App Development | Apurav Gaur · May 26, 2022 · 13 min read The micro-mobility market is no longer an experiment. Lime has completed over 500 million rides. Bird operates in 350+ cities. Beam, Yulu, and Bounce have built profitable businesses in Asia and Southeast Asia. And the global micro-mobility market, currently valued at $9.7 billion is projected to reach $214 billion by 2030. If you are building an e-scooter sharing platform in 2026, the opportunity is real. The technology is mature. The user behaviour is established. What most development guides won’t tell you is that building a scooter platform is not one app, it is three interconnected systems, each with distinct technical requirements. Get one layer wrong and the entire platform breaks. This guide covers exactly how all three layers work, what it costs to build them, and what technical decisions you need to make before writing a single line of code. Why E-Scooter Platforms Are Still a Strong Business Opportunity in 2026 Urban congestion is getting worse, not better. In every major city – Mumbai, London, New York, Singapore, Dubai, the average commute under 5km still takes longer than it should by car or public transport. E-scooters solve a specific problem that no other transport mode solves well: the first and last mile. What makes the business case stronger in 2026 specifically: The hardware has matured. First-generation scooter businesses lost money replacing hardware every 3-6 months. Current IoT controllers (Segway, Omni, Okai) have lifespans of 3-5 years. The unit economics work now in a way they did not in 2018. Regulation has clarified. Most major cities have published micro-mobility operating frameworks. The uncertainty that killed early operators has largely resolved cities now want these platforms, not just tolerate them. The gap in tier-2 cities is enormous. Lime and Bird are focused on tier-1 cities in the US and Europe. Tier-2 cities globally across India, Southeast Asia, the Middle East, and Africa are completely underserved. This is where the next generation of micro-mobility businesses will be built. What You Are Actually Building: The Three Layers Before discussing features, it is important to understand that a scooter platform is not one application. It is three connected systems that must work together in real time. Layer 1 – The Rider App (iOS + Android) What users download. What they interact with on every ride. Every friction point here costs you rides and retention. Layer 2 – The Operator Dashboard (Web) What your operations team uses to manage the fleet. Without this, the business cannot run, you cannot deploy scooters, manage batteries, handle faults, or understand revenue. Layer 3 – The IoT Hardware Layer The bridge between your software and the physical scooters. This is the layer most development agencies do not understand well and where most e-scooter builds fail. The rider app handles the user experience, the backend manages fleet operations and payments, and the IoT layer communicates directly with the scooter hardware via MQTT protocol. Layer 1: Rider App Features: What to Build and Why Must-Have Features (MVP) User registration and verification Email, phone, or social login. Driving licence verification for markets that require it. ID verification via Onfido or similar for age-restricted markets. Live map with scooter locations Every available scooter visible on the map in real time. Powered by Google Maps API or Mapbox. Location data comes from GPS pings sent by the IoT controller every 30–60 seconds when idle, every 3–5 seconds during a ride. QR code unlock User scans QR code on the scooter. App sends unlock command to your backend. Backend sends command to the IoT controller via MQTT. Controller disengages the motor lock. Confirmation returns to app within 1–2 seconds. This entire flow needs to work reliably on 2G connections not just 4G. Real-time ride tracking Live map showing the rider’s route, current speed, battery level remaining, and estimated cost as the ride progresses. Powered by Socket.io for live updates. In-app payment Stripe for international markets. Razorpay for India. In-app wallet with top-up for markets where per-ride card processing is expensive. Payment must capture before the ride unlocks not after. End ride and parking confirmation Rider ends ride. App requests a photo of the parked scooter. Backend validates parking zone using geofence check. Ride cost finalised and receipt sent. Ride history and receipts Every trip logged: start point, end point, distance, duration, cost, map replay. Required for customer support and dispute resolution. Push notifications Ride start confirmation, ride end receipt, low wallet balance alert, promotional offers. Firebase Cloud Messaging for both iOS and Android. Phase 