99Ways Academy — PostHog & CRO Guides
Sharing our optimization experiences, won or failed
One-Time CRO Redesign vs Ongoing A/B Testing: How to Choose the Right Starting Point
A tested one-time redesign can be rational when the current experience needs a structural reset. Ongoing A/B testing creates more long-term value when…
How to Evaluate a CRO Agency: Measure Deployed Business Impact, Not Win Rate
A CRO agency should be judged by the complete evidence chain it creates: trustworthy measurement, explicit hypotheses, valid experiments, verified deployments, and cumulative…
Event Tracking Plan: A Practical Framework for Connecting Traffic, Behavior, and Revenue
An event tracking plan is not an event-name spreadsheet. It is the contract that defines how business questions become entities, events, trigger rules,…
PostHog Audit: 12 Checks Before You Trust the Data
A PostHog audit should prove more than event delivery. These 12 checks follow the full measurement chain—from ad click and identity through event…
PostHog vs GA4: Which Should Be Your Source of Truth for Growth Decisions?
PostHog is usually better suited than the GA4 reporting interface to be the operational source of truth for growth decisions. It provides direct…
Experimentation System Audit: How to Know Whether Your A/B Test Results Can Be Trusted
An A/B result is only as trustworthy as the system that produced it. This audit framework examines eligibility, assignment, bucketing, identity persistence, exposure,…
Conversion Tracking Audit: How to Know Whether Your Funnel Data Can Be Trusted
A conversion tracking audit should prove more than whether a tag fires. It should verify event meaning, duplicates, identity, cross-domain continuity, attribution, consent…
What Is Growth Lever Analytics? A Decision System for Finding What Actually Moves Growth
Growth Lever Analytics is a decision process for finding the controllable variable worth acting on, building the measurement needed to trust the decision,…
How Much Does an A/B Test Really Cost?
What does it cost to run an A/B test? The obvious answer is the cost of producing it. Someone needs to research the…
