MethodologyLast updated April 2026
How we find landing pages that convert
Analyzed section by section. Updated every week. This page documents how we do it, so the ranking is reproducible and so you can disagree with it.
Document what actually converts on SaaS landing pages, with evidence.
The largest library of landing pages that convert.
Landing Page Analyzer is a research platform: a library of landing pages that convert, with deep analyses for the top entries, so you can study what actually works before you ship.
From sourcing to analysis, every week.
- 01Step 01 — Sourcing
How we curate pages
We browse. No submissions, no paid placement, no vendor endorsement. Pages enter the library when they earn a spot against existing entries.
Roughly 30 new sections enter the library per week. Existing pages get re-analyzed every 3 months, or sooner when a company ships a significant homepage change.
Companies cannot pay to appear, buy a better score, or remove a negative analysis. Commercial relationships, when they exist, are disclosed on the analysis page.
- 02Step 02 — Framework
How we score pages
Our goal is not to judge whether a page looks good — it is to judge whether it converts. The framework draws on conversion copywriting, UX research, behavioral psychology, and the recurring patterns we see on pages that actually convert. It does not judge visual design trends or aesthetic polish — a page can be ugly and still convert, and a beautiful one can still fail.
The framework runs at two levels.
- Page-level
- Each full page is analyzed against 60+ criteria covering value proposition, trust, CTA hierarchy, copy, visual hierarchy, objections, and mobile UX.
- Section-level
- Each section (hero, pricing, testimonials, FAQ…) is analyzed against 5–10 criteria specific to its job. A pricing section is not judged like a hero.
- 03Step 03 — Review
Who reviews the analysis
Conversion analysis is subjective. The framework used here was defined by Gabriel Amzallag.
Our internal agent runs the first pass against that framework — page-level and section-level. Gabriel reviews every analysis before publication. No fully-automated rankings.
Think an analysis is wrong? Tell us.
If you believe an analysis is incorrect, email contact@gabrielamzallag.com with the URL and the part you disagree with. We review every dispute and document corrections on the affected page.
Research before you build.
Browse the library, study the analyses, or run your own page through the same framework.