The ₹700 GPS Device That Tracked 10,000 Vehicles
In 2015, a GPS tracking device cost ₹4,000 minimum. If you wanted sensors — temperature, fuel level, RFID — you were looking at ₹8,000-10,000.
We built one for ₹700.
Not a stripped-down version. Not a “lite” product. A full-featured, multi-sensor-capable GPS tracking device that we deployed on 10,000+ vehicles across India.
This is the story of Fourzip — and the most important lesson it taught me about building products.
The Basement
While our engineering classmates were joining TCS, Infosys, and Wipro, four of us moved into our professor’s basement.
No funding. No connections. No clue what we were doing. Just an observation: fleet tracking in India was a racket, and someone had to fix it.
The Problem With Existing Hardware
We started by trying to use what was already on the market.
Chinese vendors (GT06, Teltonika): Cheap-ish, but completely inflexible. Want to add a temperature sensor? Buy a different device. Want custom firmware? Good luck. And their server costs would eat you alive.
Indian vendors (Itriangle and others): Better support, same problems. Locked ecosystem, limited customization, high per-device costs.
For a small transporter with 200 trucks, hardware alone would cost ₹8-20 lakh. That’s before software, before servers, before any operational value. The economics killed the adoption.
The Decision
We could have built software on top of existing hardware. That’s what everyone else was doing — a nice dashboard on top of an expensive device.
But if the hardware cost is the barrier to adoption, improving the software doesn’t solve the problem. You have to fix the economics.
So we decided to build the hardware.
Building the Device
I’m a product guy, not an electronics engineer. But constraints breed creativity.
Step 1: Start with commodity parts. We sourced GPS modules, GSM modules, and microcontrollers at commodity prices. No proprietary chips, no vendor lock-in.
Step 2: Write custom firmware. We wrote RTOS (Real-Time Operating System) based firmware from scratch. Custom scripts that handled GPS acquisition, data transmission, sensor reading, and power management. This is where the magic happened — our software made cheap hardware smart.
Step 3: Multi-sensor architecture. Instead of building separate devices for different use cases, we designed a single platform with sensor ports. Temperature probe for cold chain? Plug it in. Fuel sensor for pilferage detection? Plug it in. RFID reader for student tracking? You get the idea.
Step 4: MQTT infrastructure. We built our own server infrastructure using MQTT — lightweight messaging protocol perfect for IoT. Thousands of devices sending real-time data, and our server costs were a fraction of what the Chinese vendors charged.
Final BOM: ₹700 per device. Eighty-two percent cheaper than the cheapest alternative.
The Verticals
With hardware costs this low, we could go after markets that were previously uneconomical:
Municipal Garbage Trucks (Ulhasnagar): Our first customer. Track every garbage pickup, let citizens report missed pickups through an app. This contract opened the government sector for us.
Police & Fire Vehicles (Vijayawada): GPS + cameras on government vehicles. Real-time dispatch, route tracking, accountability. Scaled this to Jharkhand with Ecosense Enviro.
Ammonium Nitrate Transport (Deepak Fertiliser): Pilferage detection using fuel sensors and route deviation alerts. When you’re transporting explosive materials, real-time tracking isn’t a nice-to-have.
Student Safety (Schools): RFID cards + GPS + geofencing + cameras. Parents know exactly when their kid boards the bus, where the bus is, and when it arrives at school.
Agriculture IoT: MQTT-based servers for soil monitoring, weather stations, irrigation control. Different domain, same infrastructure.
Local Transporters: The bread and butter. Real-time vehicle tracking, route replay, proactive alerts for thousands of small fleet operators.
The Growth
No VC money. No angel rounds. Pure revenue growth.
- Year 1: 4 people, basement, first customer
- Year 3: 15 people, real office, multiple verticals
- Year 5: 25 people, 5,000+ vehicles, government contracts
- Year 7: 30 people, 10,000+ vehicles, 6+ verticals
We burned cash on over-ambitious solutions and government implementation timelines. Eventually pivoted to CTO-as-a-Service. But the core lesson stuck.
What This Taught Me
1. Constraints Are Features
We didn’t have money to buy expensive hardware. So we built cheap hardware. That “limitation” became our biggest competitive advantage.
Every MVP I build today follows this principle. You don’t need the expensive tool. You need the right tool at the right price.
2. Fix the Economics, Not the Dashboard
If the fundamental economics of your product don’t work, no amount of UX polish will save you. We could have built the world’s best fleet management dashboard — on top of ₹4,000 devices that nobody would buy.
When I scope MVPs for founders, I start with economics. Not features.
3. Start With One Customer
Our first customer was a municipal garbage truck fleet. Not glamorous. Not scalable (yet). But it was real revenue from a real problem, and it funded everything that came after.
Stop planning for 10,000 users. Get one user. Then get ten.
4. Hardware is Software With Consequences
When your code has a bug, you push a fix. When your device has a bug, it’s sitting in a truck in Jharkhand and someone has to physically go retrieve it.
This taught me to test obsessively before shipping. A habit that carries over to everything I build.
The Legacy
Fourzip eventually wound down — we spread too thin across too many verticals with too little capital. Classic bootstrapped startup death.
But those 7 years built every instinct I have as a builder. Resourcefulness. Speed. The belief that if you can’t buy it cheap enough, you build it yourself.
When a founder tells me “it’s too expensive to build,” I think about that ₹700 device.
It’s never too expensive. Your approach is just too standard.
Building something with hardware constraints? Let’s talk.