Skills / Startup-growth

Lean Analytics Stages

A good metric changes the way you behave.

Lean Analytics is a prescriptive, stage-based framework for building startups with data. Co-authored by Alistair Croll and Benjamin Yoskovitz, it guides founders through five sequential stages — Empathy, Stickiness, Virality, Revenue, and Scale — each anchored by a single One Metric That Matters (OMTM). The framework maps six business model archetypes to each stage, giving founders thirty specific analytical situations with concrete metric guidance, benchmark thresholds, and explicit exit criteria that determine when you've earned the right to move forward.

By Alistair Croll and Benjamin Yoskovitz  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisLean Analytics Stages

Input

“We're a B2B SaaS tool for content marketing teams. We have 340 paying customers, $18k MRR, and our team is stuck in a debate about whether to build…”

Diagnosis

Let's figure out which Lean Analytics stage you're actually in — because the right metric to track, and the right feature to build, depends entirely on that answer.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Empathy before stickiness, stickiness before virality — the startup analytics framework that prevents premature scaling

Lean Analytics organizes startup growth into five sequential stages — Empathy, Stickiness, Virality, Revenue, and Scale — each with a named One Metric That Matters (OMTM) and explicit stage gate thresholds that must be cleared before advancing. The framework's core discipline is sequential enforcement: you cannot legitimately optimize virality until stickiness is proven through retention cohorts, and you cannot claim revenue-stage thinking until LTV/CAC economics are healthy. Six business model archetypes (E-commerce, SaaS, Free Mobile App, Media Site, User Generated Content, Two-Sided Marketplace) are mapped to each stage, producing thirty distinct analytical situations with specific benchmark metrics — for example, SaaS monthly churn below 2% as the Stickiness exit criterion, or viral coefficient (K-factor) above 1.0 to claim a self-sustaining Virality stage. The Lean Analytics Cycle operationalizes the framework: identify your current stage, select the OMTM, set a threshold target, design and run experiments, analyze results, then advance or iterate. A defining discipline is the explicit rejection of vanity metrics: total registered users, page views, and gross sign-ups are systematically replaced by retention cohorts, engagement ratios, and unit economics.

The problem

Most founders track too many metrics and optimize for the wrong things at the wrong time — building viral loops before they've proven retention, or chasing revenue while customers are silently churning. The result is a dashboard full of vanity metrics that grow while the business quietly bleeds. Without a stage-based sequencing discipline, there's no principled way to decide whether building referral features or fixing onboarding is the higher-leverage move right now.

The solution

After a session with Lean Analytics Stages, you'll know exactly which of the five stages you're in, which single metric your team should obsess over this quarter, and precisely what threshold you need to hit before you've earned the right to move on to the next stage of growth.

You bring

A description of your startup's business model and the problem you solve
Your current metrics dashboard or the numbers your team tracks most often
Recent user behavior data — retention rates, engagement, sign-ups, churn, or revenue
Honest answers about where growth feels stuck or where your team disagrees on priorities

You get

Your current Lean Analytics stage identified with explicit, evidence-based reasoning
Your OMTM — the one metric to obsess over this quarter, tailored to your stage and business model archetype
A vanity metrics audit that replaces feel-good numbers with actionable, causal alternatives
The specific stage gate thresholds you need to hit before advancing, drawn from Lean Analytics benchmark data

In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“Your startup has 340 paying customers and $18k MRR growing 12% month-over-month. Your team is in a sprint planning fight: half want to build referral codes and a 'refer a friend' flow, half want to redesign the onboarding sequence. You have 23 metrics on your dashboard, regularly check MRR growth and total sign-ups, and haven't looked at a retention cohort in two months. Monthly churn is 4% but you've been telling yourself it's manageable because overall growth is positive.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With Lean Analytics Stages

“You know you're in the Stickiness stage with a stage gate problem at 4% churn. Your OMTM is monthly churn rate with a hard 2% target. You've removed 19 metrics from your dashboard, added a cohort retention chart by signup month, and discovered that customers who complete a specific workflow in week one churn at 1.1% versus 6.8% for those who don't. Every sprint for the next quarter routes through one question: does this change increase week-one workflow completion? The referral feature debate is closed.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

Lean Analytics StagesClaude

Lean Analytics Stages

Let's figure out which Lean Analytics stage you're actually in — because the right metric to track, and the right feature to build, depends entirely on that answer. Tell me about your startup: what problem you solve, who your users are, and which numbers your team looks at most often.

