Fuel station analytics built for African forecourts
AutoVision is a fuel station analytics platform built in Nairobi by DataFly Ltd. It reads the CCTV cameras already watching your forecourt and turns them into live numbers: which pumps are busy, how long vehicles stay on each pump pad, where queues form, and how many pump customers walk into the shop.
What a station manager can see
- Live status of every pump pad: free, occupied or leaving.
- Completed pump sessions and pump utilisation, by hour and by pump.
- Average and median pump occupancy time, always shown with the number of sessions behind it.
- Queue build-up at the pumps, so you know when to add staff.
- Pump-to-shop conversion: how many pump customers visit the shop.
- Vehicle and footfall counts across the forecourt, entrances and shop door.
- Safety and service signals such as vehicles going the wrong way or crowd build-up.
- Daily, weekly and monthly executive reports.
How it works
AutoVision needs no changes to your pumps or point-of-sale. Standard IP cameras stream video to an AutoVision server, where computer vision detects and tracks vehicles and people, reads number plates, and turns what it sees into sessions and counts. The results appear on a web dashboard within seconds.
- A site survey decides camera positions: pump islands, entrances and exits, and the shop door.
- Cameras stream over the local network (RTSP) to the AutoVision server.
- Detection, tracking and plate reading run continuously on the server.
- Operators sign in to the dashboard from a phone or computer.
Numbers you can defend
A dashboard is only useful if you can trust it. In AutoVision the same figure is calculated once and shown identically on every screen and report, each average carries its sample size, and we label exactly what is measured. For example, the pump figure is called pump occupancy time because AutoVision sees a vehicle on the pump pad; it does not claim to know when fuel was being dispensed.
Read the full method on our page about how AutoVision measures pump occupancy time.
Who uses it
- Station managers running day-to-day staffing and pump allocation.
- Fuel retailers comparing sites and finding the slowest pumps.
- Commercial teams measuring how well the forecourt feeds the shop.
- Security teams who need an accurate record of who entered and when.
Frequently asked questions
What is fuel station analytics?
It means using data from the forecourt, such as vehicles, pumps, queues and shop traffic, to run the station better: staffing, pump allocation and shop conversion. AutoVision produces that data from camera video.
Do I need special cameras?
No. AutoVision works from standard IP cameras that can stream video over your network. Camera position matters more than camera brand, and we advise on placement during a site survey.
Does AutoVision measure how much fuel was sold?
No. It measures vehicles, pump-pad occupancy, queues and shop visits from video. It does not read pump meters or tank levels, so volumes and sales still come from your pump and POS systems.
Where is AutoVision available?
AutoVision is built in Nairobi and runs at a Nairobi fuel station today. If you operate stations elsewhere in Kenya or Africa, contact us to discuss a deployment.
How do I get started?
Contact DataFly Ltd for a site survey. We confirm camera positions and what the dashboard will show before you commit.
Keep reading
- Pump occupancyPump occupancy analytics for fuel stations: how long vehicles occupy each pump, which pump is slowest, and when queues form. Honest, defensible numbers.
- ANPR KenyaNumber plate recognition (ANPR) for Kenyan fuel stations: unique vehicle counts, pump-visit and shop links, repeat vehicles and hotlist alerts.
- How we measureThe exact method behind AutoVision's pump occupancy figures: what counts as a session, how averages are calculated, what is excluded and what limits remain.