In this blog post, you will learn about the limitations of traditional geocoders for last mile deliveries.
.png)
Geocoders typically map an address to the center of a building's roof, which is impractical for delivery drivers who require exact locations for parking and building entrances. This inaccuracy can increase delivery times and cause dissatisfaction among drivers and customers.
Naurt addresses these challenges with tailored location data that enhances route planning and delivery efficiency.
To understand this problem more deeply, you must first understand:
A geocoder is a service that converts addresses into locations, and vice versa. Typically, geocoding services operate as APIs. Forward geocoding converts a human-readable address into a computer-readable location. Reverse geocoding does the opposite - converting computer-readable locations into human-readable addresses.
Every place on Earth has a computer-readable location; we typically use latitudes and longitudes to represent this. Memorising all these locations as numbers is challenging, and it's hard to communicate them. Therefore, we also give every address a human-readable name. However, when we want to display the position of an address on a map, the map won’t know where that is. So, you need to convert it into a computer-readable location. The other way around too - maybe you know some computer-readable location and you want to find nearby addresses.
With a lot of work! Huge national survey efforts go into creating datasets for geocoders. Some time ago, every single address needed to be visited by a person and surveyed. These days, this process can be automated a little with satellites and computer vision. But even still, without having at least approximate locations of addresses first, this can only give you building outlines.
Creating general-purpose geocoders at scale introduces a whole new set of challenges, too. Addresses are constantly changing, for example, a building changing use. Buildings are converted into flats, thus creating new addresses, or flats are fused into a larger building once more. New buildings can be built and old ones demolished. Keeping all this data fresh requires constant monitoring.
Geocoders typically provide the centre of the building roof as the location for a given address. This would be the point given to a delivery driver on a map, and the location fed into a routing and ETA system. However, the driver isn’t aiming for the centre of the building roof - they first need to park somewhere suitable, then enter a door. With such a naive final destination, drivers might be routed in the worst cases to an incorrect street without parking. After eventually finding somewhere to park, that parking spot might not even be the closest one to the building entrance.
This process of the delivery driver attempting to find somewhere to park and enter the building can cost from 30 seconds up to 5 minutes (and in some rare cases, beyond). When a delivery driver is completing 80 to 120 deliveries a day, this wasted time adds up to 40 to 60 minutes. This has huge consequences: driver dissatisfaction leading to driver churn, customer dissatisfaction, failed deliveries, and fewer deliveries made.
Why not use a system which takes you to the optimal parking spot and building entrance, cutting out all this hassle?
Our database includes millions of addresses with the optimal parking spot and building entrance for each address. This provides you with accurate location data to assist not only your drivers but also your route planning tools, ensuring drivers park on the correct side of the road. We aim to prevent drivers from wasting time having to turn around or circle the block. Additionally, we help your ETA system understand how long it takes to complete the final 100 meters of a delivery. We provide better data to your system, helping you achieve a better delivery per hour rate, increase customer satisfaction, and decrease driver churn.
Sign up at Naurt for product updates, and stay in the loop!
