Learning Flutter for 6 Months: My First App Project Paid Rs 22,000

The app went live at 6 pm. By 11 that night, the client's message read: "getting complaints, app is crashing."

I opened the laptop and found the bug within twenty minutes, a missing null check on an API response that only broke when the server responded slowly. Fixed it, tested it, pushed a new build, had it live on the Play Store by 1:30 am.

The client's reply the next morning: "appreciate you fixing this so fast." That single sentence is the moment this started feeling like something I could actually build a freelance practice around.

---

Where this started, and how little I knew going in

Web development came first, basic HTML-CSS-JavaScript work, small landing pages for local clients. One of them asked, almost in passing, whether I could also build a mobile app for his small chain of juice bars. I said yes without having the faintest idea what Flutter even was.

That yes sent me straight to Google, searching "how to build a mobile app." Flutter kept surfacing above React Native, and the performance comparisons tipped the decision.

---

The first two months were mostly confusion and a counter app

Following tutorials produced the standard first project everyone builds, a counter app where a button increments a number on screen. Simple enough, except Dart itself, Flutter's programming language, took real effort to get comfortable with.

The widget concept confused me more than anything else. Every tutorial repeated "everything is a widget," but understanding that practically, that a single line of text is a widget, a button is a widget, an entire screen is also technically one widget, took weeks to sink in properly.

Actually, the first month barely involved any real coding at all. Installing Android Studio, getting an emulator running, sorting out SDK versions, that setup alone ate close to three weeks before any actual app-building began.

---

The first real app, built purely for practice

Three months of practice led to a personal to-do list app, built only for myself. This was the first project where state management, tracking and updating data properly within the app, actually clicked, along with local storage so data survived closing the app.

That app never got published anywhere. But it proved something important: a complete, working app was possible now, not just copying tutorial code line by line.

---

The first client: from a juice bar to a salon

The juice bar client's project quietly died, budget constraints on their end. A different lead came through instead, a small salon chain wanting an appointment booking app.

Rs 19,500 for the project. The scope covered: service booking, a simple loyalty point tracker, and push notifications for appointment reminders.

Six weeks to build. Firebase integration turned out to be the hardest part, since this was the first time working with a real backend rather than local-only storage. Two full days went into a single authentication bug alone.

---

Launch night, the crash from the opening

The app went live, and the crash described at the start happened that same night. I was genuinely worried the client relationship would sour over it. Instead, fixing it fast built more trust than the bug had damaged.

One lesson stuck from that night: testing under poor network conditions matters before shipping anything. Development happened entirely on fast home wifi. Real users were on considerably slower mobile connections.

---

The second project, a different kind of mistake

A second client arrived through the salon owner's referral, a small home-decor retail brand. This time no technical mistake happened. Scope crept instead, steadily, one small request at a time.

The project started at Rs 15,000. Extra feature requests kept coming, each framed as "just one small addition," and none of them got billed separately at the time. By the end, the actual scope had roughly doubled for the same original payment.

---

The honest numbers

Six months in, total earnings sit at Rs 82,900, mostly from those two larger projects plus some smaller maintenance work in between. Averaged out, that's Rs 13,800 a month, though the distribution is uneven, some months brought in Rs 26,000, others as little as Rs 2,800.

Time cost runs 10-12 hours a week, split between continued learning and actual project work.

The costliest mistake was letting scope creep slide unchecked on that second project. Every contract now states clearly: "3 revisions or additions included, additional features billed per item." That rule exists because of exactly that one project.

And here's something nobody warns you about going in: publishing to the Play Store is its own multi-day process. Reviews take 2-3 days, and Google occasionally rejects submissions over small policy details. My first app got rejected twice, once for a missing privacy policy link, once for a permission that wasn't explained clearly enough in the listing

---

Learning iOS publishing is next on the list, everything so far has stayed on Android only. The Apple Developer account's annual fee keeps making me hesitate on actually starting that