The six customer life cycle stages are awareness, acquisition, activation, retention, revenue, and referral. Awareness is when someone learns you exist. Acquisition is their first action, usually a visit or a signup. Activation is the first time they experience your core value. Retention is coming back. Revenue is paying you, or paying you more. Referral is telling someone else.
Each stage has its own entry event, its own metric, and its own failure mode, and a customer can only be in one at a time. That is what makes the model useful operationally: the transition rate between two adjacent stages tells you where people are disappearing, which is a different question from how many people you have in total.
Which is also where most uses of the model go wrong. Drawn as six boxes it invites you to count the population in each and work on the one that looks thin, and that is close to the opposite of what it is for. Three questions make it operational instead of decorative. What marks the boundary between stages? Which transition is actually holding growth back? And what does the model get wrong badly enough to be worth naming?
The newsletter
Join our KISS newsletter
One short read a week on what actually moves revenue, in a free email. Read by 10,000+ operators and founders.
No spam. Unsubscribe in one click.
I.Six stages, and why the counts inside them are the wrong thing to read
Each stage is defined by an entry event, which is what makes the boundaries measurable, and what makes the transitions rather than the populations the useful number.
A.The six boundaries
The closest well-known relative is Dave McClure’s pirate metrics, five stages whose initials spell AARRR. The six-stage version adds awareness at the front, because most measurement gaps start before anyone has visited your site. Each boundary is an event you can instrument, and that is the only property that matters.
Awareness begins at a first impression: a search result, a social post, an ad, a podcast mention, a recommendation. They know you exist and hold no opinion. It is measured by impressions and reach, which are volume indicators, and rather better by brand search volume, which is one of the strongest signals that awareness work is landing. Acquisition begins at a first meaningful action, typically a signup or an account, and is measured by signup rate and cost per acquisition segmented by channel. Activation begins at the first experience of core value, the event you define, and is measured by activation rate and time to activate. Retention begins at a return visit after activation, and is measured by cohort curves and engagement frequency. Revenue begins at the first payment and continues through expansion, measured by LTV, ARPU, and expansion share. Referral begins when a customer produces a new one.
The six stages as one funnel
Funnels report viewB.Why the counts mislead and the transitions do not
The population in a stage is a fact about your history and your budget. The transition rate out of it is a fact about your product right now. Ten thousand people in Awareness means you bought or earned ten thousand impressions, which you already knew from the invoice. Three percent of them reaching Acquisition is new information, and it is information about the offer rather than the spend.
The practical consequence is a reversal of what most teams do. Reading counts, the biggest number is at the top and the smallest at the bottom, so attention flows to Referral, which looks catastrophically thin, or back to Awareness, which looks productive. Reading transitions, the picture usually inverts: the steepest single drop is somewhere in the middle, and it is capping every number below it. A company that excels at Awareness and fails at Activation is filling a bucket with a hole in it, and the count view shows that as a healthy top of funnel.
The second consequence is diagnostic. Awareness up with Acquisition flat is a conversion problem on your entry point. Acquisition up with Activation flat almost always means the new users are lower quality, which points at channel mix rather than onboarding. Neither of those readings is available from the populations, because both are consistent with every count going up.
II.The two transitions that are almost always the constraint
Two of the five transitions do far more damage than the rest, and neither of them is the one that gets funded.
A.Activation is the steepest drop and the cheapest to move
Activation is the moment a new user experiences your core value for the first time. On most products the largest single drop in the whole lifecycle happens here: a sizeable share of accounts are created, poked at briefly, and never used again. Measure your own drop before assuming it is small.
The reason it is the cheapest transition to move is arithmetic rather than optimism. Every stage below it is multiplied by it, so a proportional improvement in activation produces the same downstream volume as a much larger proportional increase in traffic, at none of the acquisition cost.
Getting there starts with defining the event, which is a decision rather than a discovery: creating a project and inviting a teammate for a project tool, a first purchase for a marketplace. The requirement is that the event correlates with 90-day retention in your own data, and that once fixed the definition stops moving. Then onboarding is designed to reach it fast: fewer choices before the first result, contextual guidance at the step people stall on, and a follow-up to the ones who stalled. Our guide to activation rate optimization covers the tactics.
B.Retention is the one that caps everything above it
Activation is the steepest drop; retention is the one that decides whether fixing it mattered. A business with strong retention compounds, because each new customer adds to a growing base. A business with weak retention replaces churning customers to stay flat. At 5% monthly churn you lose nearly half your customer base every year, and no acquisition budget outruns that.
