SmartMonkey, the easiest and fastest route planning platform in the market.

App illustration humans

Common Geocoding mistakes and how to avoid them

Reduce costs and maintain quality of service with AI-driven Geocoding

geo heatmaps geo
SmarMonkey allows you to track performance of your field operations

Hi, I’m Xavi Ruiz, CEO at Last week, I was talking with Muthu from Geoawesomeness about what we do at SmartMonkey. During  this conversation, I remarked that non-geo specialists have issues because of their lack of knowledge about geocoding systems. They think that it’s not a problem anymore, so when you talk to them about the quality of their master data they say: “Come on! We have Google Maps, right?”. It’s a normal reaction!

geo MAPS
Don’t rely 100% on google maps for Geo-coding

The magic wand of geocoding

Google has a good reputation as a tech giant and works well in urban scenarios. But if we ask them “Have you ever got lost using Google Maps?” it triggers bad experiences in their minds. I am a heavy user of Google Maps and I can relate to that.  

The principal question that arises is why Google Maps users rely so much on its geocoding?

Geocoding, for non-experts, means to provide geographical coordinates corresponding to a location.

They trust Google Maps because of their experience. They don’t understand how it works and it’s magic for them: they enter an address and Google magically puts a pin on a map. Magic feels perfect!  

But the bad news is that Google Maps is far away from perfection.

A study at the Pittsburgh University (2009)  about the quality of the market concludes that geocoders do not deliver enough quality. In short, 60% of data is accurate enough (< 200m), 20% have enough error which is useful for a route optimizer and the rest is not even a result (manual  geocoding is still needed).* (See reference at the end of post)

geo stats
Results from the study* at the Pittsburgh University(2009) about the quality of the geocoders (page 1090 in vol.24, No.7). See link below

Achieving a 60-70% of correctness is far from magic or perfect.  

Logistic operations cannot rely on results with an error around 30% on geolocalised places.  

When geocoding faces reality

Why is Google not magically working? There are several reasons:  

  • The main reason is the lack of information on the country side compared to the urban areas.  
  • Another reason is that names change. Streets, squares, avenues get renamed over time by the administration. Information needs to be up-to-date. And the system has to take into account that users can still  look for the old names when thinking about a place. Sometimes places  have cozy names that are just familiar to the village citizens. But these places don’t exist for Google Maps.  
  • Multilingual regions have different translations and multiple combinations for the same place.  

Living in the non-existing Star

I have a personal story to tell as an example of this trouble with Google Maps.  

My wife lived in a town close to Girona called Olot. This region is famous for its political majority towards Catalan nationalism. My wife lived at a square officially called “Plaça Espanya”. To avoid this name, the inhabitants of Olot refer to it with the cozy name of “Plaça Estrella” (star square). No need to ask why.  

geo placa estrella gmaps
Plaça Estrella. Satellite image. Copyright by Google Maps

Now you can imagine the huge problems that DHL or UPS workers had to deliver to this square!

Relying on human knowledge

How people are currently solving this issue? Obviously by making mistakes and learning from them. It’s that simple because none of us was born with full knowledge. Knowledge about the places is acquired and shared between people. Human knowledge fills the gaps of information.  

Since people make mistakes, a 100% quality service can’t be achieved with this method.

Loss of information

Not only workers make mistakes, in addition, they are not always available. The efficiency of operations depends highly on experienced people. This shared knowledge is one of the most valuable assets of a logistics company.   They realize the value of their workers when they have to replace them. It’s impossible to replace the knowledge and at the same time keep a consistent level of quality of service.  

This is a huge risk for companies that are not aware of this problem.  

Artificial Intelligence to the rescue

Artificial Intelligence is able to learn from the operations as humans do. It does learn automatically, without mistakes and AI is always available. SmartMonkey helps companies in their logistic operations and reduces their costs up to 30%.

Case Study: Suez

Suez is the biggest utility firm in the world. As a water supplier company, they have hundreds of thousands of places which are regularly visited: water meters, valves and others.

geo SUEZ 1
Geolocated water meters, valves and others. Copyright 2019 SmartMonkey

Case Study: Suez  

In some remote areas, there is just one person responsible for all the  company operations. Just one person has the knowledge of the geolocation. Finding these water meters and valves is a real challenge.  Google Maps won’t be of any help.  

Suez faced the problem when a worker needed to be replaced. SmartMonkey proved that AI helps to keep the quality of service and reducing risk, time and money.  

How did we do that? SmartMonkey processed the GPS track from Suez logistics and combined with the business data (water-meter readings) to geolocate automatically the places to visit. The result was:  

  • More than 99% of places where geolocated with a precision of maximum 25m distance
  • 60% with a precision of maximum 2m distance

Capturing the knowledge automatically makes the information more reliable and gives clear insights about the operation.  

SmartMonkey helps companies overcome challenges when geolocating and optimizing routes.  

Check out our website to know more about our solutions.  

Stay tuned for our next article about human behaviors!

* * * Reference * * *
Roongpiboonsopit, D. and Karimi, H.A., 2009. Comparative evaluation and analysis of online geocoding services. International Journal of Geographical Information Science, 24(7), pp.1081-1100. Source: