====== Why two funnels with same front-end behave oppositely ======

===== The surface similarity illusion =====

Two funnels can look identical.

Same landing page.
Same headline.
Same checkout flow.
Same upsell stack.

Front-end metrics appear similar.

Yet over time, one becomes profitable and stable.
The other collapses.

CBSplit was built to explain this divergence.

===== The front end is only the entry layer =====

Front-end similarity hides backend differences.

Profitability depends on:

  
  * Refund behavior
  * Subscription retention
  * Rebill success
  * Customer satisfaction
  * Processor stability

Two funnels can convert at the same rate and still produce opposite lifecycle outcomes.

===== Traffic source changes everything =====

Even with identical front-end assets:

  
  * Different traffic sources produce different buyer intent
  * Messaging alignment varies
  * Expectation clarity shifts
  * Refund probability changes

High-intent traffic creates durable cohorts.

Low-intent traffic inflates short-term metrics and erodes backend stability.

CBSplit segments lifecycle performance by traffic origin.

===== Messaging before the click matters =====

Pre-click messaging defines:

  
  * Buyer expectations
  * Perceived value
  * Emotional framing
  * Purchase intent depth

Two funnels with identical pages can attract:

  
  * Aligned buyers in one campaign
  * Misaligned buyers in another

The backend reveals the difference.

===== Subscription dynamics expose divergence =====

In recurring models, identical front ends can produce:

  
  * Strong rebill survival in one funnel
  * Rapid churn in another
  * Stable lifetime value versus declining cohorts

Front-end data cannot predict subscription durability alone.

CBSplit evaluates cohort stability across billing cycles.

===== Refund timing reverses conclusions =====

Early metrics may show similar performance.

After refund windows close:

  
  * One funnel maintains revenue
  * The other experiences refund clusters

Gross revenue equality hides net revenue divergence.

CBSplit recalculates profitability after refund reconciliation.

===== Upsell sequencing interacts with audience intent =====

Identical upsell flows can:

  
  * Reinforce value for aligned customers
  * Trigger regret for misaligned customers

Audience alignment determines whether upsells strengthen or weaken retention.

Front-end similarity does not guarantee behavioral similarity.

===== Scaling magnifies hidden differences =====

At small volume, divergence may appear minor.

When scaled:

  
  * Refund exposure multiplies
  * Churn patterns accelerate
  * Risk thresholds approach
  * Net profitability separates clearly

Small lifecycle differences compound under scale.

===== Profitability is defined by lifecycle durability =====

True funnel performance depends on:

  
  * Refund-adjusted revenue
  * Rebill-adjusted LTV
  * Cohort retention stability
  * Traffic-quality alignment

Front-end appearance is not predictive of lifecycle resilience.

CBSplit measures the entire system.