BlackBerry didn’t lose because it ignored ordinary customers. It lost while it was winning them over. Its cheap prepaid data plan, launched by Vodacom in December 2008, opened BlackBerry to millions of South Africans who couldn’t afford a monthly contract. Then Apple built a general-purpose phone on infrastructure that already existed, with an ecosystem BlackBerry couldn’t match.
BlackBerry lost because its system couldn’t change when the market did. A general-purpose product can swallow a specialist whenever it can do the specialist’s job almost as well.
A specialist business survives only where ordinary technology genuinely can’t do the job: where the asset can’t carry a smartphone, the environment defeats normal sensors, or the terrain breaks normal signal coverage. Three conditions decide where that holds: equipment spread over a wide area, where physical checks are expensive and unreliable; weak or missing local back-office systems; and a large amount of valuable but unconnected equipment in places a smartphone-based solution simply doesn’t reach.
None of this is unique to South Africa – the same conditions apply across much of sub-Saharan Africa and beyond. Applied honestly, the test rules out most of what gets counted as opportunity in the big internet-of-things (IoT) market forecasts.
Analyst firm Omdia forecasts 35.6 million African devices connected through mobile networks by 2030, for cellular connections alone, before counting the equipment that runs on cheaper, non-mobile wireless instead. Add utility meters, farm equipment, delivery trucks and factory equipment across 54 African countries, and the Omdia figure isn’t a ceiling – it’s a floor, and a low one.
Logistics: a R577-million problem without a platform
South Africa’s connectivity layer is being built. Vodacom paid R1-billion for a company called IoT.nxt in 2019, and now runs systems for Eskom and other large clients. MTN built a marketplace called Chenosis, meant to open the mid-market to outside developers. Both are real investments with a stated goal, but the gap between announcing an open developer ecosystem and outside developers actually building on it usually takes years to close, not months. That gap, in the middle of the market, is not an accident. It’s a problem neither operator is set up to solve on its own.
South Africa’s trucking industry loses more money to theft than its own numbers can properly show. Over the 18 months to June 2022, industry group Tapa recorded 2 670 cargo theft incidents across all nine provinces, with documented losses of R577-million drawn from only 3.4% of victims who shared financial data. The real total is likely far higher. Trucking makes up 9-10% of South Africa’s economy and employs more than a million people.
The problem isn’t a lack of tracking devices. Fleet tracking already exists and gets defeated routinely: GPS jammers that blind a tracker without touching the truck, criminals posing as police to force a stop, forged paperwork used to collect goods from a warehouse that should never have released them. What’s missing is a trusted, real-time record proving three things: that the right goods left the right place, that they arrived undamaged and that payment was automatically triggered on confirmed delivery. That gap sits between two problems – one the finance chief owns, one the operations chief owns.
Why hasn’t someone built this already? At a recent industry conference, Tom Kruger of Hussar Security Solutions made the point directly: you cannot introduce new technology into an industry with no shared coordination across companies first. Every large operator builds tracking for its own fleet. Nobody has built a shared, trusted standard across competing operators, because no single company has a commercial reason to build infrastructure its rivals then use for free. A peer-reviewed study of South Africa’s ports, based on 24 interviews and two industry workshops, found exactly this pattern.
One large operator already has most of the pieces, built for itself. DSV, after taking over DB Schenker in 2025, runs a platform called myDSV offering proof of delivery, shipment milestones tied to invoicing, and GPS tracking marketed explicitly as theft protection. That covers two of the three missing functions. But it only covers DSV’s own fleet and customers. A shared standard across every operator would let customers compare DSV against smaller rivals on equal footing – the opposite of what the platform is currently used for.
That leaves two live possibilities. DSV, now considerably larger, could extend myDSV outward and become the industry standard everyone plugs into. Or it keeps the platform closed, because openness gives away the edge it currently provides DSV’s own customers. There is no evidence yet pointing either way. If DSV chooses the second path, the one better aligned with its own commercial interest, the coordination gap stays open precisely because the party best placed to close it has no reason to.
A platform built to close that gap creates real, measurable value for both people at once: less theft and a documented paper trail for the finance chief, fewer delivery disputes and automatic invoicing for the operations chief. All it needs is a standard mobile connection. The missing piece was never the mobile connection.
