Product experiment research by Ali Abouelatta

Lazyweb Research

Short Q/A issues on paywalls, signup, onboarding, retention, pricing, and product experiments. Built for humans to skim and agents to cite.

Issues: 21
Questions: 582
Categories: 35

By Ali Abouelatta
Data as of July 2026

Read as Markdown

Side-by-side Uber Eats paywall comparison with the control grid on the left and the experiment benefit list on the right.

Featured issue · Jun 3, 2026

How did Uber Eats test a benefit list against a benefit grid in its subscription paywall?

Uber Eats tested a denser Uber One benefit list against a four-card grid, while also changing the free-period label, adding a monthly savings claim, and renaming the CTA to Join Uber One. The screenshots show the observed design change, not a measured winner.

monetization · paywall · subscription · conversion

Read the issue

35 categories · 582 questions

Browse questions by category

20 more issues

Every issue

Nov 21, 2022
How did Duolingo use streak goals to improve retention?
Duolingo added a streak-goal choice immediately after users started a new streak. The team did not store or use the chosen value, but the act of choosing created commitment; Ali reported that users retained better weeks after seeing it, and a later version with no preselected option produced stronger goal selection without causing drop-off.
Feb 6, 2023
Which Duolingo retention experiments made habit loops stronger?
Ali identified four Duolingo retention experiments from 2022: widgets that change mood when you skip lessons, special streak app icons, goal-commitment copy, and Perfect Streak Week. Each experiment made progress, identity, or commitment more visible at the moment users were likely to act.
Jan 29, 2024
When does positive friction improve retention or conversion?
Positive friction works when the added effort also increases perceived value, motivation, or commitment. Ali's examples include Duolingo streak goals, long wellness onboarding flows, and purchase flows that add explicit choice instead of hiding the decision.
Mar 1, 2024
How are top-grossing apps testing price increases?
Ali's top-100 grossing app review found that about 20% tested a price increase over six months, with an average increase of about 27%. Around a third of those tests pushed harder toward annual plans, while about 10% of observed pricing experiments were not successful.
Apr 5, 2024
Why do registration paywalls look simpler than in-app paywalls?
Registration paywalls often show less feature and benefit detail than later in-app paywalls because they appear before the user has full product context. Ali observed this pattern while reviewing more than 100 onboarding flows, especially in apps with multiple subscription tiers.
Jun 13, 2023
What can product teams learn from Netflix landing page A/B tests?
Netflix's landing-page tests show three durable lessons: revisit old hypotheses when the context changes, isolate variables even after a winner, and avoid highlighting too many things at once. Ali's review covers seven experiments across timeline reminders, CTA copy, email input, free-trial removal, input contrast, and ad-tier messaging.
Aug 2, 2024
Why should sign-up screens be boring?
Ali's review of Ring, AllTrails, Nextdoor, and Spotify signup redesigns found the functional, plain version won in each case. The pattern is that signup is a job-to-be-done moment: decorative videos, illustrations, and animation can distract from the CTA and can add download or bandwidth cost.
Aug 2, 2024
What did Ring learn from testing a boring sign-up screen?
Ring's functional signup screen beat a more delightful signup experience with a neighborhood background video. The winning version stripped the screen back to logo, promise, create-account CTA, sign-in link, and support copy, reducing competition around the primary action.
Aug 2, 2024
What did AllTrails learn from testing a boring sign-up screen?
AllTrails' simpler signup screen beat a version with a scenic hiking-trail background video. The winning screen kept the brand mark, direct signup/login language, social login buttons, and email continuation without letting the outdoor imagery compete with the auth choices.
Aug 2, 2024
What did Nextdoor learn from testing a boring sign-up screen?
Nextdoor's plain white signup screen beat a background-video version built around lively neighborhood footage. The winning version kept the neighborhood promise but removed the moving scene, making the sign-in methods and invite-code path more prominent.
Aug 2, 2024
Can a sign-up illustration reduce logins and registrations?
