Data Governance

Hors Catégorie

Henrik Gabs LiliendahlLeave a comment

Right now the yearly paramount in cycling sport Le Tour de France is going on and today is probably the hardest stage in the race with three extraordinary climbs. In cycling races the climbs are categorized on a scale from 4 (the easiest) to 1 (the hardest) depending on the length and steepness. And then there are climbs beyond category, being longer and steeper than usually, like the three climbs today. The description in French for such extreme climbs is “hors catégorie“.

Within master data management categorization is an important activity.

We categorize our customer master data for example depending on what kind of party we dealing with like in the list here called Party Master Data Types that I usually use within customer data integration (CDI). Another way of categorizing is by geography as the data quality challenges may vary depending on where the party in question resides.

In product information management (PIM) categorization of products is one of the most basic activities. Also here the categorization is important for establishing the data quality requirements as they may be very different between various categories as told in the post Hierarchical Completeness.

But there are always some master data records that are beyond categorization in order to fulfill else accepted requirements for data quality as I experienced in the post Big Trouble with Big Names.

Bookmark and Share

Howcatchem

Henrik Gabs Liliendahl3 Comments

I’m sad to learn that Peter Falk has died. Peter Falk is most known as Lieutenant Columbo in the American television series Columbo, which has been shown all over the world including in my country when I was a teenager.

I think Columbo would have been a great data quality specialist too. Underestimated by the smart guys but focused on the important details while exercising the art of “howcatchem”. Opposite to his “whodunit” colleagues who are sitting on a horseback chasing the villains down a crowded city street or doing the same in a car while smashing most other cars on the way.

Bookmark and Share

Book Review: Berson and Dubov on MDM

Henrik Gabs Liliendahl3 Comments

A few days ago Julian Schwarzenbach over at the Data and Process Advantage Blog published a review of the book “Master Data Management and Data Governance” by Alex Berson and Larry Dubov. Link to Julian’s review here.

And hey, that’s the book I have been reading too during the last months. So why not make my review too.    

I agree very much with Julian’s positive review of the book. It is a very comprehensive book – and thick and heavy I have learned from bringing it with me on travel which is where I usually read offline stuff. But master data management and related data governance is a big and heavy discipline with a lot of details that has to be dealt with.

Probably I have annoyed fellow travellers in trains and airplanes while reading the book with exclamations as: Yes, precisely, that’s what I always have said, good point and so on. Because I agree very much with many of the issues described and the solutions discussed in the book.

For the mandatory bit of criticism that must be included in every book review I will bring on my pet bashing about United States and English language centricity. Well, it’s actually not that bad, as the book at many places does indicate that other angles and pains exist than those being prominent in the United States and with the English language.

Oh, and I bear with that  my surname in the references are spelled “Sorensen” instead of “Sørensen” and that a related date are formatted like “11/22/2009” which will be the 11th day in the 22nd month of the year 2009 to me.     

Bookmark and Share

Customer Relationship Mess (CRM)

Henrik Gabs LiliendahlLeave a comment

I have several times witnessed how a sales department for a lot of good reasons has forced the implementation of a CRM (Customer Relationship Management) software package disconnected from the ERP (Enterprise Resource Planning) system and other applications where customer master data have been handled until then.

The good reasons have been that the current applications didn’t fit the business processes in a dynamic sales department and perhaps that the current monolithic enterprise solution was too inflexible for the business needs in sales.

While this move may have been a great success in sales force automation the downside is often that the single customer view has been limited to a single customer view seen from the windows in the sales department offices.

In order to have a 360 degree view of customer you have to cover all the view points in the enterprise embracing all departments being in contact with the customer and thereby accessing and maintaining customer master data.

Those who feel the pain when a company doesn’t maintain such a view is the customer and those who enjoys when a company have that view is the customer.

Lately I had two experiences as a customer. A bad experience facing a lousy approximately 110 degree customer view from a phone company and a well executed 360 degree view from an insurance company. Both cases haven’t been around one of my favorite subjects being identity resolution. Both companies have my citizen ID.

