Subscribe

Growing up

There are seven crucial steps to achieving data governance maturity.

Mervyn Mooi
By Mervyn Mooi, Director of Knowledge Integration Dynamics (KID) and represents the ICT services arm of the Thesele Group.
Johannesburg, 03 May 2013

Data governance (DG) maturity is still in its infancy stage in South African businesses. Although many organisations have actually developed industry-standard DG frameworks and have tried to deploy DG programmes, the implementation of data governance practices has largely been unsuccessful because many have failed to map DG frameworks back to their IT, project, data, information or content life cycles.

A huge challenge organisations face is that they are increasingly under pressure to comply with statutory and regulatory requirements, governance being the primary focus.

The problems auditors and the respective stakeholders have are:

1) They face multiple versions of the truth in the data, information, and content;

2) They cannot create consistent lineage or trail back to sources;

3) They cannot determine which data, information or content were used, when and under what circumstances; and

4) They cannot adequately figure out who took what decisions and the processing methods used.

Rest assured

Companies need to know, with certainty, who is responsible and accountable for their resources, being data, information and content, and be able to monitor and control the process life cycle from one end to the other, for each and every resource to achieve DG.

Many companies consider DG a necessary evil - at best there are a number of reasons why they will implement it. An organisation will benefit from data governance when it has a large portfolio of systems that cannot be properly managed using traditional management models, or when systems become so complicated that the same applies, or when cross-functional programmes have to be implemented, and, as is the case with auditors, when regulations, compliance or contractual obligations require it. With the perpetual increase in data, information or content volumes, the market for DG is growing.

In small, simple environments, deploying data governance is far easier than in large, complex environments. The demand for policies, procedures, standards and guidelines increases exponentially as size and complexity rises. Hundreds of users, potentially hundreds of databases, spreadsheets, silo- and cross-functional departments, divisions and businesses can quickly become an onerous burden on even the most dedicated teams. The resulting chaos filters through the processes that they support to appear at the coalface where customers meet the business, which rapidly impacts the top line.

Many companies consider DG a necessary evil.

The desire to deploy good DG programmes does exist in South African businesses - just take a look at the recruitment opportunities in the field and read about the businesses that have adopted data governance platforms. They range from banks to the top retailers and telcos. Job specifications require people fluent in policy institution, process management, reporting, expectation management, architecture and design, planning and familiarity with regulatory requirements, among others.

Getting the right people onboard following the will and commitment to deploy DG is very important, no doubt. But what do businesses do to achieve effective DG, for example, making it real and applicable? How do they orchestrate DG throughout the organisation?

Seven habits of highly effective data governance

1. As a first step, a data governance framework is essential to position DG as a key component or enabling service of the organisation.

2. The next step, which many have achieved or are currently undertaking, is to ensure they have relevant DG policies, procedures, standards and guidelines in place.

3. Entrench a data, information and content life cycle management programme that impacts all processes that create, store, process, consume and destroy data in the business. Key to making DG real and applicable is to map the life cycle programme phases to the organisation's implementation or project methodology, the software development life cycle, agile methods, change and release management processes, or whatever is specifically used in the business, for new systems implementations and system changes.

4. The life cycle management programme must articulate the policies, procedures, standards and guidelines as they are adopted through the compilation of applicable, real controls and checkpoints for the life cycles. Governance controls and checkpoints must be implemented at each gate or phase in the management life cycle programme and enforced. It is through these gates that the policies, procedures, standards and guidelines are articulated or instituted, and where they are made real and applicable.

5. Governance controls can be multifaceted. For example, they can be enforced from various aspects or competencies such as governance, architecture, data management, data quality and IT.

6. The life cycle controls must be extended into operations, the live environment, when new projects, system components, systems and changes to existing systems are implemented. The beauty is that most of the controls can be automated in the system processes so they do not place additional overhead on technical personnel.

This approach can be applied to both structured and unstructured information even though unstructured information may call on specialised technologies such as enterprise content management, for example.

With limited resources and budgets, few businesses will find they can approach the task from top to bottom in one fell swoop and leverage on existing, if any, DG initiatives. They'll need to figure out where to put their people to get the critical results, immediately followed by the necessary, and finally the important implementation tasks at hand. Doing so can be difficult, but having the right skills available is paramount.

Share