# How Many Detected Experiments Change CTA Copy, and What Do Companies Rewrite?

Source: Lazyweb Research
Published: 2026-07-07
Sample size: n=561
Tags: paywall, checkout, experiments, monetization, ux-patterns
HTML: https://www.lazyweb.com/research/how-many-experiments-touch-cta-copy-and-what-changes
Markdown: https://www.lazyweb.com/research/how-many-experiments-touch-cta-copy-and-what-changes.md

**Answer.** Across Lazyweb's 4,814-experiment corpus, 561 detected experiments (12%) changed CTA button copy, spanning 201 companies — the widest company spread of any tactic tracked.[1][2] On paywalls specifically, 344 of 795 detected paywall diffs (43%) changed the primary CTA text.[3] Common rewrites move between generic continuation verbs and explicit value or offer language.[4]

> 561 of 4,814 detected experiments changed CTA copy across 201 companies, and 43% of 795 detected paywall diffs rewrote the primary CTA — Lazyweb Research, July 2026.

## The finding: CTA copy is the most broadly tested tactic

561 detected experiments changed CTA button copy (where 'cta' or 'button' co-occurs with copy/text/label/wording/reword), across 201 companies — more distinct companies than any other tactic in this family.[1] That breadth matters: CTA copy is the cheapest lever, so almost everyone touches it. On the paywall-specific corpus (795 detected diffs across 146 companies), 344 (43%) changed the primary CTA text.[3]

## What the rewrites look like

The most repeated before-to-after paywall CTA swaps in the corpus:

| Before | After | Times |
|---|---|---|
| continue | try for free | 3 |
| continue with core | upgrade | 3 |
| restart subscription | complete subscription | 3 |
| get youtube premium | renew now | 3 |
| get 40% off | get 50% off | 2 |
| get nitro | get basic | 2 |

Source: paywall CTA experiment corpus.[4] The pattern: swaps move between generic continuation verbs ('continue') and explicit value or offer language ('try for free', 'upgrade', 'renew now').

## A named example

Calendly changed its Standard plan CTA from 'Try for free' to 'Get started' while keeping 'Try for free' — and the page's only bright-blue primary button — on Teams.[5] Inferred rationale: reserving trial language and the highest-contrast button for the higher-ARPU Teams plan steers volume there. This is copy plus visual hierarchy working together.

## Caveats

These are detected UI diffs with inferred rationale, not measured conversion results.[6] The 561 count is 'experiments that changed CTA copy' by keyword co-occurrence, so it includes label tweaks alongside meaningful rewrites. Use the swap table as a menu of what peers try, not a ranked winners list.

## The numbers

| Stat | Computed from |
| --- | --- |
| 561 CTA-copy experiments across 201 companies | tactic_mentions__cta_copy |
| 4,814 total detected experiments (CTA copy = 12%) | total_detected_experiments; 561/4814 |
| 344 of 795 paywall diffs (43%) changed the primary CTA, 146 companies | paywall_cta_experiment_corpus |
| Top swaps: continue→try for free (3), continue with core→upgrade (3), restart subscription→complete subscription (3), get youtube premium→renew now (3), get 40% off→get 50% off (2), get nitro→get basic (2) | qualitative: CTA copy changes (paywall corpus), 2026-07-07 pull |
| Calendly changed Standard 'Try for free'→'Get started', kept it on Teams | qualitative: CTA copy changes, calendly, 2026-07-04 |
| 0 measured conversion outcomes (learnings are LLM-inferred) | family universe note |

## Methodology

Universe: 4,814 detected experiments (362 companies) plus a companion 795-diff paywall CTA corpus (146 companies), from Lazyweb's ~800-app mobile corpus. CTA-copy tagged by keyword co-occurrence; swaps are observed before/after strings. Detected diffs with inferred rationale, not measured results. July 2026 pull.

## Sources & citations

- [1] Lazyweb Research analysis of 4,814 detected experiments (362 companies, mobile-app corpus), July 2026. 561 experiments where (cta OR button) AND (copy/text/label/wording/reword) appear, across 201 companies.
- [2] Lazyweb Research analysis of 4,814 detected experiments (362 companies, mobile-app corpus), July 2026. 561 of 4,814 total detected experiments (12%).
- [3] Lazyweb Research analysis of 795 detected paywall CTA experiments (146 companies), July 2026. 344 of 795 paywall diffs changed the primary CTA text.
- [4] Lazyweb Research analysis of 795 detected paywall CTA experiments (146 companies), July 2026. Top repeated before→after CTA swaps from the paywall corpus, July 2026 pull.
- [5] Lazyweb Research analysis of 4,814 detected experiments (362 companies, mobile-app corpus), July 2026. Calendly (2026-07-04): 'Try for free'→'Get started' on Standard; 'Try for free' retained on Teams.
- [6] Lazyweb Research analysis of 4,814 detected experiments (362 companies, mobile-app corpus), July 2026. Learnings are LLM-inferred hypotheses, never measured A/B outcomes.

## Related questions

- [Do Companies That Add Discount Badges to Their Paywall Keep Them?](https://www.lazyweb.com/research/do-companies-keep-discount-badges-after-testing-them)
- [Which Page Areas Get Tested Most — And Where Does the Hero Rank?](https://www.lazyweb.com/research/how-many-experiments-touch-hero-media-swaps)
- [Do Companies Remove Pricing Tiers — and How Often Is It Tested?](https://www.lazyweb.com/research/how-many-experiments-touch-pricing-tier-removal)