The problem every fuel station already knows
Fuel stations are one of the most heavily camera-covered environments in any industry. A typical European forecourt runs five to twenty cameras, recording around the clock. The infrastructure is already paid for. The footage is already being captured, and almost none of it gets acted on. Drive-offs are the clearest example. Every year, the average UK forecourt loses around £11,530 to drive-offs and unpaid fuel, roughly 28 incidents per station per quarter. Across the network, that adds up to more than £100 million in annual losses and an estimated 1.5 million incidents a year. The cameras captured every one of them.
The same pattern runs through every other incident type. Spills go unnoticed for minutes or hours. Unauthorised access to restricted areas happens overnight with no response. Equipment faults go unlogged until they become breakdowns. The cameras see all of it. The operations team sees almost none of it, not because the footage doesn't exist, but because no one can watch thousands of feeds across thousands of sites in real time.
The result is an estate of cameras that records everything and acts on nothing.
What Manako changes
Manako deploys Vision Agents directly on the cameras a fuel station network already owns. No new cameras. No new hardware. No engineers. No code. A Vision Agent is built for a specific operational situation, fuel spills, unauthorised access, vehicles in restricted zones, equipment left running, queue build-up at the forecourt. The operations team describes what matters. Manako builds the Vision Agent and deploys it across every relevant camera on every site. Vision Agents run locally on hardware already on site. Footage never leaves the premises. There is no cloud dependency and nothing new to install. How it works in practice
When a Vision Agent identifies an event, it writes a structured record directly into the dispatch, maintenance, or compliance systems the operator is already using (Slack, Whatsapp, Telegram, Email integrations). The right person receives the event on the device they already carry, and acts on it while there is still time to change the outcome.
A fuel spill gets flagged within seconds. Cleanup starts before it spreads. The station stays open. Unauthorised access outside of hours triggers an immediate security response, not a discovery the next morning. Every event is timestamped, linked to the camera, and stored with a clip of the evidence. The audit trail builds itself from the moment the network goes live. Deployment at scale
Manako has partnered with AVIA to deploy Vision Agents across their fuel station network in France, with a target of 400 stations. The deployment runs on cameras and hardware already on site, no new infrastructure, no engineers, no specialist staff required to operate it. As the operations team identifies new situations worth acting on, additional Vision Agents can be deployed across the network.
The economics
The cost model is simple. The cameras already exist. The hardware already exists. Manako adds the Vision Agent layer, the part that turns footage into action. No capital expenditure on new infrastructure. No ongoing cost of human reviewers. No six-figure integration projects per site. Vision Agents are sub-50MB, running on standard hardware at a fraction of the cost of dedicated systems. Why this matters
The pattern is not unique to fuel stations. Any industry that operates physical sites with cameras, retail, manufacturing, warehouses, stadiums, logistics, faces the same gap between what cameras capture and what the business acts on.
Manako closes that gap. One platform. Any camera. Every environment.
Join the waitlist to be the first to build with Manako, or get in touch to discuss an enterprise deployment across your operation.