Mobile App Development Process in Malaysia: From Idea to App Store

You have an idea. Maybe it’s an app for your gym, your tuition centre, your delivery network. You’ve been quoted somewhere between RM 8,000 and RM 200,000 by different people. The numbers are wildly different because nobody told you what’s actually involved. Here’s the full process, in seven honest phases, with where most projects in Malaysia stall.

This is the process we use, and it’s roughly what every reputable agency in KL uses too — only the names change.

Phase 1: discovery (week 0)

Before we touch design or code, we have a 30–60 minute conversation. Goal: understand what success looks like.

What we ask:

• Who uses this app, and how often?
• What problem does it solve that a website wouldn’t?
• What’s the one thing that, if it works, makes this worth building?
• What’s the budget range and the deadline?

What you should bring:

• A page or two of notes about the idea
• Examples of apps you like (and ones you don’t)
• An honest budget number (not the price you hope to pay)

Where projects fail here: “we’ll figure it out as we build.” If discovery is skipped, scope explodes by month two and the budget runs out before launch.

Phase 2: scoping & quote (week 1)

We turn the discovery into:

• A feature list, in priority order: must-have, should-have, nice-to-have.
• User journeys — 3 to 8 short flows: “tuition parent logs in → checks fee → pays via FPX → gets receipt”
• Tech stack proposal: native (Swift / Kotlin), cross-platform (React Native / Flutter), or hybrid.
• A fixed quote, broken into milestones with prices.
• A timeline with realistic dates.

Most apps for SMEs in Malaysia work fine in React Native. It builds for both iOS and Android from one codebase, which roughly halves cost vs going fully native. Native makes sense if you need: complex camera/AR, intensive Bluetooth, certain background tasks, or App Store/Play Store review polish for a consumer app.

For the broader picture see mobile app development in Malaysia.

Phase 3: design (weeks 2–4)

Two outputs:

• Wireframes: black-and-white sketches of every screen. Cheap to change. We iterate here.
• Hi-fi designs: pixel-perfect screens in your brand. These approximate the final look.

Tools we use: Figma. You get the file with edit access from day 1, so you can comment directly on screens.

What you should NOT do: design committee. Get one or two final decision-makers; rotating opinions through six stakeholders kills schedules.

Where projects fail here: approving designs without thinking about content. Real menus have 47 items, not 4. Real product names are 38 characters long. Push real data through every screen before you sign off.

Phase 4: build (weeks 4–12)

We code in vertical slices. That means: instead of “do all the screens, then all the backend, then all the data,” we build one feature end-to-end (login → fetch user → show dashboard) and demo it. Then the next.

Why: you see real progress every week, and we catch architecture problems early.

Typical weekly cadence:

• Monday: short call (15–30 min) — what we’re shipping this week.
• Tuesday–Thursday: heads-down dev.
• Friday: demo on TestFlight or Firebase App Distribution. You play with it on your phone.

Tools you’ll touch:

• TestFlight (iOS) and Firebase App Distribution (Android) — internal beta.
• Sentry or Bugsnag — crash reporting.
• Notion or Linear — feature tracker.

Where projects fail here: silent dev. If you don’t see a working build every week, the project is in trouble — either we’re stuck and not telling you, or we’re building something nobody asked for. Demand the demo.

Phase 5: integrations (overlap with build)

The least glamorous and most variable phase. Estimating this badly is how RM 30k apps become RM 90k apps.

Common Malaysian integrations:

• Payments: Billplz, Stripe Malaysia, Curlec, FPX direct. RM 2,000–6,000 per gateway with full webhook handling.
• WhatsApp Cloud API for notifications: RM 1,500–3,000 setup + Meta’s per-message charge.
• Maps: Google Maps SDK is free up to ~28k loads/month; after that it’s measurable. Mapbox is sometimes cheaper at scale.
• Push notifications: Firebase Cloud Messaging is free; the cost is in writing the trigger logic correctly.
• SMS OTP: Twilio, MessageBird, or local Malaysian SMS gateways. Per-message cost matters more than setup.
• POS / ERP sync: depends entirely on the system. Some have clean APIs (Storehub, Loyverse), some need scraping.

