Data Quality Dimensions in Motion

For the fifth year Dan Myers of DQMatters is making an Annual Dimensions of Data Quality Survey.

There are some very interesting findings when looking at the trend in the previous years surveys as seen in the figure below.

Data Quality Dimensions 2015 to 2018

Among the data quality dimensions included in this survey we see that the use of consistency, validity and not at least completeness has increased significantly over these years.

The possible use of consistency and completeness was examined here on the blog in the post Multi-Domain MDM and Data Quality Dimensions. Another dimension included in this post was uniqueness, which is a frequently addressed data quality dimension for customer master data in the quest of fighting duplicates in databases around.

You can now be part of the 2019 Annual Dimensions of Data Quality Survey here.

Even With 20 Entities MDM Can be Hard

This week I attended the Master Data Management Summit Europe 2018 and Data Governance Conference Europe 2018 in London.

Among the recurring sessions year by year on this conference and the sister conferences around the world will be Aaron Zornes presenting the top MDM Vendors as he (that is the MDM Institute) sees it and the top System Integrators as well.

Managing an ongoing list of such entities can be hard and doing it in PowerPoint does not make the task easier as visualized in two different shots captured via Twitter as seen below around the Top 19 to 22 European MDM / DG System Integrators:

20 entities

Bigger picture available here.

Now, the variations between these two versions of the truth and the real world are (at least):

  • Red circles: Is number 17 (in alphabetical order) Deloitte – in Denmark – who bought Platon 5 years ago or is it KPMG.
  • Blue arrow and circles: Is SAP Professional Services in there or not – and if they are, there must be 21 Top 20 players with two number 11: Edifixio and Entity Group
  • Green arrow: Number 1 (in alphabetical order) Affecto has been bought by number 8 CGI during this year.

PS: Recently I started a disruptive list of MDM vendors maintained by the vendors themselves. Perhaps the analysts can be helped by a similar list for System Integrators?

Data Quality in Different Languages

The term ”data quality” exists in many different languages.

As reported in the post Häagen-Dazs Datakvalitet, the Scandinavian word for data quality is datakvalitet. Well, actually there is no such language as Scandinavian, but datakvalitet is used in Danish, Swedish and Norwegian all together. Maybe even in both Norwegian languages, though Google Translate only know of one Norwegian language.

In other Germanic languages the words for data quality are close to datakvalitet. In German: Datenqualität. In Dutch: Datakwaliteit.

The above terms are compound words. Even though English is also classified as a Germanic language we see a Latin influence as “data quality” is two words in English. And that goes for all English variants. It is only when it comes to if we have to standardise this or standardize that we are in trouble. British English is best when we have to select if data quality improvement is a program or a programme.

In true Latin languages we have three words. French: Qualité des données. Spanish: Calidad de datos.

And then there are of course terms in other alphabets than latin and other script systems:

data quality in different languages

Bookmark and Share

Häagen-Dazs Datakvalitet

There is a term called foreign branding. Foreign branding is describing an implied cachet or superiority of products and services with foreign-sounding names

Häagen-Dazs ice cream is an example of foreign branding. Though the brand was established in New York the name was supposed to sound Scandinavian.

However, Häagen-Dazs does sound and look somewhat strange to a Scandinavian. The reason is probably that the constellation of the letters “äa” and “zs” are not part of any native Scandinavian words.

By the way, datakvalitet is the Scandinavian compound word for data quality.

Getting datakvalitet right in world wide data isn’t easy. What works in some countries doesn’t work in other countries, not at least when we are talking datakvalitet regarding party master data such as customer master data, supplier master data and employee master data.

One of the reasons why datakvalitet for party master data is different is the various possibilities with applying big reference data sources. For example the availability of citizen data is different in New York than in Scandinavia. This affects the ways of reaching optimal datakvalitet as reported in the post Did They Put a Man on the Moon.

As part of the ongoing globalization handling international datakvalitet is becoming more and more common. Many enterprises try to deploy enterprise wide datakvalitet initiatives and shared service centers handles party master data uncommon to the people working there. This often results in finding a strange word like Häagen-Dazs.

Bookmark and Share