Data management maturity is a widely debated topic. The word ‘maturity’ can have different meanings put into different contexts. For this article, I will use the following definition: ‘Maturity is a measurement of the ability of an organization to undertake continuous improvement in a particular discipline’ (Maleh 2018, Security and Privacy Management, Techniques and Protocols). To ensure continuous improvements, a maturity assessment allows a company to do the following:
- to assess the ‘as is’ status
- to specify the desired ‘to be’ situation
- to evaluate the gap between ‘as is’ and ‘to be.’
- to benchmark the results against peers in the industry.
A company creates a foundation for developing strategies, roadmaps, plans, and actions to reach the ‘to be’ status by doing a maturity assessment.
The advantages of using a maturity assessment for a data management function seem apparent. Each company manages data in one way or another. Some companies have already formalized the data management function. Others have not. Regardless of this, any company can still measure its data management maturity.
Maturity assessment should not be a ‘one-off’ event. The key aim is to develop a tool that will assist the company in constantly measuring its performance and progress and benchmarking the results against other companies.
If a company decides to measure its data management maturity, there are a couple of questions that need to be answered first:
- Is it possible to use existing data management/governance models?
- What is the definition of ‘data management’ within the company?
- How can our data management maturity be measured?
In this article, you will find answers to each of these questions. Let’s start with the first one and directly jump into the challenges of the existing models.
Challenges with the existing data management models
Several well-known data management/governance maturity models like DAMA-DMBOK2, DCAM, CMMI CERT-RMM (Data Management Maturity Model) by CMMI, IBM Data Governance Council Maturity model, Stanford Data Governance Maturity Model, Gartner’s Enterprise Information Management Maturity Model. You can find the detailed analysis and comparison of these models in my article’ Data management maturity models: a comparative analysis.’
There are several key challenges with the usage of these models:
- All the above-mentioned maturity models can hardly be compared as they have substantial differences in their metamodels.
- Each company that wants to assess its data management maturity should first align/map the metamodel of data management used in their company with the chosen maturity model.
- The two above-mentioned challenges do not allow companies to reach one of the key goals of maturity assessment: to benchmark the maturity status of their company against their peers in the industry.
These challenges led me to the idea of developing a standard metamodel (the ‘Orange’ model of data management) that can be used for both the implementation of data management and maturity assessment. I have described the key principles of this model in the article ‘What does data management have in common with… an orange?’ and my book The “Orange” Model of Data Management.
Now let us find the answers to the other two questions mentioned above. I will do it by demonstrating a way for a company to design and perform a data management maturity assessment.
Design and perform a data management maturity assessment and benchmark the results.
A company needs to take the following steps to perform a maturity assessment and benchmark the results.
Step 1. Specify the definition, scope, and key sub-capabilities of data management.
The basic data management model includes five core sub-capabilities of data management, as shown in Figure 1. But first, let’s agree on the definition of data management. Data management is a business capability that safeguards the company’s data and information resources. It optimizes data and information value chains (further in the text, ‘data chain’) to ensure the effective conduction of business. Data chains support the delivery of key business value propositions. A set of key data management sub-capabilities provides an optimized design of these chains. These sub-capabilities traditionally include (but are not limited to) data management framework (data governance), data modeling, information systems architecture, and data quality. The list of these sub-capabilities depends on the chosen perspective on data management (broad or narrow). This perspective relies on the relationship between data management and the IT function. Supporting IT and other sub-capabilities contribute to the optimized functioning of data chains.
Figure 1. Key data management sub-capabilities
Each company should first specify its understanding and scope of data management. The next step is the identification of data chains that relate to the key business value chains. Such analysis will result in a list of key data management and IT-related sub-capabilities that will ensure the performance of identified data chains.
Very often, a company has already specified company-specific data management capabilities and deliverables that do not match the standard models. It is a good solution but does not allow benchmarking the results against other companies. To align the company’s vision of data management with peers in the industry, a company should map with a standard data management and data management maturity model.
Step 2. Map the company’s data management sub-capabilities with the standard model
First, to make such a mapping, let’s specify the definition of data management (sub-)capability and its dimensions.
A (sub-)capability is the ability of a company to reach some goals and deliver some outputs. In our case, these are data management outputs. The following dimensions enable a (sub-)capability: role, process, data (input and output), and tools.
A role describes the participation of people in business processes. Roles can represent business units, functional jobs, a set of data management-related accountabilities, etc.
A process means a business process on different levels of abstraction.
Data in this context means output artifacts of a specific process. In the broader context, data can also represent input and output data required to conduct a process.
Tools include information technology systems, applications, and resources required to enable the (sub-)capability, i.e., budgets.
The full data management capability and separate sub-capabilities can be described using these four dimensions.
Below in Figure 2, you can see an example of the sub-capability of a data management framework from the ‘Orange’ model13 with a detailed description of each of the four dimensions.
Figure 2. Four dimensions of data management framework capability
Now you can map your company’s data management model’s processes, roles, deliverables, and tools to the standard one.
Step 3. Specify maturity levels and define indicators (KPIs)
As shown in my previous article on data management maturity models, no standard number of maturity levels exists. This number varies between five and six. So, to proceed with developing a maturity model, a company should first specify the measurement indicators and the number of maturity levels. In the ‘Orange’ model, each indicator is linked to a specific sub-capability and to one of four dimensions, as shown in Figure 3. In the Maturity Scan which is based on the ‘Orange’ model, each question is linked to a specific indicator.
Figure 3. The maturity measurement approach
If your company decides to develop its maturity model, you should specify indicators per data management capability and then per dimension of capability.
The specification of standard indicators with related levels of maturity per each indicator will allow your company to:
- to measure the ‘as is’ level of maturity
- to specify the desired level of maturity
- to focus on the most important capabilities and its dimension.
The last step will be benchmarking your achievements against your peers in the industry.
Step 4. Benchmark the company’s achievements with peers in the industry
Benchmarking is not a mandatory step, but it is still interesting to compare the achievements of your company with those of others. Because there are plenty of different data management/governance maturity models, it isn’t easy to get a common basis for comparison.
If you don’t know where to start, I invite you to perform your own Data Maturity Scan HERE. Afterward, you can compare your results with the average results of other participants, presented in ‘The Data Management Maturity Assessment Review 2019’. You can download it for free HERE.
In the following articles of the series, I will provide an in-depth overview of how to measure maturity per each data management sub-capability and demonstrate the benchmark results.
For more insights, visit the Data Crossroads Academy site:
