Every boardroom is talking about AI. Organisations are investing heavily in analytics platforms, predictive models, automation tools and operational dashboards. The expectation is that better technology will lead to better decisions. But there is a problem hiding beneath the excitement: AI is only as intelligent as the data feeding it.
For Rory Atkinson, director at Orange Logistics and an electrical and electronic engineer by profession, this is not a theoretical discussion. It is something his business deals with every day.
Orange Logistics moves FMCG goods across South Africa, operating in an environment where theft, route disruptions, signal interference and operational complexity are all part of the landscape.
Like many businesses, Orange Logistics had access to tracking systems, location data and reporting platforms. What it did not always have was consistent visibility. And that distinction matters.
There is a growing assumption in many organisations that more data automatically leads to better outcomes. In reality, incomplete data often creates a false sense of confidence.
A dashboard may look impressive. Reports may appear accurate. AI models may generate sophisticated recommendations. But if the underlying data becomes unavailable at critical moments, every decision built on that information becomes compromised.
When risk becomes real
This challenge is particularly relevant in logistics. When a vehicle deviates from its route, stops unexpectedly or enters a high-risk area, timing becomes critical. The sooner an organisation becomes aware of an issue, the greater the opportunity to respond.
“When a vehicle goes quiet, that’s not a technical issue. That’s the moment risk becomes real.”
In South Africa, this challenge is amplified by organised cargo theft and the increasing use of signal jamming technologies designed to disrupt GPS and GSM-based tracking systems. When that happens, visibility disappears. And when visibility disappears, operational intelligence disappears with it. This is where many digital transformation projects begin to encounter difficulties.
Organisations spend significant resources building sophisticated systems to analyse data yet often devote far less attention to ensuring that the data itself remains available under real-world conditions.
Orange Logistics recognised this challenge early on. The company began looking beyond tracking and focused instead on a broader question: how do you maintain visibility when conditions are no longer ideal?
The video below explores how Orange Logistics approached this challenge and why resilient visibility has become a critical part of its operational strategy.
🎥 Watch the Orange Logistics customer story
Rather than replacing existing systems, Orange Logistics introduced Sigfox 0G technology as a complementary connectivity layer. The objective was not to collect more data. It was to ensure that critical data continued to flow when primary communication channels were unavailable. “If your system depends on one network, you’re exposed the moment that network is disrupted.”
Sigfox operates independently of traditional GSM infrastructure and is designed to transmit small, critical data messages over long distances using minimal power. That may sound simple, but simplicity is often underestimated.
In many IoT deployments, the most valuable piece of information is not a continuous stream of telemetry. It is a single event that indicates something has changed. A route deviation. An unexpected stop. A security breach. A movement alert. The value lies in knowing that something happened and being able to respond. “The difference between loss and recovery is often just a few minutes of visibility.”
The real challenge
This thinking is becoming increasingly important as organisations move toward AI-driven operations. The conversation is often focused on algorithms, automation and analytics. The real challenge is ensuring the quality, consistency and reliability of the data feeding those systems.
AI does not fail in the cloud. It fails at the edge. If data disappears during critical moments, predictive models become less accurate, automated workflows become less effective and decision-making becomes less reliable.
For Orange Logistics, resilient visibility has enabled faster response, reduced operational risk and improved confidence in the information being used to manage the business.
But the lesson extends far beyond logistics. Whether organisations are managing assets, infrastructure, utilities or supply chains, the same principle applies. Operational intelligence starts with operational visibility. And visibility only has value if it remains available when conditions become difficult.
“You don’t need perfect systems. You need to know you’re never completely blind,” says Atkinson.
As AI adoption accelerates and connected systems become increasingly central to business operations, organisations may need to rethink where they focus their attention.
The future of intelligent operations may not be determined by the sophistication of the algorithms. It may be determined by the reliability of the data that feeds them.
To learn more about how Sigfox South Africa is helping organisations build resilient visibility across logistics, asset tracking, utilities and infrastructure environments, visit sigfoxsa.co.za. For more information about Orange Logistics, visit www.orangelogistics.co.za.