Still active, by signup cohort
Cohorts report view| Cohort | Day 7 | Day 30 | Day 60 | Day 90 |
|---|---|---|---|---|
| Januaryn=1,240 | 61% | 38% | 29% | 25% |
| Februaryn=1,380 | 64% | 41% | 33% | 28% |
| Marchn=1,510 | 66% | 45% | 36% | |
| Apriln=1,490 | 68% | 47% |
The two transitions also constrain each other, which is why they belong in the same part. Improving activation without retention pushes more people into a stage they leave, and the lifecycle dashboard shows a win at one transition and no change in revenue. Improving retention without activation improves the experience of a group too small to matter. The order that works is activation first, because it is faster to move and it enlarges the population any retention work applies to, then retention, measured on the cohorts that arrived after the activation change.
Here the model reaches its own limit, and it is worth being blunt about it. The lifecycle tells you retention is the constraint. It has nothing whatsoever to say about why people leave, and the honest answers are usually product answers: the value was thinner than the first session suggested, or a competitor is better, or the use case was occasional all along. Exit surveys, churn split by segment, and behavioural comparison of churned against retained users are how you find out. Our guide to diagnosing SaaS churn covers the method.
III.Where the model stops being a line, and where it breaks
The lifecycle is a loop rather than a funnel, and it assumes a kind of measurement most analytics stacks cannot actually perform.
A.The stages feed each other backwards
Drawn as a funnel, the last stage is the smallest and least important. Drawn honestly, it is an input to the first. Referral feeds Awareness at no acquisition cost. Revenue funds the marketing that drives Awareness. Retention data should decide which audiences Acquisition targets, because the channel that produces the flattest cohort curve is the one worth more budget regardless of its cost per signup.
Referral is the clearest case of a stage that pays back upstream. A widely cited Wharton study of a German bank found referred customers were worth roughly 16% more in lifetime value and churned less than customers from other channels. The exact figure will not generalise, but the mechanism does: a referred user was pre-qualified by someone who already understands the product, so they self-select into better fit and arrive further along than a cold visitor. That makes Referral an Awareness channel with an activation advantage built in, which is a very different thing from the thinnest box on a diagram.
None of that survives a straight-line reading. Neither does the most useful thing the model does, which is to price a transition. Calculate what one percentage point on each transition rate is worth in annual revenue and the ranking is usually surprising: a point of retention beats a point of acquisition many times over, and without the calculation teams default to acquisition because it is the more visible spend.
B.What the model assumes about measurement, and usually does not get
The model assumes you can hold one person across the whole sequence, because every number it produces is a transition, and a transition is a statement about the same individual at two points in time. Awareness to Acquisition can be weeks. Acquisition to Revenue can be months. Session-based analytics cannot make that statement: it sees a person as several unrelated anonymous rows across devices and weeks, so the transitions it reports are comparisons of two populations that happen to be roughly the same size. Person-level analytics is what turns the model from a diagram into a measurement.
The gaps are not evenly distributed either. Awareness is the worst measured stage in almost every stack, and it is getting worse, because a growing share of first impressions now happens inside ChatGPT, Claude, Gemini, and Perplexity, and those visits arrive labelled direct. The LLM Acquisition report separates them out, which is the difference between an Awareness stage you can attribute and one you can only estimate.
Practically, that means three things to build: funnel reports on the five transitions, cohort curves for the retention one, and a person key that survives across devices so the first two mean anything. In KISSmetrics you describe the stage in the chat and it builds the metric and the report behind it, with autocapture and an autoconfiguration pass handling the underlying tracking, so the work is defining the boundaries rather than instrumenting them.
Verdict
The six stages are a diagnostic, not a plan, and the only correct use of them is to rank the five transitions by what one percentage point is worth and work the top of that list. Run that calculation and for most companies it returns the same order: activation first, because it is the steepest drop and the fastest to move; retention second, because it decides whether the activation win survives; and awareness last, whatever the current marketing budget suggests. Referral is not a stage to fund, it is what a healthy retention curve produces on its own.
The model earns none of that if you read the boxes instead of the arrows. A lifecycle dashboard showing six populations is a restatement of your spending history with a framework drawn over it. A dashboard showing five transition rates, each measured on individuals rather than sessions, names your constraint in a single screen and updates it when the constraint moves. Build the second one, work one transition at a time, and stop buying awareness until the two below it have flattened.
This is part of Analytics metrics, defined and explained, under growth and lifecycle. The guide puts the rest of the pieces in order.
Continue Reading
The Customer Lifecycle: A Framework for Tracking What Actually Matters
Most businesses only track acquisition. The customer lifecycle framework gives you a structured way to measure awareness, desire, purchase, repeat purchase, and advocacy so you can grow at every stage.
Read articleBuilding Your First Funnel: A Step-by-Step Guide to Tracking Conversions
A well-built funnel tells you exactly where you are losing customers and how to fix it. This guide walks through building your first funnel, benchmarking each step, and testing improvements.
Read articleThe Churn Prevention Workflow: Detecting and Saving At-Risk Customers Automatically
By the time a customer tells you they are leaving, it is too late. Automated churn prevention workflows detect risk signals weeks before cancellation and trigger interventions that actually work.
Read article