Smart metering: the gap the roll-out didn’t close
Eskom is 87% behind its own target for installing smart meters by 2027. Municipalities owed Eskom R98.5-billion in early 2025; that has since passed R110-billion, still rising despite the roll-out. National treasury’s response – a R2-billion programme installing meters and routing payment straight to Eskom – reaches only around 250 000 meters over three years, against a national base of millions of connections.
That response solves the wrong problem. People generally do pay their electricity bills. The failure happens after that, somewhere in the chain between collection and Eskom.
Maluti-a-Phofung proves the point. In May 2023 it signed an agreement handing billing to Eskom directly. Payment levels rose from 17-18% to 40% by October 2024, mostly from written-off debt and interest, then fell back to 24-25% by March 2025 and stayed there. Eskom’s own explanation: billing errors, wrong time-of-day pricing and prepaid vendors who collected money and never passed it on. That’s not a faulty meter; it’s a failure of how money moves through the system after collection, whatever billing technology sits underneath.
A platform-based solution has to do two things at once. It has to get billing and reconciliation right, closing the specific gaps Eskom named at Maluti-a-Phofung. And it has to settle directly with Eskom rather than routing funds through municipal treasury, where they don’t reliably arrive. The municipality keeps the customer relationship and its political standing, but not custody of the money in transit. Maluti-a-Phofung’s own agreement already routed billing to Eskom directly, and it still fell back to 24-25%. Direct settlement alone was not enough; it has to be paired with back-office accuracy, not substituted for it.
There’s already a legal route for a private company to take this role: a municipal public-private partnership, governed by its own chapter of the Municipal Finance Management Act. Midvaal is already part-way through procuring a 20-year electricity concession under this exact process. But the route is legally real and practically empty: a 2025 legal review found very few PPPs have ever been completed at municipal level. National treasury streamlined the national-level PPP rules in February 2025. It left the older municipal-level rules untouched.
Eskom is not a neutral bystander. It already occupies the exact role a private platform would compete for, and is expanding it: it has singled out the 14 worst-indebted municipalities for its own distribution agency agreements (DAAs), with reporting suggesting that could grow to 30. Under the three DAAs already signed, customer payments go straight into Eskom’s own bank account, with Eskom deciding the order in which that money gets applied, leaving the municipality getting, in one report’s words, “small change if anything”. Eskom has a direct commercial reason to keep expanding its own version of this role, not make way for a competitor.
The more realistic opening runs through Eskom, not around it. Its own DAA execution is demonstrably weak, so a private platform’s more plausible customer may not be a municipality shopping for a PPP partner. It may be Eskom itself, buying in the accuracy its own roll-out doesn’t deliver, while keeping the settlement relationship and the political credit. The technology already exists, inside the banks, mobile networks and payment processors that handle far more volume than South Africa’s 257 municipalities combined. What’s missing is the deal that connects that capability to Eskom, not the capability itself.
The operator that cannot be designed out
The word “operator”, by this point, means something broader than a mobile network: whoever holds the platform position. In both verticals examined here, a version of that operator already exists, and neither is a telecoms company. Eskom runs the closest thing to a metering platform. DSV runs the closest thing to a logistics verification platform. Neither is finished – Eskom’s version is badly executed, DSV’s doesn’t extend past its own network – but the space isn’t empty. It’s occupied, imperfectly, by incumbents who got there first, and neither needed to be a mobile operator to get there.
That changes the question. It isn’t whether the platform layer gets built. It’s whether the incumbent finishes the job, or whether the gap in its own execution stays open long enough for someone else to get in. For metering, whoever closes that gap most plausibly sells accuracy to Eskom rather than trying to replace it. For logistics, the outcome depends on a decision DSV hasn’t obviously made yet. A new specialist without an existing foothold in either market has a narrower opening than the size of the underlying problem suggests.
Simpler solutions, built on infrastructure that already exists, tend to arrive before the purpose-built alternative is ready. BlackBerry learned that lesson while it was still winning. The specialist that survives will be the one that builds its platform before a general-purpose solution makes it pointless – and understands that the time it has to do that is shorter than it looks.
- The author, Pambos Soteriades, has spent 28 years in mobile telecommunications, including executive roles at Vodacom Group and Telkom Kenya. He is not affiliated with, employed by or invested in any operator, institution or company mentioned in this article
- Subscribe to TechCentral’s daily newsletter
- Get breaking news alerts on WhatsApp
