# What are companies A/B testing on their landing-page CTAs right now?

Source: Lazyweb Research
Published: 2026-07-07
Sample size: n=232
Tags: landing-page, experiments, ux-patterns, mobile, web
HTML: https://www.lazyweb.com/research/what-are-companies-ab-testing-on-landing-cta
Markdown: https://www.lazyweb.com/research/what-are-companies-ab-testing-on-landing-cta.md

**Answer.** CTA is the fourth most-tested landing area with 232 detected experiments, behind hero (488), offer (301), and pricing (254).[1] The active themes are free-trial wording (77, 33%), price-in-button (53, 23%), and demo paths (25, 11%).[2] All figures are detected before/after diffs with inferred rationale, not measured lift.

> 232 detected CTA experiments — the #4 most-tested landing area, July 2026.

## Finding: CTA sits mid-pack in test volume, mobile-heavy

CTA draws 232 experiments and skews mobile (141 mobile vs 91 web).[1][3]

| Area | Experiments | Avg impact |
|---|---|---|
| HERO | 488 | 3.28 |
| OFFER | 301 | 3.69 |
| PRICING | 254 | 3.52 |
| CTA | 232 | 3.16 |

The mobile tilt matters: more CTA experimentation happens on phones than on desktop.

## Breakdown: the CTA test menu

The 232 CTA experiments cluster into a handful of themes (non-exclusive):[2]

| Theme | Count | Share |
|---|---|---|
| Free-trial wording | 77 | 33% |
| Price in button | 53 | 23% |
| Demo path | 25 | 11% |
| Waitlist | 1 | <1% |

Beyond wording, the qualitative rows show structural swaps — removing a CLI install widget or an AI prompt box in favor of a plain account-first button.

## Breakdown: structural CTA swaps

Two detected examples go beyond copy (observed change + inferred rationale):[3]

- **Factory (web):** a hero CLI install widget ('curl ... | sh' + OS tabs) replaced with a single 'START BUILDING' button. Inferred rationale: enterprise-buyer traffic isn't terminal-ready, so an account-first funnel guides installation later.
- **Replo (web):** an interactive 'Build With AI' prompt box replaced with a standard 'Sign Up For Free' / 'Book a Demo' pair. Inferred rationale: serious buyers want a workflow platform, not a toy demo.

## The numbers

| Stat | Computed from |
| --- | --- |
| CTA 232 experiments, #4 area (hero 488, offer 301, pricing 254) | experiment_area_leaderboard / cta_experiments_total |
| Free-trial 77 (33%), price-in-CTA 53 (23%), demo 25 (11%), waitlist 1 of 232 | cta_change_themes |
| CTA platform split 141 mobile / 91 web; Factory and Replo examples | cta_platform_split / qualitative[factory, replo] |

## Methodology

Universe: 232 detected CTA experiments within 2,160 annotated, July 2026. Method: area annotation counts, keyword incidence, and named-company qualitative rows. Caveat: observed changes with inferred rationale, not measured A/B lift.

## Sources & citations

- [1] Lazyweb Research analysis of 2,160 annotated experiments (detected-experiment corpus), July 2026. Area annotation counts; impact is a model-assigned 1-5 relative score.
- [2] Lazyweb Research analysis of 232 CTA experiments (detected-experiment corpus), July 2026. Non-exclusive keyword incidence over summary+learning text.
- [3] Lazyweb Research analysis of 225 named-company CTA experiments (detected-experiment corpus), July 2026. Platform split and observed before/after CTA changes with model-inferred rationale.

## Related questions

- ['Book a demo' vs 'Start free': which primary CTA should my landing page use?](https://www.lazyweb.com/research/book-demo-vs-start-free-primary-cta-choice)
- [Should I put the price inside my landing-page CTA button?](https://www.lazyweb.com/research/should-i-put-price-inside-my-cta-button)
- [When should I drop the free-trial CTA and route everyone to pricing?](https://www.lazyweb.com/research/when-do-incumbents-drop-the-free-trial-cta)