Why I built
PetroControl.

The software industry abandoned you by design.

Fleet software is thirty years old. In its entire history, it was never built for fleets under two hundred vehicles. Not by accident. By unit economics.

The incumbents (Geotab, Frotcom, Cartrack, MiX Telematics) bundle a hardware tracker with field service, multi-week onboarding, and a six-month sales cycle. The minimum viable contract value sits above thirty thousand dollars a year. A thirty-vehicle fleet cannot justify the capex. A hundred-vehicle fleet might, but only after an approval committee that will not exist for another two quarters.

Geotab serves four million vehicles by going aggressively upmarket. The mid-tier, fleets of ten to two hundred vehicles, was structurally abandoned. Construction subcontractors, regional distributors, private security companies, FMCG haulers, mining contractors, agricultural cooperatives. The operators that move the actual economy. Skipped. I am building for them, at a price they can absorb, on infrastructure they already own.

Why I built it now, not five years ago.

Four conditions converged in the last eighteen months. None of them existed when the previous cohort of African logistics startups tried this and failed. Five years and 1.8 billion dollars in venture capital, mostly deployed against the wrong product: routing apps and matching marketplaces that sat on top of the system, never inside it (Breega, Africa Logistics & Transport, 2026 outlook).

First, the cost of building per-customer software at the bottom of the market collapsed by two orders of magnitude. The same AI economics that broke the incumbents are the economics I run on.

Second, the African Continental Free Trade Area moved from treaty into operational reality. Trucks now cross borders at scale, burning fuel that nobody audits.

Third, the subsidy era ended. In Angola, the price of diesel tripled in eighteen months. Fuel theft was tolerable when fuel was cheap. It is no longer.

Fourth, the African venture cohort matured. The next thesis demands platforms that own the underlying cost, not thin software layers that decorate it.

How big is the leak, really.

PetroControl recovers between fifteen and thirty percent of monthly fuel cost. That range is not a marketing claim. It is the conservative middle of what published industry research has been measuring for years, and I list the sources at the foot of this page so you can check the figure yourself.

WEX, the largest fleet payments network in the United States, published a 2024 benchmark placing uncontrolled fleet fuel leakage at approximately twenty percent of fuel spend. MiX Telematics, the South African telematics provider with the longest fleet dataset in the region, places the figure between twenty-two and thirty-six percent for fleets without real-time validation. Frost & Sullivan, in its African commercial vehicle telematics analysis, reports fleet fuel losses between eighteen and twenty-five percent across operators without instrumentation.

The fifteen to thirty percent I commit to is the conservative floor of the published evidence, not the ceiling. I lead with the lower end because I would rather under-promise and be proven right twice.

What I built.

Every driver in Angola already uses WhatsApp. Not most of them. Every one of them. It runs on a five-year-old Android with three megabytes of free storage. It works on a connection that drops every fifteen seconds. It is the only piece of software a driver opens more than once a day, every day, without being asked.

When I did the math on building a dedicated app, the numbers never closed. Installation friction. Support tickets. Training sessions. Lost devices. The quiet, correct resistance from drivers who read another app as another surveillance tool.

WhatsApp closed the math. The driver sends a photo and three numbers. The bot validates against the vehicle's historical norm, reads the receipt with a vision model, and surfaces a deviation to the manager the moment anything is off. Ninety seconds, beginning to end. Nothing to install. Nothing to teach. Nothing to lose when the phone breaks. The whole loop closes before the driver leaves the station. Every refuel becomes an auditable financial event. The audit trail is the product.

What I will not do.

I will not turn PetroControl into a GPS tracker. Phones are unreliable for that work, and hardware is somebody else's business. I will not pretend that yelling at drivers is the answer. The system has to make the right behaviour the easy behaviour, or it does not work. I will not chase logos I have not earned. And I will not ship a feature that adds friction to a flow that costs the operator nothing today.