Sway AI · B2C · dating profile feedback app
Open onboarding with short videos explaining the app before asking any questions
Added after emails asking what the app does; no numbers given
What they did
What happened
- Stage
- monetization
In their words
Daniel Heintzma - I speedran a new app from 0 to $2m/year in 6 months (here's how)
Daniel Heintzman… the I actually like I can share like some like the onboarding flow, as well as like the paywall design maybe. So yeah, this is like the onboarding. The way I think about onboarding is like it needs to like The goal of onboarding before a paywall is to like raise your motivation and increase like your intention to purchase the app. There's like a lot of visuals explaining what the app does, and then questions to personalize the app experience.
Daniel HeintzmanStarted by like clarifying like why the user's there. It's I feel like it's very similar to like a sales funnel where like you clarify like why they're there, explain and label their problem, and then sell the dream result. Like you'll see like the dream result of like what other members are experiencing, um personalizing through the questions, and then like as well as explaining why this app is for them. And then we show them the dream result also on the paywall. And so this was like the first version of the paywall, where you can
Daniel Heintzmanafter you go through that onboarding, you would be able to see your action plan now. And if you like tap Continue, you get to unlock the entire app. And that was like the first version of the paywall, and just wanted like to convey the feeling that you get a lot of value from this subscription. So you see it's like a carousel of like eight slides of all the things you get as part of your subscription. And so that was one thing that we had did with the paywall. I think the other unique thing that I think is similar to the way like Cal AI thinks about it is
Daniel Heintzmanthe onboarding is like very sequential and like a one-flow process where the user doesn't really need to think, but they're like guided on this like one-path journey towards the paywall. And then after the paywall, you show them like the aha moment of the app. And so like which is getting the profile rated. And so they would go through this process of like uploading screenshots of their profile. And then it would analyze it. And then similarly to …
Joseph Choi… the psychology a little bit again? Uh it was really interesting to hear you kind of describe the journey of like, you know, showing the dream outcome, like explaining what the app is, and that whole sequence of questions. Like is there some sort of method to the way that you order the screens? Like it looks like right here there's like you just describe the app with really clear images. Is there a a method to kind of like the ordering of this?
Daniel HeintzmanYeah, I think this is a something that I had learned a lot when it comes to like we got a lot of feedback in written emails of people We didn't have this initially. So, initially we would just go straight into the questions. And so I think this is a lot a lot of what I see on other apps doing is they they just look at I think like Cal AI and it just goes straight into the questions. But I think it misses the whole point, and what we would get a lot of questions about is like, "What does this app even do? Like what am I even subscribing for?" These are actually videos, but they're not playing right now, but
Daniel Heintzmanthey're videos that explain the problem, as well as explaining how the app solves that problem. And then once people like have that context, then asking questions makes sense, right? It's like because otherwise you're just you're asking someone questions, and they have no context why they're giving you this information.
Joseph ChoiYou're You're giving a lot of information and value even before they have to ask uh answer any questions, and then in between your questions, you're also giving continuously giving more value as well. Is there a Is there an ordering to the to the types of questions that you ask? Is it basically just problem problem questions and then solution questions?
Daniel HeintzmanI think less about the ordering, and more about the simplicity of how easy it is to come up with an answer. So I think that's why a lot of these um companies like will offer a uh like multiple choice, just being that it's so easy to for the user to like quickly tap one and give an answer, because I have seen other apps where like they ask these questions that like are actually like kind of hard for you to come up with an answer for. The whole point of onboarding is you don't want to slow people down. You want to get them ramped up and get them going before the paywall.
Joseph ChoiThis is really helpful to to see the whole onboarding and everything. It It's I can I can see how a lot of this translates to Cal AI too, with like making the the onboarding smooth and making it high converting. Um but then I want to like dig into that story now. Like how did you how did you get into Cal AI? Like how did you meet the team and um and get started working with them?
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
- Grow an Instagram account on Hinge prompt tips before building, with a preorder linkSome videos took off within weeks; audience existed before launch
- Charge followers $15 by DM for a manual profile review before buildingPeople paid before any app existed; proved willingness to pay
- Copy the text of a men's self-improvement account's most viral slideshows into your own designsSlideshows averaged 10-50K views; one hit 600K
- Make a month of TikTok slideshows in 1-2 hours from a spreadsheet and templateSeconds per slideshow; a month of posts in an hour or two
- Post unbranded slideshows until views come, then brand every slide with the appOne branded Christmas Day post got ~4M views and $5-6K revenue in a day
- Have one dating influencer make promo videos, then run them as ads from his pageTook revenue from $30-40K/month to about $2M a year; creative supply became the limit
- Ask for profile screenshots after the paywall, not before itSimple onboarding before the paywall performed a lot better; no numbers
- Sell extra analysis credits to people who rerun the profile checkAbout 11% of revenue
- Sell AI photo packs at $14.99 for one and $49.99 for tenBiggest upsell in the app; no spending cap lets big spenders pay more
- Preselect the yearly plan on the paywall to collect cash upfrontWeekly plan also tried; grew a lot slower
- Redesign the meal logging screens so people keep paying after the 3-day trial
- Let users join friend groups and topic groups with streak leaderboards
- Add daily logging streaks, copied from Duolingo
- Let users post progress photos, meals and streak badges to Instagram storiesA lot of users share progress and meals; no numbers
- Make the 3 screens creators film (camera, loading, food detail) instantly clear
- Grow through fitness creators who sell coaching, so the app doesn't compete with themMain growth driver; guaranteed views vs hoping to go viral
- Ramp up paid ads once influencer growth is workingMarked Cal AI's big growth inflection point; no numbers
- Launch in a month with game-like branding unlike bank-style crypto appsWent really viral; idea to launch in about a month
Tags: ios, sales-funnel-onboarding, explainer-video, multiple-choice, dream-outcome