A practical and pragmatic approach to the implementation of data management that delivers quick wins is one of the key challenges of any data management professional. Sooner or later, you will deal with this at one point in your career.

In the series of presentations on Practical implementation or optimization of data management with the “Orange” model, I share with you my practical experience of the past 10 years. This experience has led me to develop a new model and practical method for the implementation and optimization of data management. This method is a collection of techniques and templates that can be used for performing various tasks related to the development and optimization of data management in your company. In this article, we discuss how to optimize a particular data management capability.

Using the “Orange” model for developing and performing a data management maturity assessment

Now, many companies have been running data management initiatives for several years.

What you often see in practice is that some data management capabilities have been developed better than others. I will share some striking insights I have received due to the data management maturity assessment review 2019. Companies put much effort into developing data management documentation and implementing tools. And at the same time, they often do not reach the expected results in implementing data management-related processes and delivering required artifacts. It gives us a good understanding of optimizing some existing data management functions.

In this article, let’s discuss the following questions:

  1. What are the commonly used data management capabilities?
  2. Why should companies optimize already implemented capabilities?
  3. How to optimize a particular data management capability to get quick wins?

Widely used data management capabilities.

Each company chooses its metamodel of data management. Still, some capabilities are commonly used in any company. The “Orange” model of data management includes several such capabilities. One of the key “Orange” model schemes can be seen in Figure 1.

Figure 1. The key components of the “Orange” model.

The “Orange” model describes data management as a set of business capabilities that enable data value chains. One of the key value propositions of data management is to deliver data of the required quality to internal and external stakeholders for different purposes. Data management sets up data value chains that transform raw data into meaningful information. Different data management capabilities should enable these data value chains. The core data management capabilities in the “Orange” model are data modeling, information systems architecture, data quality, and data governance. In Figure 1, they are marked orange. Data management professionals perform these capabilities. Other capabilities belong to different domains like IT, security, and other business support functions.

The key reasons to optimize a particular data management capability

The key reasons are listed in Figure 2:

Figure 2. Key reasons to optimize a particular data management capability.

  1. We need to improve particular capabilities to be able to deliver practical results. Quick wins are a key means to show the effectiveness of data management to top-management and key stakeholders.
  2. A capability can be more or less critical in delivering value to stakeholders. So, the most critical capabilities should get the most attention.
  3. The performance and effectiveness of each data management capability should be assessed and improved.
  4. You might also want to optimize the ratio between revenue/cost contribution.

Now, how can we do this?

A 7-step practical approach to optimize a particular data management capability

One of the models to answer this question is a 7-step approach shown in Figure 3. As an example, I will use data quality capability.

Figure 3. The 7-step approach for optimization of a data quality capability.

Step 1. Specify key stakeholders and their concerns.

First, you should identify the key stakeholders and their concerns.

Data quality has different groups of stakeholders, internal and external. Each group has different concerns and needs regarding the quality of data. I will share with you one practical example. Once, I delivered a workshop to a group of 20 professionals that formed a virtual working group. The group comprised business, IT, data management, and architecture professionals. They performed several data quality projects for different stakeholder groups, and their concerns and needs were completely different.

Step 2. Identify the business value proposition per stakeholder group.

Business value proposition means products or services or benefits associated with them. In our case, we decided to assess the value proposition for each stakeholder group. For that, we used the technique called Business canvas. This is a well-known technique that is usually used for the business modeling of an entire business. I have explained this method for a particular business function in one of my previous articles.

Step 3. Design data quality capability dimensions.

Four dimensions enable data quality capability. These are processes, tools, roles, and data. You can find the description of these dimensions in detail in the first article of this series.

The “Orange” model offers two different options to design a business capability.

Option 1. You describe each dimension as shown in Figure 4 one by one.

Figure 4. Option 1 is to design data capability dimensions.

You could start, for example, with the key outcomes of the capability. Examples are lists and definitions of data quality dimensions, data quality requirements, etc. To deliver these artifacts, corresponding business processes, such as gathering and providing data quality requirements, should be operational. Roles will be involved in performing the processes and will belong to each category of data stewards. Tools should enable the function of business processes.

Option 2. You first decompose the data quality capability into sub-capabilities at a lower level and then describe each sub-capability using four components. This option is shown in Figure 5.

Figure 5. Option 2 is to design data quality dimensions.

The final result should be similar to the result of option 1.

Step 4. Map business capabilities and value chains.

One of the high-level values of DQ capability is delivering data of the required quality. To produce this value, a company should perform a set of actions that form a value chain, as shown in Figure 6. Such activities include identifying DQ requirements, profile data, etc. To perform these steps, a company should possess specific capabilities such as data quality checks and controls management, DQ monitoring and reporting, data modeling, data lineage, stakeholder management, etc.

Figure 6. The mapping between value chains and business capabilities.

Step 5. Assess the current and desired level of maturity

The fifth step is the assessment of either maturity or performance of the chosen business capability. The “Orange” model offers you two methodologies. The first one is to perform a maturity assessment. I have explained this in depth in the third article of this series. There are also plenty of free materials on our website about this subject. Another option is to apply techniques that allow for assessing the performance of business functions. I will cover this topic in the seventh and last article of this series. We are returning to our case. If you recall, I mentioned three different data quality projects. Stakeholders and value delivered varied for each of the three projects. We have used this methodology to measure the maturity level for each data quality project. To our surprise, the level of maturity of each component varied. For one project, for example, processes were perfectly developed and implemented and got the ‘green’ status. While the same methods of other projects were assessed as ‘red.’

After assessing the current and desired maturity levels, it is time to move on to make a GAP analysis.

Step 6. Perform GAP analysis.

The differences between the current and desired levels of data management form gaps that need to be identified.

Step 7. Develop a roadmap and plans, and implement them.

Once you finish the analysis, you can proceed with the roadmap and its implementation.

In the following article, I will share my experience designing data management/governance roles.

For more insights, visit the Data Crossroads Academy site: //academy.datacrossroads.nl.