Scaling IoT on a Start-Up Budget: Flutterlab’s Journey with The Tech Dept

Problems We Had To Solve

1. The client had a brilliant hardware team who had created their IP, but no experience building the cloud and software platforms needed
2. The platform needed to be able to launch with a small number of users, but scale to handle thousands without rework
3. There were some basic off the shelf platforms but they didn’t provide the flexibility required
4. This had to be achieved with a start-up budget, without distracting the client from core product development and sales

The Plan

1. Facilitated a North Star Workshop with the client to understand what was, and wasn’t key to them and their potential clients
2. Design a solution that could deliver what mattered right now, but could evolve without significant rework
3. Implement and test this with a group of beachhead customers
4. Defer questions that didn’t need to be solved now until we had real customer feedback

The Challenges

1. Developing a completely novel solution
2. Changing hardware requirements
3. Potential for huge volumes of devices and data
4. Limited resources

What We Did

1. Rapid prototyping to quickly understand what would and wouldn’t work
2. Close collaboration with the hardware team to establish standards that were sufficient to move forwards without being restrictive
3. Created a scalable MS Azure based IoT (Internet of Things) infrastructure that autoscales as more devices are added - combined with optimisation of ‘hot’ and’cold’ data storage to limit the amount of expensive cloud resources used
4. Developed a modular solution that only did what was exactly needed at the start but which we have been able to extend cost effectively over time

Technologies Utilised

• MS Azure - IoT Hub
• Device twinning
• Stream analytics
• React
• .Net Core 8.5
• sMQTT
• PostGres

Results

• Hundreds of Connected Devices
• First Customers Onboarded