The ten-stage revenue lifecycle, and who owns each one

The AARRR funnel is nearly twenty years old and it has one structural flaw that matters more every year: it was designed to describe a customer's behaviour, not to assign responsibility for it.

That was fine when growth lived in one team. It isn't fine now, when the reason a customer doesn't renew is usually something that happened in a stage owned by somebody else.

We model ten stages, and every one has a named owning team. The number of stages matters less than the ownership. Here's the model and, more importantly, the reasoning about who's accountable at each point.

Key takeaways

• Stages exist to assign accountability, not to describe behaviour. If a stage has no owner, it has no advocate when something breaks.
• Price and Support are stages, not functions. Both leak revenue continuously and neither appears in a standard funnel.
• The stage where a loss is recorded is rarely where it was caused. Churn is the extreme case.
• Every stage has one owner and at least one dependency. The owner is accountable; the dependency is who they can't fix it without.
• Churn is owned by everyone, which sounds like a cop-out and is actually the point.

The model

Stage	Owner	Core question	Primary dependency
Acquire	Marketing	Are we buying customers who last?	Finance
Activate	Product	Do they reach first value?	Support
Price	Finance	Are we holding margin?	Sales
Adopt	Product	Do they use what they pay for?	Customer Success
Retain	Marketing	Do they come back?	Product, Engineering
Expand	Sales	Do they grow with us?	Product
Support	Support	Are problems systemic?	Engineering
Renew	Customer Success	Do they stay?	Finance, Engineering
Advocate	Marketing	Do they bring others?	—
Churn	Everyone	Why did we lose them?	All

Why these ten

Acquire — Marketing

Not signups. Quality of cohort. A channel is only good if the customers it brings still exist at month six with positive contribution. Measuring Acquire on volume is how businesses grow their way into worse unit economics.

Depends on Finance, because the discount policy that made a channel look efficient is a pricing decision made elsewhere.

Activate — Product

The single highest-leverage stage in most businesses, and the most commonly under-instrumented. A customer who pays and never reaches first value is a refund, a churn, or a support ticket that hasn't happened yet.

The hard part is defining first value concretely enough to measure. Not "logged in." First order delivered. First workflow completed. First teammate invited. If you can't name it in one sentence, your team is optimizing something they haven't agreed on.

Depends on Support, because the drop-off point shows up in ticket text weeks before it shows up in a funnel.

Price — Finance

Missing from almost every lifecycle model, which is why discount dependency runs unchecked for years. Price is a stage because pricing decisions happen continuously — at acquisition, at renewal, at expansion — and each one alters the customer's reference price permanently.

Depends on Sales, who hold the lever and are measured on something else.

Adopt — Product

The gap between what a customer pays for and what they use. In B2B it's the clearest renewal risk available months in advance. In B2C it's the subscription that lapses without a complaint.

Depends on Customer Success, who need the signal early enough to act.

Retain — Marketing

Repeat purchase and its decay. Marketing owns the number, which is the source of most organizational friction here — because the causes are frequently things Marketing doesn't control: delivery times, fee presentation, stock availability, a checkout deploy.

Depends on Product and Engineering. This dependency is the most consequential in the model, and the one most companies handle worst.

Expand — Sales

Upsell and cross-sell readiness. Usually visible as a usage signal well before it's visible as an opportunity, and usually missed because nobody instrumented the threshold.

Depends on Product, who own the limits that generate the signal.

Support — Support

A stage rather than a cost centre, because support contact is a revenue event. A customer who contacts you about a problem and then doesn't return has told you the price of that problem.

The reframe that matters: support teams are measured on resolution time, which optimizes for closing tickets fast rather than noticing that four hundred of them share a cause.

Depends on Engineering, who own most systemic causes.

Renew — Customer Success

Where voluntary and involuntary churn must be separated before anything else happens. A failed payment and a considered departure require completely different responses, and most renewal dashboards combine them into one number.

Depends on Finance and Engineering, who own the payment stack and the retry logic.

Advocate — Marketing

Referral behaviour. Usually happening and rarely instrumented, which means paid CAC efficiency is systematically overstated in businesses with strong word of mouth.

Churn — everyone

The deliberate exception. Churn has no single owner because churn is not a stage — it's a terminal event whose cause always lives in an earlier stage.

The discipline is attribution back to source. A churn caused by a failed payment belongs to Renew. One caused by never reaching first value belongs to Activate. One caused by a competitor genuinely being better belongs to nobody, and that's fine — natural churn exists and pretending otherwise leads to retention spend aimed at people who were never coming back.

Strong association across the businesses we work with, when churn is attributed back to source stage, the distribution is consistently more concentrated than teams expect. The instinct is that churn has many causes. The data usually says it has two or three that matter.

Why one owner and one dependency

The two failure modes here are symmetrical.

One owner, no dependency produces blame. Marketing owns Retain, repeat rate falls because of a deploy, and Marketing is asked to explain a number they didn't move. They run a campaign, because a campaign is what they control. It doesn't work, because the customers aren't hesitant — they're annoyed.

Shared ownership with no accountable party produces drift. Everyone assumes someone else is watching. The leak runs for months and the post-mortem concludes that "we should have better cross-functional communication," which is not a fix.

The model that works is one accountable owner plus a named dependency who is expected in the room. The owner can't be blamed for a cause they don't control, and the dependency can't ignore a problem that isn't in their dashboard.

In Flolyt, when an agent opens a Room on a Retain-stage leak that traces to a deploy, both Marketing and Product are pulled in automatically. Not as a notification — as participants, with the evidence already assembled.

Adapting the model

Two variants worth noting.

B2B account mode. Stages stay the same; the unit changes from individual customer to named account. Adopt becomes seat-level usage against entitlement. Expand becomes formal upsell motion. Renew becomes contract renewal with a date.

B2C consumer mode. Retain carries far more weight, Renew is often implicit rather than contractual, and Support volume-to-repeat correlation becomes one of the strongest available signals.

The stages don't change. The relative size of the leaks does.

Frequently asked questions

Why ten stages instead of the usual five? Because Price, Support, Adopt, and Advocate all leak revenue continuously and don't appear in a standard funnel. Collapsing them means those leaks have no owner.

What if we don't have separate teams for each stage? Then one person owns several stages, which is fine. The point is that each stage has a name attached, not that you need nine departments. Smaller companies often have one growth lead owning four stages, and the model still helps by making the handoffs explicit.

Isn't "everyone owns Churn" a cop-out? It would be if churn were a stage where work happens. It isn't — it's where losses are recorded. The actual work is attributing each churn back to the stage that caused it, and that stage has a real owner.

How is this different from a customer journey map? A journey map describes what the customer experiences. This describes who in your company is accountable when it goes wrong. Related, but they answer different questions.

Where do you start if you're instrumenting from scratch? Renew first — splitting voluntary from involuntary churn is one-time work with immediate payoff. Then Activate, because activation failures cause losses you won't see for months. Then Retain.