Hily · B2C · dating app
Launch a test every day and review yesterday's results together each morning
Credited with finding the winning channel and reaching positive unit economics
Workedexperimentationteam-process
What they did
In the early struggle, launched something every day and met the next day to review results and decide next steps. Tests covered onboarding, paywalls, ad channels, UGC and creatives. Now: analytics on almost everything; every A/B test compares control vs test on many core and 'healthy' guardrail metrics.
What happened
Guest credits it with surviving, finding Snapchat and lifting LTV to positive unit economics.
- Stage
- pre product-market fit through today
“we were experimenting literally every day. Like, every day we were launching something, the next day we were together like this had a discussion”
“When we do some tests, we compare to different kind of uh metrics. We have a lot of core metrics and like uh healthy metrics that we look at to compare all of those results.”
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 Hily
All 24 Hily tacticsMore from this episode
- Hily: Launch a dating app in California only, then add Texas, to keep users closeCAC ran about 3x LTV for 6-9 months; didn't fix economics alone
- Hily: Let people into the app only after they pay
- Hily: Put most ad spend on Snapchat with creator videos made for SnapchatMuch lower CAC than other channels; 1-1.5M downloads a month until rivals copied
- Hily: Use a long quiz-style onboarding that personalizes the app, then sellGuest says it still works eight years later; lifted early LTV
- Hily: Ask dating users to sign up first instead of pitching or letting them browsePitch screens and account-free browsing both converted worse than plain signup
- Hily: Add short 'Nice to meet you' and 'Awesome' cheers between onboarding stepsRaised the share of users finishing onboarding; no numbers
- Hily: Add livestreaming to a dating app after paid growth stallsWeak traction and lost focus; dropped after about 1-1.5 years
- Hily: Cut side features and focus everything on better matches and dates20-40% yearly growth in daily users and revenue; top 3 US dating downloads
- Hily: Sell a cheap one-time 'be seen 10x more' boost right after onboardingWorked better than folding the boost into the subscription; no numbers
- Hily: Show a 'visibility: low' badge on the home screen that opens a boost offerGuest says intent-matched offers convert much better than generic 'buy' screens
- Hily: Change the app icon now and then for seasons and holidaysBrings extra organic installs and spikes in returning users each time
- Hily: Draw a notification badge on the app icon (no longer allowed)Raised conversion 'like crazy'; now prohibited by the stores
- Hily: Add invite-a-friend and sharing features to a dating appDidn't work; people won't share dating apps via links
- Hily: Work on when and how you ask for push permission, and explain the benefitHigher opt-in drove retention, guest says; no numbers
- Hily: Pause pushes after 4 ignored in a row; tailor over 30 push rules
- Hily: Send pushes about real events like 'someone liked you', not generic come-back offersGeneric promotional pushes did 4-5x worse than real-event pushes
- Hily: Show lapsed users to more people so real likes bring them backGuest says matching tweaks drive retention; no numbers
- Hily: Add a Discover tab of themed profile collections and push them daily
- Hily: Hide extra profile details until users add their own photos and infoGuest says it works fine; no numbers
- Hily: Make separate store pages for each ad channel, gender and search keyword
- Hily: Ask for a rating only after someone sends a like or a messageWorks well for them, guest says; no numbers
- Run Google Play in-app events to show up twice in search resultsGot a second search placement besides the organic listing