Observe the existing page to find hypotheses that may be blocking conversion. New landing page creation is linked to marketing-landing-page, and e-commerce detail pages are linked to the relevant skill in moai-seller.
Required materials
- The current landing page URL or screen materials, and which device, country, and browser will be used to view them
- The linked ad copy and the actual post-click destination URL
- The analysis period, and the denominators and sources for visits, key actions, and purchase or application metrics by traffic channel
- Verified evidence such as product price, shipping, refunds, warranty, reviews, and certifications
If there is no access permission, state the scope of judgment based only on the provided screens. If the screen could not be opened, do not write that a screen review was performed.
Review order
- Check whether the ad's promise matches the product, benefits, and conditions on the destination page. The wording does not need to be identical, but the meaning and conditions must not differ.
- Check whether the target customer can understand the value offered and the next action on the first screen.
- Check whether the price, shipping, refunds, privacy handling, and contact channels match actual policies and are easy to find.
- For reviews, ratings, figures, and before-and-after photos, verify the source and usage rights. Do not create social proof that does not exist or newly promise a refund guarantee.
- If the input, buttons, and payment or application flow can be directly checked on mobile and desktop, record the behavior. If there is no permission or environment, leave it unverified.
- For CTR, conversion rate, and bounce rate, compare after checking the denominators for the same period and segment. Do not judge the landing page as the sole cause based only on high click-through rate and low conversion rate.
Results
| Observation | Evidence and screen location | Customer impact hypothesis | Improvement proposal | Required verification | Priority |
|---|---|---|---|---|---|
| Items actually verified | URL, screen, data | Mark as estimate | Actionable changes | Measurement, policy, development verification | Evidence |
Distinguish problems, hypotheses, and confirmed causes. Do not present arbitrary total scores, expected conversion lift rates, or fixed step-by-step schedules. After applying an improvement, compare actual metrics or design an experiment. If a copy review skill is installed, use it, but cross-check the original text and figures; if not, review directly.
Connections
- content-copywriting: headline and CTA drafts tailored to verified product conditions
- marketing-landing-page: new landing page implementation
- marketing-growth-experiment: experiment design to verify the effect of changes
- marketing-performance-report: actual performance comparison