2 Features (Post-MVP) Scheduled booking: Reserve a scooter up to 30 minutes in advance Referral and rewards program: User referral codes, loyalty points, free ride credits Helmet rental integration – for markets requiring helmets Multi-vehicle support – e-bikes, cargo bikes, mopeds on the same platform Group ride – multiple users sharing one booking session Need help choosing the right stack? We can guide you Layer 2: Operator Dashboard Features- What Your Team Needs Fleet Management Live map showing every scooter with real-time status: available, in-ride, low battery (below 20%), offline, needs maintenance. Filter by zone, battery level, or fault type. Click any scooter to see full history. Geofencing Draw riding zones (where scooters can be ridden), no-go zones (restricted areas), and parking zones (where rides must end). App automatically warns riders approaching boundaries and charges penalty fees for out-of-zone parking. This is critical for city permit compliance. Battery and Maintenance Alerts Automatic alert when a scooter drops below 20% battery. Fault alerts when the IoT controller reports a hardware issue – brake sensor failure, motor fault, battery degradation. Each alert assigned to a field team member for resolution. Charging Team Management Assign low-battery scooters to charging teams. Track pickup confirmation, charge status, and redeployment. Revenue lost per hour of downtime visible per scooter. Revenue and Analytics Dashboard Rides per day, revenue per scooter, revenue per zone, peak hours, average ride duration, top 10 performing zones. Required for operational decisions and investor reporting. User Management View rider accounts, ride history, payment history. Issue credits, handle disputes, suspend or ban users. Search by phone number, email, or ride ID. Pricing Management Set base fare, per-minute rate, unlocking fee, and penalty fees by zone. Run promotional pricing for specific time windows or user segments. Layer 3: IoT Hardware Integration – The Technical Reality This flow runs across three layers: the rider app, the backend MQTT broker, and the IoT controller on the scooter itself. Every step must work reliably on 2G connections, not just 4G. This is the layer that determines whether your platform works in the real world. Every scooter needs an IoT controller board that communicates with your backend. The most widely used options in 2026 are Segway IoT, Omni IoT, and Okai – each with different API documentation, firmware update processes, and reliability characteristics. Your choice of hardware vendor affects your backend architecture significantly. How the communication works: Your backend communicates with scooter controllers using MQTT – a lightweight messaging protocol designed for IoT devices operating on low-power, low-bandwidth connections. This is the industry standard and what Lime, Bird, and Beam all use. When a rider scans a QR code: Rider app sends unlock request to your API server API server authenticates the request and publishes unlock command to MQTT broker MQTT broker delivers command to the specific scooter’s IoT controller Controller disengages motor lock and sends confirmation back via MQTT API server receives confirmation and notifies the rider app Ride starts – GPS pings begin every 3–5 seconds What your backend needs to handle: MQTT broker – Mosquitto for smaller fleets, AWS IoT Core for scale Redis for real-time scooter state (current location, status, battery, lock state) PostgreSQL for historical trip data, user data, payment records WebSockets for pushing live fleet updates to the operator dashboard The most common IoT integration mistakes: Building for 4G only scooters in urban canyons frequently drop to 2G; your MQTT implementation must handle reconnection gracefully Ignoring firmware updates – IoT controllers need periodic firmware updates; build an OTA (over-the-air) update process from day one No fault tolerance in the lock/unlock flow if the unlock command fails, the user is standing in the street with a locked scooter; build retry logic and a manual override flow Tech Stack for an E-Scooter Platform Layer Technology Why Rider app (iOS + Android) React Native One codebase for both platforms – faster development, easier maintenance Operator dashboard React.js Fast, component-based web UI for real-time fleet updates Backend API Node.js Non-blocking I/O handles thousands of concurrent GPS pings efficiently Database PostgreSQL Structured trip data, user records, payment history Real-time state Redis Sub-millisecond reads for live scooter location and status IoT communication MQTT (Mosquitto / AWS IoT Core) Industry standard for IoT device communication Mapping Google Maps API Live map, geofencing, route calculation Real-time updates Socket.io Live GPS updates to rider app and operator dashboard Payments Stripe (global) / Razorpay (India) PCI-DSS compliant, reliable, well-documented Push notifications Firebase Cloud Messaging iOS and Android from one integration Cloud infrastructure AWS or GCP Auto-scaling for peak usage, global availability Development Timeline Phase What Gets Built Timeline Phase 1 — MVP Rider app (core flow) · Basic operator dashboard · IoT integration (1 scooter model) · Payment integration 10–14 weeks Phase 2 — Operations Full operator dashboard · Geofencing · Charging team management · Analytics · Multi-zone pricing 6–8 weeks Phase 3 — Scale Multi-city support · Multi-vehicle types · Advanced analytics · API for city integrations 8–12 weeks Most startups launch Phase 1 in one city, validate unit economics, then build Phase 2 before expanding. How Much Does It Cost to Build an E-Scooter App in 2026? Cost depends on the scope of build and the location of the development team. Scope India-based team US-based team MVP (Phase 1) $40,000 – $80,000 $120,000 – $240,000 Full platform (Phase 1 + 2) $80,000 – $160,000 $240,000 – $480,000 Enterprise / multi-city $150,000 – $300,000+ $450,000 – $900,000+ Building with a dedicated development team in India delivers the same senior engineering quality at 60–70% less than equivalent US or UK rates. Full IP ownership, NDA-first, and EST/PST timezone overlap are standard on every Deorwine engagement. What drives cost up: Supporting multiple scooter hardware models (each has different IoT API behaviour) Multiple cities with different pricing and geofence configurations Multi-language support Advanced analytics and ML-based demand forecasting Custom IoT controller hardware (rather than off-the-shelf) What keeps cost down: Starting with one city, one scooter model Using off-the-shelf IoT hardware with documented APIs Launching Android-only first (rider app), then adding iOS in Phase 2 Building the operator dashboard as a web app, not a native mobile app Planning a SaaS product? Talk to Deorwine about a dedicated team What Deorwine Has Built That’s Relevant Deorwine has built real-time GPS tracking platforms, IoT-connected mobile products, and on-demand apps handling 40,000+ daily transactions. Specifically relevant to e-scooter development: Freight Country – a real-time freight management system with live GPS tracking, multi-role mobile apps (shipper, driver, carrier), automated dispatch, and digital BOL. The same GPS architecture, multi-role structure, and real-time tracking that powers Freight Country applies directly to a scooter platform. The Laundry Machine – an on-demand platform handling 40,000+ completed orders across 4+ cities. The same booking flow, payment integration, real-time order tracking, and operational dashboard architecture that powers The Laundry Machine is the foundation of a scooter rider and operator system. We understand the technical complexity at the app layer, the backend layer, and the IoT layer and we build with the production architecture that scales, not the prototype architecture that needs to be rebuilt six months after launch. Choosing the Right Development Partner Building a scooter platform is not a job for a generalist agency. The IoT layer alone requires specific expertise that most mobile app developers do not have. When evaluating any development partner for a micro-mobility project, ask: Have you built real-time GPS tracking at scale before? Show me the architecture. How do you handle IoT connectivity failures what happens when a scooter drops its MQTT connection mid-ride? What is your approach to firmware updates on deployed hardware? How do you structure the geofencing system to handle multiple concurrent zone configurations? If the answer to any of these is vague, the partner does not have the experience to deliver a production-grade scooter platform. Ready to Build Your E-Scooter Platform? Deorwine places dedicated development teams, senior engineers across mobile, backend, IoT, and platform architecture within 3–7 days. EST and PST timezone overlap guaranteed. Full IP ownership. NDA-first on every engagement. If you are building a scooter sharing platform, a cargo bike service, or a multi-modal micro-mobility product, talk to our team before you write a single line of code. Get a cost estimate for your e-scooter platform Contact us Share Facebook Twitter LinkedIn The Author Apurav Gaur Co-founder, Deorwine Infotech I'm Apurv Gaur, Co-founder of Deorwine Infotech, with 15+ years of experience in building digital products. I started my journey as a developer, but over time, I grew into a business-focused technologist, helping companies scale through technology, strategy, and AI-driven solutions. Today, I focus on AI-led development to build faster, smarter, and more scalable products.