We're a B2B SaaS tool for content marketing teams. We have 340 paying customers, $18k MRR, and our team is stuck in a debate about whether to build referral features or improve the editor. I track DAU, MRR growth, total sign-ups, and LinkedIn followers.

Based on what you've described, I'd place you in the Stickiness stage — possibly approaching the Virality transition, but you haven't earned the right to cross it yet. Here's the diagnostic: 340 paying customers and $18k MRR tells me you've cleared the Empathy stage — real people pay real money for a real problem. But the fact that your team is actively debating referral features while you're tracking total sign-ups and LinkedIn followers suggests stickiness isn't locked down. Before we even discuss virality, I need one number from you: what's your monthly churn rate?

Churn is around 4% monthly. And honestly our DAU/MAU ratio is something we don't really track — we mostly look at MRR and sign-up volume.

4% monthly churn is a stage gate failure for SaaS. The Lean Analytics benchmark to exit the Stickiness stage is under 2% monthly churn — at 4%, you're losing close to 40% of your customer base annually before accounting for compounding. Building referral features right now would be pouring water into a leaky bucket. Your OMTM for this quarter is not MRR growth. It's monthly churn rate, with a hard target of getting below 2% before any virality work is authorized. As for MRR growth and total sign-ups: those are vanity metrics at your current stage. MRR can grow while your retention problem gets worse if you're acquiring faster than you're churning — and that's exactly the trap that hides stage gate failures.

That's a tough call — but you're right. We've been celebrating MRR growth without ever looking at cohorts. So what should our actual dashboard look like right now?

For a B2B SaaS product in the Stickiness stage, your dashboard needs exactly three things: (1) Monthly churn by cohort — not a blended average, because cohort view tells you whether newer customers retain better than older ones, which is the signal that your product is improving; (2) Feature adoption rate — which core features do retained customers use in week one that churned customers don't? That delta is your stickiness signal and your actual product roadmap; (3) Time-to-first-value — how quickly do new customers reach their first meaningful outcome in your tool? Everything else — LinkedIn followers, total sign-ups, MRR growth rate — comes off the dashboard until you've crossed the 2% churn threshold. The referral feature is tabled. The editor improvement is on the table only if cohort data shows it's a retention driver.

This is exactly the clarity our team needed. So once we hit under 2% churn — what does the Virality stage gate actually require?

Reply to Lean Analytics Stages…

What the skill is doing

01Stage Diagnosis

02OMTM Selection

03Vanity Metric Audit

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

LALean Analytics Stagestyping…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

We're a B2B SaaS tool for content marketing teams. We have 340 paying customers, $18k MRR, and our team is stuck in a debate about whether to build re…9:14✓✓

Let's figure out which Lean Analytics stage you're actually in — because the right metric to track, and the right feature to build…9:14

Tell me about your startup: what problem you solve, who your users are, and which numbers your team looks at most often.9:15

Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

◦Reads the situation, names the pattern, returns one concrete next move.

◦Delivered in seconds, inside a conversation that already exists.

Specimen · WhatsApp Business API · live

Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Stage Diagnosis

Determine which of the five Lean Analytics stages your startup currently occupies, using your metrics, user behavior patterns, and business description as evidence. The diagnosis surfaces where you've genuinely earned the right to be and where you may be prematurely optimizing — often revealing that a startup celebrating revenue growth is still in the Stickiness stage with an unaddressed churn problem.

◦ MethodBased on Croll and Yoskovitz's sequential stage logic: a startup cannot claim Stickiness without cohort-based retention proof, cannot claim Virality without K-factor evidence above 1.0, and cannot claim a Revenue stage without LTV/CAC economics that make sense — each stage has specific evidential criteria, not just time-based progression.

CapabilityC-02

OMTM Selection

Identify the One Metric That Matters for your current stage and business model archetype — the single number to be obsessed with above all else for this period. This metric changes deliberately as you advance through stages, and selecting the wrong one causes misaligned sprint priorities, wasted experiments, and the illusion of progress against a metric that doesn't actually gate the next stage.

◦ MethodDrawn from the book's 6×5 business model × stage matrix, which specifies the OMTM and benchmark targets for each of thirty analytical combinations — for example, DAU/MAU retention ratio for a SaaS product at the Stickiness stage, viral coefficient (K-factor) at the Virality stage, or LTV/CAC ratio at the Revenue stage for a two-sided marketplace.