Yes. Ali's Spotify example shows that even adding an illustration to an otherwise plain signup screen lost against the simpler control. The change did not add a new decision, but it added visual competition around the same signup, social-login, and login actions.
Jun 2, 2026
How should apps show limited-time subscription discounts in paywalls?
Make the discount impossible to miss, then guide users toward the intended plan. Snapchat changed the paywall headline to a 60% savings claim, added multiple discounted plan cards, marked annual as BEST VALUE, and put a SPECIAL OFFER badge on the purchase surface.
Jun 3, 2026
Should AI subscription paywalls use comparison tables to explain free versus paid plans?
Yes when the upgrade is feature-dense. ChatGPT moved from a benefit checklist to a Free-versus-Plus comparison table, making the paid gap explicit and putting price inside the primary upgrade CTA while keeping a separate web-purchase link.
Jun 3, 2026
Should subscription paywalls reduce copy density around free trials?
Often yes. Strava reduced paywall density by shifting from a benefit-led annual trial card to a broader first-month-free message and a shorter trial timeline, keeping the first decision focused on immediate access instead of feature inventory.
Jun 2, 2026
Is texting clearer than chatting for mental health subscription features?
Texting can be clearer when the value depends on concrete product mechanics. Headspace replaced a chat-with-coach hero and live coach bubble with a text-coaching headline and product-style illustration, making the paid coaching add-on easier to understand.
Jun 3, 2026
Should free-trial CTAs describe the user goal instead of the trial?
Yes when the product goal is more motivating than the billing mechanic. Apple Fitness moved from trial-first messaging to workout-first messaging and changed the primary CTA from trying the trial to starting a workout, while keeping the annual upsell available.
Jun 2, 2026
Where should social proof appear on a subscription paywall?
Put social proof next to the premium benefit it supports. Speechify paired advanced AI voices with voice-avatar imagery and a weekly-starts proof bar, making the offer feel both differentiated and widely adopted on the same decision surface.
Jun 2, 2026
Should mobile apps offer a cheaper web checkout CTA?
A cheaper web checkout CTA can work when the savings are simple enough to understand. SoundCloud tested a mobile paywall that included a web-savings option, then moved to a more price-first layout that traded extra discount routing for faster comprehension.
Jun 2, 2026
Does urgency copy like Accept Now improve subscription offer CTAs?
Urgency copy helps when the offer itself is concrete. Apple Music moved from broad catalogue-and-content messaging to a three-months-for-the-price-of-one intro offer and used acceptance-language CTA copy, making the economic upside easier to act on.
Jun 3, 2026
Should subscription upsells explain premium features before showing price?
Yes when the feature is the reason to upgrade. YouTube changed an ad-free detail screen from generic uninterrupted-play copy to an explicit Premium-exclusive frame with shorter supporting copy, tying the benefit more directly to the paid plan.
35 categories582 questions, answered with dataBrowse by category

How did Uber Eats test a benefit list against a benefit grid in its subscription paywall?

Browse questions by category

Every issue

How did Duolingo use streak goals to improve retention?

Which Duolingo retention experiments made habit loops stronger?

When does positive friction improve retention or conversion?

How are top-grossing apps testing price increases?

Why do registration paywalls look simpler than in-app paywalls?

What can product teams learn from Netflix landing page A/B tests?

Why should sign-up screens be boring?

What did Ring learn from testing a boring sign-up screen?

What did AllTrails learn from testing a boring sign-up screen?

What did Nextdoor learn from testing a boring sign-up screen?

Can a sign-up illustration reduce logins and registrations?

How should apps show limited-time subscription discounts in paywalls?

Should AI subscription paywalls use comparison tables to explain free versus paid plans?

Should subscription paywalls reduce copy density around free trials?

Is texting clearer than chatting for mental health subscription features?

Should free-trial CTAs describe the user goal instead of the trial?

Where should social proof appear on a subscription paywall?

Should mobile apps offer a cheaper web checkout CTA?

Does urgency copy like Accept Now improve subscription offer CTAs?

Should subscription upsells explain premium features before showing price?