Show HN: I spent 4 months building Duolingo but for your life
three-cells.comSome history: A couple of days ago my app finally got approved and released on app store.
Overall really happy with the release. Now I finally have something that I use every single day. Whereas previously I was using various different apps to journal, track habits, track my weight and manage tasks etc. Now I have all of that in one place.
Initially it was a website that I built for myself but I realised that something like this is better built in an app after speaking with lots of people.
I used Convex dev for my backend and that honestly made the backend part of creating this part very easy because I could just use the same structure I had for the web app. Convex was my choice because even for a website I wanted the data to always be in sync whether I access it on my phone or laptop.
The most annoying part about building an app is having to go through the review process of app store. For a website, I can just make a change and it can be live in less than a minute. But the app store review process alone took me 2 weeks to release this.
Using AI made migrating the website into an app really easy. I had some components like heat-maps and graphs which would have been really difficult to migrate over if I was doing this a couple of years back.
The idea itself actually came from reading lots of productivity books and then stumbling upon an interview from Jim Collins who talks about how he tracks his own life and makes sure it's going in a direction that he wants
Would love to see less front loading on the registration side - I fell off onboarding because I couldn’t get through the 12(!) page questionnaire.
The value proposition is clear, just let me use the app. Notes (my current solution for this) doesn’t make me read summaries of other people’s research every time I open the app :-)
An alternative could be something like "Make the most of your days".
I find the duolingo course very slow paced, gems and gamifications are just bad.
And of course there are competitors, many of them. There are also many free language learning resources. But, you did not said which language you are learning and whether you are beginner or not. And most of free resources are made for specific language.
I've bought a few courses from various places, but I want bite size and daily learning.
I'm getting a lot out of Anki with premade decks these days, combined with watching tons of video content.
It is very different from duolingo though. No gamification, only two types of cards (reading and writing) every card has basically all the information about that kanji available from the dictionary. Content is sourced directly and unaltered from a couple of open sourced dictionary, so no AI or content writing either.
I built it for myself as I wanted to practice writing in my target languages, while also wanting to learn new words... The idea was that hopefully I would remember the words if I could associate them to my journal for that day.
It's a little clunky, but give it a go if you're interested!
Right now it's a bring-your-own claude token model, but let me know if you're not comfortable with that.
It's stored in a VPS hosted convex backend. I'm currently building functionality to export data :)
I was also keen to see how the whole system fits into my life before I paid, and since it take 66 days ;) I thought it would be nice to see if it actually worked before I decided to pay. Just a thought. I almost made it all the way through!
Curious how you used AI to migrate the website into an app. Could you share more about that process?
Great job -- shipping something is always exciting and doing so in such a short timeframe is something to be proud of.
This is true for every app with IAP. It’s how I typically decide if I’ll download an app with IAP.
But I'm not even going to bother looking if an app can't clearly advertise its price.
It just feels like a dark pattern, like they're intentionally trying to trick me from the very start.
Maybe the price will turn out to be fine, but it's sure not building trust from the start.
> Three Cells Weekly $2.99
> Lifetime Access to Three Cells $29.99
> Three Cells Weekly $2.99
The ratio here feels a bit off, $155.48/y or $29.99 for life screams "trying to push you to pay a lot for an app I know you won't stick with" rather than "we've got great value options for any type of user". There is a 7 day trial as well, so even more confusing as to which kind of user is supposed to want the weekly payment.
Duolingo is just another mobile game, but pretending to be a learning app.
They are also easy to build - you need one primitive: basic text storage, although scheduled push notifications are great to have. No need to sync stuff, no sharing/permission model, no scale issues you can’t solve with a b-tree SQLite index.
I think another factor is an increase in productivity-lifestyle content influencers, the sort of people who talk about Notion on TikTok. Speaking of Notion there’s like a zillion user-created habit tracker templates for Notion too. I work at Notion but don’t use it daily outside of work.
I agree. I've been tempted lately to write my own local todo + notes + calendar app that fits the way I think about tasks and time. Kind of like developing a software glove for ones mental model. It's no wonder there are so many "gloves" in this space, everyone's model is unique.
100% this. They fall into the same camp as "self-help books", "life coaches", and to certain extent "spiritual gurus".
There are only a limited number of game mechanics, people install for the theme and come back because of it.
You can take a 1:1 copy of a game, reskin it 12 different ways and 1 of those might be a hit.
For apps like this there aren't that many ways to track what you're doing, but people like different kinds of UX.
Usually what happens is a lot of other people in the world also feel the same way as me and if they like how I have approached the app then they would download it and use it and I think that's how you get a lot of different types of habit trackers coming up all the time
That said, I bounced off at the pricing. The $30 lifetime price isn’t something I find inherently too expensive, but I need to see if the app works for me before committing to it. It was weird that if I went forward with the free trial it would automatically put me on the exorbitant $3/week price. That option was repellent and got me worried about forgetting to either cancel or make the purchase. Compounding the issue was uncertainty about whether I even _could_ make the lifetime purchase after accepting the free trial.
Then I lost momentum and started thinking about how I was about to drop $30 on an app that’s just some HN poster’s 4-month project, and I have no clue how crippled it will be if (when) you decide to shut down the API.
If you’re confident the app itself is habit-forming, I’d recommend just letting people use it for a couple weeks and then hitting them with the paywall. And when you’re asking for that kind of money and using the word “lifetime”, I’d describe how you’re going to guarantee that to the user, even if they’re the only person who ended up buying your app.
Edit: Now I’m stuck on the payment options screen with no way to delete my account. Not happy about that.
But I also would have bounced on the "free trial, auto payment after" I understand the thinking around them, but for me it's just not a pattern im ever going to opt into. It feel predatory, you will probably forget and then ill get some amount of "months" off you. Like gym memberships.
I would be much more likely to convert if the "trail" as months maybe years instead of days or weeks (this is a lifestyle thing after all, if customers find it genuinely useful and buy into the system for months then $30 a year is nothing)
OR
free-trial followed by locked out. No "automatically start billing"
- I will add an option for the user to delete their account upon hitting the paywall if they don't want to continue. - I didn't want the app to be free to begin with as it doesn't attract serious users and also because I'm an indie maker and free users is not something that I can ultimately afford at this current stage. - You're right, I should have a way for the user to trial the app and then pay once instead of the trial being on the weekly subscription only.
As for guaranteeing lifetime access, a lot of web based products offer lifetime access and I guess it's just a matter of trusting the maker if they will support it. For my particular app I know that I've been involved in the productivity space for quite some years and only now making an app that suits my needs. I imagine myself using this for all the years to come and if I stop using it, it's self hosted on a server I own and I will keep it live forever. If the user doesn't trust that then that's completely fair and fine. No issues with that.
Lots of useful feedback, thanks again for the write up! Still building and learning and trying to be as genuine as possible.
Unfortunately I'm not leaving Android for iOS, so I wish you success and eventual expansion to Android.
It isn’t a bad deal at $30 but it’s just enough for me to go “oh” and not really open the app/push through.
2 more comments available on Hacker News