A consumer app in stealth · B2C · guest's newest app
Test the same video idea at 10, 25 and 30 seconds with different hooks
Unknownorganic-short-videoexperimentation
What they did
For one content format she tests a short script (~10s), a medium one (~25s) and a longer one (~30s), plus variations in hook, on-screen text, filming style and emotional delivery, since these can decide whether a video goes viral.
“I would test out, you know, a shorten script, so it's 10 minutes long, a 10 seconds long, a medium script that's 25 min- 25 seconds long, a longer one that's 30. So I will test out different things like the length, the hook, the text, the the way you film it, the the emotions that go into it”
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 this episode
- GlamUp: Swap text makeup tips for a face scan showing your before and afterWent viral on TikTok; app hit 1M users in 6 months
- GlamUp: Push a text-only feature through influencers, faceless accounts, own videos, Reddit and DiscordDidn't work over 3-4 months; the feature itself was the problem
- Mock up a new feature in Figma, have creators post it, build it only if it goes viralFounder's default way to test features now; no numbers
- Pay creators $10-30 a post once it hits 500 views, plus bonuses at 10K-50K+ viewsThe standard deal across consumer apps; no results given
- Recruit creators from Discord groups, friends' referrals, and small influencers averaging ~10K views
- Have creators open new accounts and warm them up for 3 days by following small niche accounts
- GlamUp: Interview every creator yourself and accept about half or fewer
- Give every creator a 30-60 minute video course and a bank of ready scripts
- GlamUp: Make the videos yourself until one goes viral, then hire creators to copy itFounder says it's how every app started; no numbers
- Copy formats from tiny accounts that got one 1M-view video, not from influencers
- GlamUp: Ask 10+ onboarding questions about the user's goals so the app feels made for themFounder says it primes users to pay; no numbers
- GlamUp: After each answer, show a stat like '80% of our users have acne too'
- Sprout: Make people pay before they can swipe right to apply for a jobFounder says it works well; app reached $250K MRR by month 8
- GlamUp: Raise prices by 20-40% and check whether more people payOften converts better, founder says; no exact numbers
- Re-test the paywall and prices every couple of months
- Delete features and improve the few that matter
- GlamUp: Start with $10-20K, skip paid ads, and put all revenue back into apps99% of growth came from organic; funded later apps
- Pick a niche with lots of TikTok content but few apps, some making $50-100K a month
- Spend day one researching the niche and day two sketching the screen meant to go viral
- Focus on making the product good so people tell friends and familyFriend-and-family users convert about 2x better
- Test $9.99 against $6.99 a week, with a $50 yearly plan$9.99 slightly ahead on small numbers; more people pick yearly
- Lower the price after no one subscribesStill zero subscribers; hosts blame the weak paywall, not the price