Two jobs: knowing where vehicles are and deciding where they go
Software for delivery businesses usually combines two functions. The first is tracking: showing where vehicles are and what they have done. The second is route planning: deciding which stops go on which vehicle and in what order. They often come in one package, but they solve different problems and are worth evaluating separately.
How fleet tracking works
A tracking system typically uses a small device installed in the vehicle or an app on a driver’s phone. It reports position at regular intervals using satellite positioning and sends it over a mobile network to a dashboard. Depending on the product, it may also record:
- Speed and stops, including how long a vehicle was parked at each location.
- Harsh braking or acceleration, which some businesses use for driver coaching.
- Engine data such as fuel use and fault codes, where the device connects to the vehicle.
- Arrival and departure times at customer addresses.
For a dispatcher, this means answering the question “where is the van?” without phoning the driver. It also allows more accurate arrival estimates for customers and a record of when deliveries happened if there is a dispute.
How route planning works
Planning a day of deliveries is a version of a classic puzzle: given a set of addresses, vehicles and constraints, find a sensible order of stops. The number of possible orders grows very quickly as stops are added, so software usually uses methods that find a good answer rather than a perfect one.
Constraints the tools may handle include:
- Delivery time windows promised to customers.
- Vehicle capacity by weight or volume.
- Driver working hours and break rules.
- Different vehicle types, such as refrigerated vans.
- Traffic patterns, often estimated from historical data.
Some tools describe this as AI, and some of it uses machine learning to estimate travel times or the time spent at each stop. Much of the core planning is classic optimization. Either way, the output is only as good as its inputs. If addresses are misspelled, stop times are guessed too low, or a van’s capacity is entered wrongly, the plan will look neat and still fail. Related ideas about transportation routing at a wider scale also apply.
Where it helps small businesses
Hypothetically, a florist with three vans and thirty daily deliveries might spend a long time each morning arranging routes by hand. A planning tool could produce a first draft in minutes, which a dispatcher then adjusts using knowledge the software lacks, such as a customer who is never home before noon or a loading dock that is awkward for a large van. Tracking might then help the shop tell a customer when to expect the delivery. The value usually comes from saved planning time, fewer wasted miles and better customer updates, although the size of the benefit varies and should be measured rather than assumed.
Limits and pitfalls
- Travel time estimates are estimates. Roadworks, weather and local access problems are hard to predict.
- Plans meet reality. Drivers know about narrow streets, difficult parking and unreliable gates. If the planned route ignores this knowledge, drivers may work around the software, and the data becomes less useful.
- Rural or poor coverage. Tracking depends on signal. Gaps in data can make reports look odd.
- Over-optimizing. A route that saves five minutes but ignores a driver’s required break is not a good route.
- Messy data. Customer address lists with duplicates or missing apartment numbers cause trouble. Basic data entry quality has real effects.
Privacy and employee trust
Tracking a vehicle means tracking a person. Before installing devices, consider:
- Telling drivers clearly what is collected, why, and who can see it.
- Whether the vehicle is used privately outside working hours and whether tracking can be switched off then.
- How long location history is kept, and who is allowed to export it.
- Local laws on monitoring workers, which vary a lot. Employee monitoring is an area where checking with a qualified adviser is wise.
Businesses often get better cooperation when tracking is explained as a way to help with routing, safety and customer updates, and when the data is not used in ways drivers did not expect.
Choosing and trying a tool
A sensible approach is to define the problem first: late deliveries, too much planning time, high fuel costs, or customers asking where their order is. Then ask vendors what data the tool needs, what hardware and monthly fees apply, how data can be exported if you switch, and how it behaves when a stop is added mid-day. Test it on one vehicle or one route for a few weeks and compare the results with your notes from before. Include the total cost, covering devices, installation and training, rather than only the headline subscription. If it does not clearly help, it is fine to stop.