It is just so that some companies cares more about single department business needs than true customer relationship management. IT’s a mess.

Bookmark and Share

Where is the Business?

Henrik Gabs Liliendahl6 Comments

In technology enabled disciplines we often like to divide an organization into two distinct parts being IT (Information Technology) and “the business”.

I am aware that we do that to emphasize that our solutions has to be business centric opposite to technology centric. We mustn’t fall into the trap of discussing technology too early and certainly not selecting certain technology brands as the first step of our solutions.

A problem however is where to find “the business” in an organization. The top management surely represents all of the business (including the IT part of the business). But in order to find the so called subject matter experts we are looking down the levels in the organization where people don’t belong to “the business” but to sales, marketing, customer service, purchase, production, human resources, finance and so on.

Some technology enabled disciplines belong to a certain department. But disciplines as (enterprise wide) data quality and master data management are supposed to support most departments. The business. So where do we find the business? And who are we by the way?

Call them?

Assuming it doesn’t matter who we are: Let’s go find “the business”. I guess it doesn’t help calling the reception and ask them to put us through to “the business”. Actually the manned reception probably doesn’t exist today. And it will be surprising to get a machine asking:

  • Do you want to speak with IT? Press 1.
  • Do you want to speak with “the business”? Press 2.

If we are in my home country Denmark we also have a linguistic issue. If I ask google to translate “the business” from English to Danish I get the word “forretningen”. If I ask google to translate “forretningen” from Danish back to English I get the word “shop”. So calling “forretningen” will probably get me to the shop floor. Not a bad place, a true gemba, but maybe not the only one.

Everyone belongs to “the business”

In data quality and master data management there is a question used all over to exemplify a common challenge within these disciplines.

The question is: What is a customer?

The challenge is that people from different departments will have different definitions. Marketing defines a customer one way, sales tend to do it a bit different, finance sees it yet in another way and production has their view point. And the stereotype IT guy defines a customer as a row in the customer table.

So now we are asking for Alexander the Great from “the business” to come cutting the Gordian Knot.

That is probably not going to happen.

More likely someone from any business unit will be able to negotiate a proper conceptual solution covering all requirements from the different business units. And from what I see around it may often be someone who’s human resource master data record is related to the IT part of the business. Or was. The main point is having a holistic view of the business where everyone belongs.    

Bookmark and Share

Things Change

Henrik Gabs Liliendahl7 Comments

Yesterday I posted a small piece called So I’m not a Capricorn? about how astrology may (also) be completely wrong because something has changed.

On the serious side: Don’t expect that because you get it Right the First Time then everything will be just fine from this day forward. Things change.

The most known example in data quality prevention is probably that it is of course important that when you enter the address belonging to a customer, you get it right. But as people (and companies) relocates you must also have procedures in place tracking those movements by establishing an Ongoing Data Maintenance program in order to ensure the timeliness of your data.

The other thing, so to speak, is that having things right (the first time) is always seen in the context of what was right at that time. Maybe you always asked your customers for a physical postal address, but because your way of doing business has changed, you actually become much more interested in having the eMail address. And, because What’s in an eMail Address, you would actually like to have had all of them. So your completeness went from being just fine to being just awful by following the same procedure as last year.

Predicting accuracy is hard. Expect to deal with Unpredictable Inaccuracy.       

Bookmark and Share

A Data Quality Immaturity Model

Henrik Gabs Liliendahl7 Comments

There are several maturity models related to data quality out there. I have found a good collection in this document from NASCIO.

I guess the mother of all maturity models is the Capability Maturity Model (CMM). This model is related to software development.

There is also a parody model for that called the Capability Immaturity Model (CIMM). Inspired by an article yesterday by Jill Dyché on Information Management called Anti-Predictions for 2011 I have found that the CIMM model is easily adapted to a data quality immaturity model with levels from zero to minus three as this:

0 : Negligent

The organization pays lip service, often with excessive fanfare, to implementing data quality processes, but lacks the will to carry through the necessary effort. Whereas level 1 assumes eventual success in producing and measuring quality data, level 0 organizations generally fail to have any idea about the actual horrible quality of the data assets.

