CBSplit CMS Wiki

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Why CBSplit focuses on outcomes, not events ======

===== The event obsession =====

Most tracking systems are event-driven.

A click fires an event.
A page view fires an event.
A button press fires an event.

Dashboards fill up quickly.
Charts move in real time.

Yet revenue understanding barely improves.

CBSplit was built to break this pattern.

===== Events describe activity, not success =====

Events answer questions like:  
  * Did something happen?
  * Where did the user click?
  * Which variation was shown?

They do not answer:  
  * Did the payment succeed?
  * Was the order clean or fragile?
  * Did the user refund later?
  * Did revenue survive over time?

Activity is not the same as outcome.

CBSplit makes that distinction explicit.

===== Funnels can generate events and still fail =====

A funnel can be full of events:  
  * Page views increase
  * Clicks look healthy
  * Conversions appear stable

Yet still produce:  
  * High refunds
  * Retry-heavy payments
  * Early subscription churn
  * Processor risk

Event-heavy funnels often hide outcome failures.

CBSplit surfaces them.

===== Outcomes arrive late and out of order =====

Real outcomes do not happen instantly.

They arrive:  
  * After retries
  * After fulfillment
  * After customer support
  * After rebills
  * After refund windows close

Event-based systems prefer immediacy.

CBSplit tolerates delay because truth is delayed.

===== Events are cheap. Outcomes are expensive =====

Events are easy to generate.

Outcomes involve:  
  * Money
  * Risk
  * Compliance
  * Accountability

Because outcomes are harder, most tools avoid them.

CBSplit embraces outcome complexity instead of simplifying it away.

===== CBSplit treats events as inputs, not conclusions =====

CBSplit does not ignore events.

It uses them as signals.

But decisions are driven by:  
  * Final payment resolution
  * Net revenue after refunds
  * Subscription survival
  * Long-term customer behavior

Events inform.
Outcomes decide.

===== Outcome-based systems enable real optimization =====

When optimization is outcome-driven, you can:  
  * Fix retry logic instead of boosting clicks
  * Improve offer alignment instead of inflating conversions
  * Protect accounts instead of chasing volume
  * Scale revenue quality, not just traffic

Event-based optimization cannot do this reliably.

===== Why traditional tools stop at events =====

Most tools focus on events because:  
  * Events are immediate
  * Events are browser-visible
  * Events are easy to visualize
  * Events look impressive in dashboards

Outcomes are slower, messier, and harder to sell.

CBSplit was built for operators, not dashboards.