Canvas fits on a single page, and you can map out your business assumptions on it in 20 minutes. Ash Maurya created it in 2010 to replace the business plan with 9 blocks that you test and revise week after week.

A Simple Guide to Lean Startup: From Definition to Practice

Lean Startup from definition to practice: Eric Ries's method for shortening product development cycles with an MVP

SEO for Startups with Amandine Bart

Recap of Amandine Bart's talk. The founder of SEO Sans Migraine covers the three pillars of SEO, a method for choosing keywords, and an SEO checklist for startups.

AI Tools for Startups: The 2026 Stack, from Pre-Seed to Series A

One starter AI tool per function, prices checked on September 26, 2026 against official pricing pages, and three costed stacks, from the solo pre-seed founder to the Series A team.

Latest in this category

Go to market: definition, strategy and examples for startups

Go to market explained: the definition, the five building blocks of a GTM strategy, four sales models illustrated by Doctolib, Atlassian, HubSpot and Pennylane, and a six-step plan.

Launching Your Product in the US: Hacker News, Product Hunt and Reddit in 2026

How to launch your product in the US in 2026 with Hacker News, Product Hunt and Reddit: the rules, real numbers from YC startups, and Anthony Riera's method.

B2B cold email: the method of a founder who sends 500 emails a day

Segment in batches of 100-200, run sequences of 5 to 9 emails, aim for a reply (never a sale): the B2B cold email method of a founder who sends 500 emails a day.