-1 : Obstructive

Processes, however inappropriate and ineffective, are implemented with rigor and tend to obstruct work. Adherence to process is the measure of success in a level -1 organization. Any actual creation of quality data is incidental. The quality of any data is not assessed, presumably on the assumption that if the proper process was followed, high quality data is guaranteed.

-2 : Contemptuous

While processes exist, they are routinely ignored by the staff and those charged with overseeing the processes are regarded with hostility. Measurements are fudged to make the organization look good.

-3 : Undermining

Not content with faking their own performance, undermining departments within the organization routinely work to downplay and sabotage the efforts of rival departments. This is worst where company policy causes departments to compete for scarce resources, which are allocated to the loudest advocates.

Bookmark and Share

Christmas Tree Options

Henrik Gabs Liliendahl3 Comments

Today the last Sunday before Christmas seems to be a good day for selecting a Christmas tree.

We are considering two different options:

  • As most times before we will find a tree as wide and high as possible for the room so it may be decorated with as much of different stuff we have collected during the years as well as some of the precious things passed down from previous generations. It will be cut over the root, but that’s not a problem since we will throw it away after Christmastide.
  • Another option is having a smaller tree still with the root on planted in a pot. We will then have to carefully select the decoration. The advantage is that it can be reused on the terrace during the year and then, a little taller, as Christmas tree again next year.   

Well, not that different from the considerations about data quality, data warehouse and business intelligence projects and programs from my workdays.

Bookmark and Share

Now, where’s the undo button?

Henrik Gabs LiliendahlLeave a comment

I have just read two blog posts about the dangers of deleting data in the good cause of making data quality improvements.

In his post Why Merging is Evil Scott Schumacher of IBM Initiate describes the horrors of using survivorship rules for merging two (or more) database rows recognized to reflect the same real world entity.

Jim Harris describes the insane practices of getting rid of unwanted data in the post A Confederacy Of Data Defects.

On a personal note I have just had a related experience from outside the data management world. We have just relocated from a fairly large house to a modest sized apartment. Due to the downsizing and the good opportunity given by the migration we wasted a lot of stuff in the process. Now we are in the process of buying replacements for these things we shouldn’t have thrown away.

As Scott describes in his post about merging, there is an alternate approach to merging being linking – with some computation inefficiency attached. Also in the cases described by Jim we often don’t dare to delete at the root, so instead we keep the original values and makes a new cleansed copy without the supposed unwanted data for the purpose at hand.

In my relocation project we could have rented a self-storage unit for all the supposed not so needed stuff as well.

It’s a balance. As in all things data quality there isn’t a single right or wrong answer to what to do. And there will always be regrets. Now, where’s the undo button?

Bookmark and Share

Snowman Data Quality

Henrik Gabs Liliendahl6 Comments

Right now it is winter in the Northern Hemisphere and this year winter has come earlier than usual to Northern Europe where I live. We have already had a lot of snow.

One of the good things with snow is that you are able to build a snowman. Snowmen are beautiful pieces of art but very vulnerable.  Wind and not at least rising temperatures makes the snowman ugly and finally go away sooner or later.

Snowmen have this unfortunate fate common with many data quality initiatives.

Many articles, blog posts and so on in the data quality realm focuses on this fate related to technology based initiatives. The common practice of executing downstream cleansing of data using data quality tools is often criticized. As a practitioner in this field I have to admit that: Yes, I am often making the art of building snowman data quality.

An often stated alternative to using data quality tools is improving data quality through change management including relaying on changing the attitude of people entering and maintaining data. Though it’s not my area of expertise I have seen such initiatives too. And I am afraid that I am not convinced that such initiatives unfortunately also sooner or later have the same fate as the snowman.

As said, I’m not the expert here. I am only the little child watching how this snowman is exposed to the changing winds in many business environments and how it finally disappears when the business climate varies over time.

Now, this is supposed to be a cheerful blog about happy databases. I am ready for getting into some warm clothes and build a beautiful snowman of any kind.  

Bookmark and Share