How to know if Mr. Müller and Mr. Muller are the same person

Esther Labrie
Posted by Esther Labrie Content and Brand Manager Tuesday, December 11, 2018 - 18:03

Esther Labrie is language specialist and content manager at Quadient. Joining the company in 2010, Esther specialized in upcoming themes in online marketing like digital communications, omni-channel and Big Data. Esther creates content that focuses on building a bridge between online marketing and customer centric selling. She enjoys music and literature and likes to spend time with friends and family.

Customer Experience Update
umlaut

The Internet and digital communications capabilities have allowed organizations in every industry to expand business beyond their home borders and attract customers from around the globe. If companies want to maintain their global footprint and continue to grow, it’s imperative that they can trust their global data.

 

For companies doing business internationally, many aspects of their business are reliant on the validity, completeness and accuracy of their global data. Customer data integration, master data management, compliance to laws and regulations and operational excellence are just a few of these aspects,… and then there are their customers.

 

emarketer

 

 It doesn’t matter where in the world customers are from, there is one expectation that every customer has no matter what their address—they expect companies to get it right, whether it’s the spelling of their name, their address or the offer that they’re sent.

 

efforts

 

Why global data is more challenging

They way computers store data is they assign a number to each letter and character. The standard code used to do this is called Unicode. It makes it possible to store and exchange over 95,000 characters—more than enough to accommodate the world’s many alphabets, scripts, ideograms and more. So what’s the problem then? There are a few.

1. Diacritics

A diacritic is a mark added to a letter in order to change the sound-value i.e. alter the pronunciation. Here are a couple of examples:

diacritics

In many data systems, it’s impossible to represent every diacritic. To compensate, IT will rewrite the rules that are applied. Many times these rules are rewritten so the spelling and pronunciation will be normalized to the organization’s home country.

 

For example, the German name Müller will be “normalized” to Muller in the Netherlands. This solution may be acceptable to head office, but what about the actual customer? How will they feel when they receive a correspondence from the company and instead of their name being spelled Müller, it’s spelled Muller?

 

2. Different writing systems

Many organizations employ more than one data writing system. For example, the way a customer’s name is spelled in the CRM may be one way and the way it is spelled in the billing system could be slightly different (i.e. Müller  vs. Muller). This small discrepancy can cause confusion and doubt as to whether Mr. Müller and Mr. Muller are the same person.

 

3. Transliteration

Transliteration is a process within a data system where the representation of letters in one script is converted into the letters of another script. Here is an example of a name that has gone through the transliteration process and how it is spelled in a variety of languages:

script

 

As you can see the issue with transliteration is, depending on which script it is converted to, it can look very different from the native script, making it difficult to confirm whether two names in two different scripts are the same person.

 

How to feel confident your global data is correct

 

When processing global data you want to determine, with as much probability as possible, that the names in your database are who they say they are and that the name associated with that person is correct.

 

In order to do this, your data system must incorporate robust, intelligent comparison methods. This comparison process is called “normalization”. Once a name has gone through the normalization process it will be translated in a common script that all the other names in the data base will be written in as well.

 

Once normalization has been completed, each name goes through one final stage of assessment called “matching”. During this stage, each name undergoes a phonetic comparison which determines, with high accuracy, whether or not the name refers to the same person.

 

As the final stage of assessment, matching is crucial because it is the final determination when checking an individual against sanction lists to prevent fraud, money laundering or financing criminal organizations.

 

Learn more about how to overcome the challenge of processing global data and the processes you require in order to do so. Download the white paper “How are you dealing with the challenge of processing global data?”