Fully charged: How Drive Terra observes its entire e-bike business in Grafana, from battery health to inventory

Drive Terra runs a fleet of electric delivery bikes and battery swapping stations across the Middle East. Instead of using separate tools for IoT monitoring and business reporting, the team built everything in Grafana.

In this session, Terra CTO Mohammad Omar shares how the company uses Grafana for two very different jobs. On the technical side, real-time vehicle telemetry, battery health monitoring, and notifications are all connected to Terra's databases directly. On the business side, Grafana  powers swap station utilization, bike availability, inventory tracking, and operational dashboards used by the ops team, executives, and partners. Learn practical lessons from a small team running Grafana as its single source of truth – what worked, what was tricky, and why they skipped Power BI entirely.

Mohammad Omar (00:01):

I assume everyone of you, there was one time someone from the product, the CEO, or from the business, they came to him and they said, "We need to monitor our processes. We need to monitor our data." So if you work in a software company, you know, engineering are the software engineers we have or the SREs, but we all know what they are. But at Drive Terra, so this is not working. Okay, so what we do at Drive Terra is we have five components. We have a battery of stations, so a station of batteries. We have a bike and we have the batteries, we have the cloud, we have the mobile app, and we have a lot of operation going on the ground. So when a driver come to a station, you just need to swap a battery and not charge it. So when speaking about this, we have several team working together to ensure that things are working in a good way.

(01:13):

The first are the engineers, and of course they are our software engineers. We have our IoT and the hardware engineers working on the firmware, on the connectivity, on the device, on the hardware and all the IoT ecosystem. You have the electrical engineers working on the cell of the batteries, working on the charge cycle and working. And we have the mechanical engineers working on everything related to the bike, faults, working on the gears, working on the steering wheel, everything we know about bikes. The second team is operation, which is related to the fleet managers, ensuring that the fleet is running all the time. We have our logistic coordinator coordinating all the stations stuff, battery charging, and we have the maintenance team going for chasing every bike if it goes down for maintenance. And we have have our warehouse and inventory team working under the operation. And then you move on to the leadership, starting with the CEO, that is always asking for some visibility, moving to the partners manager where they need to see all the bikes working under specific partners.

(02:28):

Let's say the bikes working for Amazon, the bikes working for Kareem, for Uber, and we have the finance team. They need to understand their KPIs. And last you have the partners themselves where they need to see the drivers running on the ground. So the problem here is between the non-technical team and between the technical team. Our non-technical team, they need to see all the operation cost or the operation KPIs. They don't go and wanted to do the WhatsApp stuff without any system. The leadership want to see weekly reporting. The old way was they use Excel and the new way is, you'll see later on. So, and every time a partner call us to ask what's happening with the bikes, what's happening with a certain driver and why his bike is going down. On the other way, the way we know, are the software engineers, the engineers, they can track the IoT, they can track the servers, and they can track from an electrical perspective all the electrical wiring inside the bike itself.

(03:39):

So here I can say that after my experience, observability is not just for the engineers, it's for the whole company.

(03:51):

So what we did is we used the Grafana for this. The local communities in Dubai and Lebanon and the Middle East, the first word outta their mouths was, "You are crazy to do this." And then when I show them this dashboard, it's like "Wait, it's not crazy, it's doable." If you have the right infrastructure, if you have the right database, and the right data pipeline, you can monitor the bike live the dashboard auto refresh every 30 second, every five second, every one second. You can configure it the way you want. You can keep track of all the faults. I can use the laser here, fault by fault by fault by fault, enabling everyone to understand how the hardware is performing. This is a full digital twin right there in Grafana. You can monitor the battery levels. We have two battery in each bike, so you can see them directly.

(04:51):

And here at in the bottom you can see the lifecycle on how the battery goes in a bike to a station, to another bike, to another station, to another bike, to another station. So speaking a little bit more technical, I'll take the bike as example. We have an IoT box in every bike, GPS update every ten second. It includes the voltage, just short cycle, the fault codes, and the motor data. Once you receive this data from the device, you need to process it. One of the cool things that the Grafana has out of the box are the alerts. So just imagine that if Grafana is reading our data every five second, the alerts can come directly from Grafana. The same way we have an on-call software engineer, we have an on-call battery operator or station operator just to be able to monitor the lithium batteries in Dubai heat.

(05:53):

So we take the data from the bike, we move down to, we live stream it to our servers, using MQTT and then it's getting saved in the cloud using MongoDB for IoT. And we have all the metadata, the inventory data, coming from PostgreSQL itself. All of those end up together PostgreSQL, MongoDB, RabbitMEQ in one place to visualize it, which is Grafana. So we have a lot of dashboards. Every five to six dashboards, they have a different category. The battery operators have their own dashboard. The station, they can even use, we use the Grafana. There is a camera feature, button feature, to control some functionality on the station. So those buttons that they can call an API or endpoint and the station operator, they don't need even to go to our main system from Grafana. If they see something wrong, they can stop the station or they can start the fire process.

(06:54):

They can do everything from one place. What we learned from this journey is that Grafana is not just for engineers, it's for the whole company, from the CEO to every and each one of our employees. Your team will ask questions and eventually Grafana was the answer for us. The way we started, we started with one use case, the battery health, then we move to the fleet status. Then you can pick up any single KPI, work on it, show it, gather the feedback from the team, and then build the right dashboard for the non-technical team. I mean, for the technical team, you can always go and find some templates, start with them and kick off, but for the non-technical, this was the hard part of it. So, in this journey, the non-technical team is your champion. When the CEO or anyone ask question, you know the answer.

(07:54):

Thank you, everyone, stay fully charged.