Each integration adds 1–2 weeks of work and a real-money cost. Map them all in scoping; surprises here destroy budgets.

Phase 6: testing & store submission (weeks 11–14)

Two parallel tracks.

Functional testing:

• We write automated tests for critical flows (login, payment, push).
• You run UAT on real devices — not the simulator. Get a colleague who isn’t your designer.
• We test on at least one low-end Android (RAM ≤ 3GB) and one older iPhone (iPhone X-era). Apps that fly on a Pixel 8 die on a low-budget phone.

Store submission:

iOS App Store:

• You need an Apple Developer account (USD 99/year) under a company DUNS number for organisations, or your personal account for solo apps.
• App Review takes 1–3 days now. Rejections in Malaysia are most often: privacy policy too vague, IAP rules around real-world goods, or location-permission text not specific enough.

Google Play Store:

• USD 25 one-time for the developer account.
• Closed testing (20 testers, 14 days) is now mandatory for new accounts before public release.
• Approval takes 1–7 days. Common Malaysian rejections: data safety form not matching the actual code, gambling-adjacent metaphors in metadata.

Budget 2 weeks for store back-and-forth. It’s almost never zero.

Phase 7: launch & post-launch (week 14+)

Launch day:

• Submit to stores — you (the client) press the publish button. Symbolic but matters.
• Sentry / crash dashboards on screen.
• Analytics events flowing.
• A first-week support window where we monitor + patch.

What people forget:

• Version 1 is not done. Plan three minor releases in the first two months — typo fixes, performance, things you only notice with real users.
• Ratings prompts: don’t ask for a review on first launch. Ask after a successful action, after the third use.
• Onboarding: you’ll learn within a week which screens drop users. Be ready to redo onboarding in month 2.

Real budget breakdown (Malaysia, 2026)

Typical SME app — login, profile, list of items, push, payment, basic admin web panel:

Phase	Hours	Cost (RM, mid-range)
Discovery + scoping	8–16	2,000
Design (Figma, ~12 screens)	30–50	7,500
iOS + Android build (RN)	200–300	35,000
Backend (Node + Postgres + auth)	80–120	14,000
Integrations (1 payment, push, maps)	40–60	8,000
Testing + store submission	30–50	5,500
Buffer (always real)	~15%	11,000
Total		~83,000

Solo dev in Malaysia: 60–75k for the same scope, slower timeline. Mid-size KL agency: 100–150k. Big-4-style consultancy: 250–500k.

Pick based on risk tolerance, not the cheapest number.

When NOT to build an app

In half the briefs we get, the answer is “you don’t need an app, you need a good website.”

• If users open it less than once a week → progressive web app or website.
• If it’s mostly content (catalog, info, brochure) → website.
• If you’ve never validated the idea on a landing page → start there.
• If your budget is under RM 25k → start with the web flow, build native after revenue.

We say no to a quarter of app inquiries for these reasons. Not because we don’t want the work; because a failed app costs more than a working website.

Frequently asked

React Native vs Flutter vs native — which?

For most Malaysian SME use cases, React Native. Big React Native ecosystem, faster hiring, easier handover. Flutter is great too — just rarer in the Malaysian dev market. Native is for apps where the OS fights you (advanced camera, BLE, complex maps, AR).

Can I get an app for under RM 10k?

Only if it’s a wrapper around an existing web app, or a simple static-content app (PDF reader, brochure). Anything with login, payment, or live data: no.

How long until I can publish to the store?

Realistic: 12–16 weeks for a clean MVP. Anyone promising 6 weeks is either skipping testing or building something so trivial it should have been a website.

What about Apple’s IAP rules?

If your app sells digital content consumed inside the app, Apple takes 15–30%. If you sell physical goods or services delivered outside the app (restaurant orders, salon bookings, gym memberships), IAP doesn’t apply — Stripe / Billplz / FPX is fine. Get this wrong and you’ll be rejected.

Do you do post-launch maintenance?

Yes — month-to-month, no contract lock. Typical retainers RM 1,500–4,000/month for bug fixes, OS upgrade compatibility, minor features. Cancel any time.