Weather Up · B2C · weather app with widgets
Pre-select the yearly plan on the paywall, not the monthly one
Monthly default got ~10% more subscribers but 60% lower revenue per user in month one
Workedpaywallexperimentation
What they did
A/B test (50/50) with RevenueCat Experiments scheduled to start at 12:01 AM on launch day: arm A pre-selected the $40/year plan, arm B the $4/month plan; nothing else changed. Annual default: ~800 annual, ~400 monthly. Monthly default: ~1,100 monthly, ~100 annual. Judge on realized LTV and revisit after the first annual renewals (day 366).
What happened
Monthly default converted ~10% more (first stated as 4.5%, called statistically significant) but realized LTV was 60% lower under a month in; with >30% of monthly subs already off auto-renew, the curves are unlikely to cross before year one. Launch-press cohort may differ from later search traffic.
- Stage
- launch
“And so, the test I did was defaulting to annual being the pre-selected option on the paywall versus monthly being the pre-selected option on the paywall.”
“realized LTV is 60% lower on the people who saw monthly.”
“10% more people became subscribers when they saw $4 versus seeing 40.”
Related topics
Get tactics like this every Monday
New tactics from the week's founder interviews, each linked to where it was said.
More from Weather Up
All 12 Weather Up tactics- Ship a big update that gets featured by Apple and the press~$100K in the update year with almost no ads; halved the next year without updates
- When someone leaves a hard paywall, send a push offering 7 free daysFew took it and few of those paid; being replaced
- Pull Google AdSense ads and move to a hard paywall with a 7-day trialAds caused bad reviews and complaints; app lost money for ~a year without them
- Let people close the paywall, give them 7 free days, then lock the app
More from this episode
- Weather Up: Pitch journalists with a short demo video in a Twitter DMCovered by TechCrunch, 9to5Mac, MacRumors and BGR, even by writers who rarely cover apps
- Weather Up: Hold new features back for one big launch to get attentionDownloads fell to ~15 a day during the 4-year wait; guest calls it a big regret
- Weather Up: Launch just before Vision Pro ships to avoid 'where's your Vision Pro app?'Landed an hour before Apple's DMA news; fewer articles, no social buzz, no Apple feature
- Weather Up: Launch a weather app at $40 a year with no launch saleClose to 2,000 subscribers, but press readers balked at $40 next to a $30 competitor
- Weather Up: Relaunch with one or two new features and a limited-time sale
- Weather Up: Put a feature demo video on the paywall, with hosted price buttons belowWent better than expected; built in under a day; guest gives no numbers
- Weather Up: Test $40 against $39.99 for the yearly plan
- Build a mirror app and rank for the word 'mirror' in the App StoreCaptured 6-8K of 10-20K daily 'mirror' searches at peak
- Make a $0.99 app free and sell themes plus a $10 bundleMany bought the $10 bundle; earned $10 a sale instead of $0.99
- Contrast: Buy ads for paid-upfront appsLost money; spent thousands of dollars in the early App Store
- Contrast: Give developers a revenue share instead of paying them by the hourStopped losing money on stalled projects; partners working on the side ship slowly
- Launch Center Pro: Save features for big paid-app launches that spike revenueUp to $60K in a launch week vs hundreds of dollars a day otherwise
- Dark Noise: Add interactive widgets when iOS 17 ships to land in Apple's listsPicked first in an editorial list; 'massively impactful', no numbers; other lists did little
- RevenueCat: Launch again and again, dressing up each fundraise as a launchFirst Reddit post got some traction; no numbers