CRO Methodology

The Dalayon CRO loop has four stages: Investigation, Research, Optimization, and Evaluation. The goal is to move from evidence to action without turning every opinion into a test.

The Loop

Stage	Purpose	Typical Inputs	Output
Investigation	Find where performance is leaking	Analytics, funnel data, page speed, product performance, customer journey notes	Prioritized problem list
Research	Understand why the leak may exist	Session recordings, heatmaps, customer language, reviews, support tickets, competitor scan	Evidence-backed hypotheses
Optimization	Ship the smallest useful improvement	Test design, mockups, theme changes, QA checklist, launch plan	Experiment or direct rollout
Evaluation	Decide what to keep, change, or stop	Test results, pre/post read, revenue per visitor, CVR, ATC, AOV, qualitative notes	Decision log and next action

Principles

1. Start with a business question, not a tactic.
2. Separate traffic quality problems from on-site conversion problems.
3. Prefer clear measurement windows over noisy snapshots.
4. Treat every result as directional until the sample is strong enough.
5. Record the decision, not just the numbers.

When To A/B Test

Use an A/B test when the change is reversible, traffic is sufficient, and the outcome is uncertain enough to justify splitting traffic.

Use a direct rollout when the change is clearly corrective, required for brand or technical reasons, or the traffic level cannot support a useful test. In that case, write a pre/post measurement plan before launch.

Standard Decision States

Decision	Meaning
Roll out	Variant clearly improves the primary metric and does not create a material trade-off
Keep control	Variant underperforms or adds risk without enough upside
Iterate	Signal is promising but the variant needs adjustment
Re-run	Data quality, tracking, timing, or sample size makes the result unreliable
Monitor	Direct rollout shipped and needs a clean post-launch read