Menta

Build habits together.

Comment

Inspiration

My friends and I kept trying to build habits or learn things together. We used texts, Discord and spreadsheets. Most check-ins were just, "Did you do it?" followed by "I did" or "I didn't." Usually there was no photo or any other proof. It helped, but every group had to invent its own system.

I also wanted something for myself. I've been trying to read the Bible, study, run and get to the gym more consistently. I found plenty of habit trackers, but none worked the way I wanted once friends were involved.

Duolingo made me think about why a streak can keep somebody involved. BeReal made me think about how a simple photo can make a check-in feel more real. I wanted Menta to stay simple, clear and fun without turning accountability into public embarrassment or tracking everything a person does.

What it does

Menta starts with a promise. You say what you're going to do, when you'll do it and what proof will count. Today shows the promise that needs your attention next.

When proof is due, you can add text, a photo or a video. You can keep a promise private or make it with a group. For a group promise, another person reviews the proof. If the proof is not enough, they explain why and you can try again.

Confirmed follow-through can earn Momenta, which people can use in the store or on recovery tools. Menta Pro is the paid tier. RevenueCat handles the subscription purchase and restore process.

I built sharing into Menta because habits often start with somebody asking a friend to do something with them. Referrals help people bring that friend into the app. Events take the same idea into the real world: an organiser can invite people, help them join, let them check in with a photo and show the group who took part.

How I built it

I built Menta in TypeScript with Expo and React Native. Supabase handles accounts, application data, proof storage, live updates and server functions.

RevenueCat loads the subscription plans and handles purchases and restored subscriptions. Menta's server confirms Pro access and Momenta changes. The app does not treat a purchase as finished just because the payment screen closed.

Paper was my design source. I used Codex throughout the rebuild to compare the working app with those designs, follow each action from start to finish and find states I had missed. That was especially useful when something looked finished on the phone but an upload, review or purchase was still waiting for confirmation.

I used AI to help build Menta, but I did not use a model to decide whether somebody's effort counts. For a group promise, another person reviews the proof. That judgement should stay with the people involved.

Challenges

The hard part was making the promise feel real without making Menta tiring to use. A quick tick can keep a streak alive after the habit has stopped meaning much. I worried that too many proof steps would create a different problem. I had to make the promise, proof and result clear while still letting each person choose what to share and who can see it.

I kept finding the same type of reliability problem. A proof upload, group review, purchase or account-recovery step could look finished on the phone while the server was still working. I added separate states for uploading, waiting, failing, restoring and confirming so the app does not tell somebody that an action succeeded before it knows.

Some of the product questions are still open. I'm working out how valuable the streak feels, what Momenta should be worth and where the free and Pro boundaries should sit. I also want reminders to feel more like a personal trainer who wants you to succeed, rather than another generic notification. I need real use to tell me which of those ideas actually helps.

What I learned

Building Menta showed me that the promise, proof, review and result cannot be treated as separate features. If Menta marks a promise complete before the proof or review is actually finished, the streak stops meaning anything.

I also learned not to confuse a finished screen with a finished action. I had to test the app on a real device and wait for RevenueCat and Menta's server to confirm purchases and restored access. That changed both the interface and the way I tested it.

What's next

My immediate goal is to get Menta's first public version through Apple review and let people use it. I want to learn which reminders bring people back, whether the streak still feels valuable after the first few days and whether groups and events actually help friends keep doing something together.

I also need to refine the business model. I want the free tier to be useful while giving Menta Pro a clear reason to exist. I do not want to pretend I know the right price or the exact value of Momenta before people have used the app.

Built With

Share this project:

Updates