CapabilityC-03

Vanity Metric Audit

Review your current dashboard and classify each metric as vanity (looks impressive, doesn't change behavior or indicate causation) or actionable (changes decisions, is comparable over time, and reflects a real causal lever). Vanity metrics get flagged with their actionable counterparts identified, and the dashboard gets trimmed to the minimum set that actually drives decisions.

◦ MethodBased on Croll and Yoskovitz's four criteria for a good metric: it must be comparative (a ratio or rate, not an absolute), understandable (a non-technical team member can act on it), behavioral (it changes how you make decisions), and causal rather than merely correlated. Total registered users and page views fail; retention cohorts and conversion-by-funnel-step pass.

CapabilityC-04

Stage Gate Assessment

Evaluate whether your startup has hit the specific quantified thresholds required to legitimately advance from one stage to the next. Stage gates prevent the common and expensive mistake of moving to growth and virality tactics before core retention mechanics are proven — a trap that wastes capital and masks fundamental product-market fit problems.

◦ MethodBased on the specific benchmark thresholds in the Lean Analytics framework: SaaS monthly churn below 2% as the Stickiness exit criterion, viral coefficient (K-factor) above 1.0 to claim a self-sustaining Virality stage, and healthy LTV/CAC ratio (typically 3:1 or better) as the Revenue stage gate before scaling spend.

CapabilityC-05

Archetype-Specific Metric Mapping

Classify your business into one of the six Lean Analytics archetypes and map the correct KPIs to your current stage. The right metric for a two-sided marketplace in the Revenue stage is structurally different from the right metric for a SaaS company at the same stage — using the wrong archetype's metrics is a common source of false confidence or misplaced pessimism.

◦ MethodBased on the six-archetype taxonomy (E-commerce, SaaS, Free Mobile App, Media Site, User Generated Content, Two-Sided Marketplace) and the book's stage-specific metric tables for each archetype, which identify the key benchmark numbers, typical thresholds, and most common failure modes for each of the thirty stage-model combinations.

Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces

OutputD-01

Stage Diagnosis Report

A structured assessment identifying which of the five Lean Analytics stages your startup currently occupies, with the specific evidence supporting that determination, what it means for your next sprint priorities, and what you need to address before advancing.

OutputD-02

OMTM Selection Card

Your One Metric That Matters for the current stage and business model archetype — with the specific benchmark target you're aiming for, why this metric was chosen over alternatives, and the top experiments most likely to move it.

OutputD-03

Vanity Metrics Audit

A line-by-line review of your current dashboard that classifies each metric as actionable or vanity, replaces vanity metrics with their causal counterparts, and identifies the minimum set of three to five numbers that should remain on the wall.

OutputD-04

Stage Gate Threshold Checklist

The specific quantified thresholds — drawn from Lean Analytics benchmark data for your business model archetype — that you need to clear to have earned the right to advance to the next stage, with your current status marked against each criterion.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Alistair Croll and Benjamin Yoskovitz

Alistair Croll is a Canadian entrepreneur, author, and conference organizer who co-founded Coradiant (acquired by BMC Software) and helped launch O'Reilly's Strata Data Conference. He has spoken on startup analytics at the Lean Startup Conference, SXSW, and MicroConf. His co-author Benjamin Yoskovitz is a startup founder and investor with multiple early-stage companies to his name. Together they wrote Lean Analytics (O'Reilly, 2013), now a standard text in MBA entrepreneurship programs and widely adopted in startup accelerators including Y Combinator.

Status · Inspired by Alistair Croll and Benjamin Yoskovitz’s work — not yet claimed. Are you Alistair Croll and Benjamin Yoskovitz?

Primary sourceS-01

Lean Analytics: Use Data to Build a Better Startup Faster (O'Reilly, 2013)

by Alistair Croll and Benjamin Yoskovitz

Alistair Croll: serial entrepreneur (Coradiant/BMC Software acquisition), O'Reilly Strata Data Conference co-organizer, keynote speaker at Lean Startup Conference and SXSW; Benjamin Yoskovitz: multi-time startup founder and early-stage investor.

Read the original ↗

Citationleananalyticsbook.com

In the build queue

Be first to run it.

Lean Analytics Stages is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail

At launchI have paying customers and growing MRR, but my team is debating whether to build referral features or fix our onboarding — and I can't figure out which is actually the right move. Can you help me figure out which Lean Analytics stage I'm in and what my one metric should be right now?