Preparing Hounty for launch on mobile stores

It was a busy week and it feels a lot better now that most of the legal stuff is sorted, but damn it was so much more work than I anticipated 😅.

Sole trader registration was the easiest of all; at least here in Sweden it can be done fully online. The hardest part was to estimate my income for this year: how would I possibly now?!

In the end I just calculated based on how much preliminary taxes I am ready to pay (I'm thinking of it as a weird way of saving up, since all overpaid taxes always come back around here automatically).

I got approved almost the next day - super happy about it! There’s a possibility to apply for trademark or a business name but I decided to keep using my name and spare extra expense. Registration as a sole trader itself cost nothing.

I did however talk with a "new business center" representative and got some informational support. For instance, they recommended to open a separate bank account to make it easier to track business related expenses - which I did.

Speaking of expenses

TL;DR Ability to publish to mobile stores will cost you $124 & it takes several days to get approved even after you paid.

I was aware of Apple development program costs quite a bit - it’s $99 a year, but what I didn’t know was that it can take several days or even weeks to get access after you pay. So I was quite anxious about it, even though there isn’t much to release just yet. I got my developer account within 72 hours or so in the end.

As for Google Developer account, I had one, but I needed to re-verify it, since it was old - I opened it back in Russia. I got verified in a couple of days, but discovered that my payment profile is missing (I might have deleted it myself) and the only way to get a new one is to open a new account 🤦‍♀️

Ugh, $25 one time fee. I was more annoyed that I need to submit all the documents for verification again rather than anything else.

I am, however, happy, that I did it all now - would've been so much worse if the app was ready and I wouldn't have been able to release it.

But enough about money, let’s talk about something what matters much more 🫶

User interviews

Thank you so much to everyone who contacted me and let me interview you! It was a great experience and I learned so much from you!

Firstly, I got more confidence that the problem I'm trying to solve is real: managing chores is hard and there’s no quick fix. Best case scenario a parent needs to remind the kid to clean their room, usually more effort is required. Much more.

What I heard more than once and can attest to myself is that kids don’t pay attention to the mess they leave. They just don't see it!
Which images should be able to help with ;)

But, as I was rightly pointed out, it wouldn't matter if kids won't use it.

My kids lost interest in wireframe before I was able to ask all the questions I wanted, but they liked the home screen design:

Simba on the dishwasher was particularly popular.
The illustration generated with OpenAI was also a hit, though it took a couple of tries to get it right:

Another very important part I managed to clarify during interviews is the required level of privacy. Turns out end-to-end encryption is an overkill and parents are happy with baseline data protection & transparency regarding how it's handled.

I am thinking of using short-lived signed URLs with a family specific key, though I didn't have much time to test it out just yet.

Progress so far

I spent most of the time either setting up accounts or remembering how to Flutter. What do you think? Almost pixel perfect, isn't it:

I'm not nearly as proficient with Flutter as I am with web, but it's going somewhere, so that's a relief!

I started with the backend as well, there are still some questions to answer, but I'm quite settled on GCP Cloud Run / Functions, Firestore and Firebase auth until something goes wrong. I was ready to write everything myself, with a separate bare metal SQLite for each family, but luckily I got help getting out of that rabbit hole!

And I went down so many different rabbit holes...

AI rabbit hole

For example, being able to generate a task title and description is a low hanging fruit, but it turned out to be very challenging for OpenAI and Gemini (they outright refused describing the image); Claude was up for a challenge, but found a washing machine that wasn't on a picture:

Not blaming them though, my laundry is, in fact, overwhelming!

I will probably end up adding AI sooner rather than later, but so far it seems that I'll need to train a small model for each family specifically (and make sure that users can actively opt in, not the other way around).

I'm thinking about using LiteRT and bundle it into the device itself, but let's see how it goes - I'm way over my head here to be honest so I intend to dedicate more time getting myself up to speed with ML and especially how to make it safe for children before I commit to it. Wouldn't want AI recommending climbing up the shelves or mixing chemicals or god knows what else 😱

Either way, I wouldn't mind AI taking my laundry, cleaning and cooking jobs asap 😎

But I won't build it just yet...

What am I building then?

Here is what I scoped out for the first release:

• Family account
• Kids accounts
• Visual todos with the ability to assign task to a family member
• Bounty cleaner game
• Rewards system: earn by completing bounties and contribute to family rewards

It has been a struggle to limit the app to the barebones functionality. I have so many things I want to add: animated avatars, digital asset store, ability to generate rewards illustrations using those avatars and, as I mentioned, AI.

But ultimately, I can't really release Hounty to production untill it solves a problem with my household and so I intend to make the first version, run a closed alpha with friends, family and whoever else is willing to try it, see what works with different kids and what doesn't and only then add more features.

What about families without kids?

I heard more than once "I don't have any kids, but I wonder if Hounty would help with my partner/roommate/spouse". I feel you guys! And I'll definitely look into fine-tuning Hounty for artistic (or even ADHD ridden) adults. I hope you can still use the free web version till then and do let me know if something doesn't work or could be improved there!

Thank you for all your support, ideas and collaboration! If you'd like to stay updated on Hounty's progress or participate in the closed alpha, you can write to [e-mail] or on LinkedIn

Next update will be about the release - stay tuned! 🎮✨