Do Apps Add or Remove Trial Wording From the Paywall Button When They Rewrite It?
- By Ali Abouelatta
- n = 304
- Published
- Updated
- 2 min read
- paywall · trials · monetization · experiments · ux-patterns · mobile
Short answer
Among 304 detected paywall buttons whose copy was rewritten, 31 added 'trial' or 'free' wording that wasn't there before, while 24 removed it — so adds outnumber removes, but only modestly (Lazyweb Research).source[1] Another 54 kept trial/free wording on both sides.source[1] These are detected UI diffs with inferred rationale, not measured A/B lift.
Of 304 rewritten paywall buttons, 31 added 'trial'/'free' wording and 24 removed it — a net tilt toward naming the trial (Lazyweb Research, July 2026).
2×2
How 630 paywalls combine offer framing × price-in-button
Most paywalls (70%) are direct with no price in the button; putting a price in the button is the exception in every quadrant. Source: Lazyweb Research, 630 paywalls.
← Direct (no trial)Trial-led →
- top-left: Trial-led · no price
- 108 (17%)
- top-right: Trial-led + price
- 20 (3%)
- bottom-left: Direct · no price
- 443 (70%)
- bottom-right: Direct + price
- 59 (9%)
← No price in buttonPrice in button →
Bar chart · Rewrites (of 304)
Rewrites (of 304) — Trial wording on the button
One block is about 1.9 Rewrites (of 304)
- Trial/free added (absent → present)
- 31
- Trial/free removed (present → absent)
- 24
- Trial/free on both sides
- 54
The finding
Among the 304 rewritten paywall buttons, trial-word moves lean slightly toward adding: 31 rewrites introduced 'trial' or 'free' wording that was absent before (22 companies), while 24 removed it (18 companies).source[1] A further 54 buttons kept trial/free wording on both sides.source[1] The net direction favors naming the trial in the button, but 24 apps went the other way, so it is a tilt, not a rule.
Trial wording on the button
| Button trial-word state | Rewrites (of 304) |
|---|---|
| Trial/free added (absent → present) | 31 |
| Trial/free removed (present → absent) | 24 |
| Trial/free on both sides | 54 |
Adds edge out removes 31 to 24. The presence of 54 'both sides' rows shows many apps iterate the button around an already-trial-framed CTA rather than adding or dropping the trial word itself.
Observed examples both directions
Adding: SoundCloud changed 'Continue' to 'Try free', naming the trial in the button.source[2] Removing: Headspace moved 'Start my free trial' to 'Start subscription', dropping the trial word to make the offer feel like a committed plan while still reducing surprise-billing risk via a shorter reminder window.number[3] Both directions have coherent inferred rationales — which is why the corpus shows a split, not a consensus.
How to apply it and caveats
If commitment anxiety is your main friction, the corpus leans toward naming the free trial (as SoundCloud did); if you want the plan to feel committed, dropping the trial word (as Headspace did) is a real, observed alternative.source[2]number[3] The 304 base includes some annotation-noise rows, and trial detection is a regex for 'trial'/'free', so borderline phrasings may be miscounted.number[4] Every row is a detected UI diff with inferred rationale, never measured A/B lift.
Real in-market screens
Screens from real, in-market apps, tracked by Lazyweb Research.
Kahoot1290 × 2796opens the full image 
Habits1206 × 2622opens the full image 
Nyt Games1206 × 2622opens the full image 
Citizen1290 × 2796opens the full image
The numbers
| # | Stat | Computed from |
|---|---|---|
| [1] | Trial/free added 31 (22 cos), removed 24 (18 cos), both sides 54; base 304 | trial_word_added_to_button |
| [2] | SoundCloud 'Continue' → 'Try free' | qualitative.CTA verb.soundcloud |
| [3] | Headspace 'Start my free trial' → 'Start subscription' (12-day → 7-day reminder) | qualitative.trial length wording.headspace |
| [4] | Some of the 304 rows are annotation noise | smallSampleWarnings.data_quality |
Methodology
Universe: 304 detected paywall buttons whose copy actually changed, from 795 detected before/after changes across 146 companies in Lazyweb's ~800-tracked-app mobile corpus, July 2026. Trial/free classified by regex; some rows are annotation noise. Detected UI diffs with inferred rationale, never measured A/B lift.
Sources
- [1]
Lazyweb Research analysis of 304 rewritten paywall button texts (within 795 detected changes, 146 companies), July 2026
Base = rows where cta_text_before and cta_text_after differ. Trial/free markers detected by regex; add=31, remove=24, both=54.
- [2]
Lazyweb Research analysis of 795 detected paywall CTA changes (146 companies), July 2026
Named examples SoundCloud and Headspace from qualitative sample. Detected UI diffs with inferred rationale, not measured A/B outcomes.
Source: Lazyweb Research — proprietary analysis of real, in-market app screens. Cite as Lazyweb Research, 2026-07-07.
Explore the evidence
- Full 45-frame August 2026 capture
- Three captured screens · August 2026