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 clicks are a misleading KPI for affiliate businesses? ======


{{ :kpi.png?nolink&600 |}}

In traditional affiliate tracking, **clicks** are often treated as a primary KPI.

CBSplit exists because this assumption breaks down in real revenue systems.

Clicks describe **movement**, not **outcomes**.


===== How traditional affiliate tracking thinks =====

Most affiliate stacks stop tracking at the click or at the initial conversion.

The logic usually looks like this:

  * Ad → Click
  * Click → Offer
  * Offer → Sale (assumed success)

In reality, this model ignores everything that happens **after** the click.

CBSplit was built specifically to observe what traditional tracking ignores.

===== What a click actually means (in CBSplit terms) =====

From a CBSplit perspective, a click means only one thing:

A visitor crossed a boundary.

It does **not** confirm:

  * Payment completion
  * Payment retries
  * Refund eligibility
  * Subscription continuity
  * Rebill success
  * Fraud or velocity checks
  * Chargebacks
  * Net revenue attribution

A click is an **entry event**, not a revenue event.

===== Why click-based optimization fails at scale =====

When affiliates optimize primarily for clicks:

  * Traffic quality degrades over time
  * Conversion volatility increases
  * Refunds rise after payout windows
  * Rebill performance drops
  * Merchant trust erodes

From CBSplit data, many high-click campaigns show:

  * Strong front-end metrics
  * Weak back-end revenue
  * Negative long-term ROI

Clicks scale faster than truth.

===== What clicks hide from standard dashboards =====

Standard dashboards typically show:

  * Click count
  * Initial conversion count
  * Gross revenue

CBSplit reveals what those dashboards hide:

  * Failed payment retries
  * Delayed refunds
  * Subscription churn after day 30
  * Rebill suppression
  * Processor-level failures
  * Traffic-source-specific decay

Clicks remain constant while real revenue silently collapses.

===== CBSplit treats clicks as a signal, not a KPI =====

In CBSplit, clicks are treated as **diagnostic signals**, not success metrics.

They help answer:

  * Which sources send volume
  * Which creatives trigger curiosity
  * Which funnels attract attention

They do **not** answer:

  * Which traffic is profitable
  * Which partners scale safely
  * Which offers survive refunds
  * Which campaigns generate durable revenue

===== What CBSplit measures instead =====

CBSplit focuses on **post-click truth**.

Key metrics include:

  * Net revenue after refunds
  * Subscription survival across rebills
  * Revenue decay curves
  * Time-delayed attribution
  * Source-level refund ratios
  * Offer-level long-term performance

These metrics reflect **business reality**, not surface activity.

===== Why clicks create false confidence =====

Clicks produce fast feedback.

Revenue truth arrives slowly.

This timing mismatch causes:

  * Premature scaling
  * Misleading A/B test results
  * Overpayment to traffic sources
  * Late discovery of bad actors

CBSplit aligns measurement with **when money actually settles**, not when curiosity spikes.