Voices | How Data Governance is evolving

Arun Chandar, VP Strategic Service Development explores how data governance is matching pace with the rest of the industry, adapting and evolving with change. Learn how good governance allows you to align with business goals, manage AI risks, and leverage metadata.

As technologies advance and data ecosystems become increasingly complex, data governance must adapt and evolve too. I want to share how effective governance impacts decision-making and operational efficiency, and how learning and adopting proactive strategies can help manage risks associated with emerging technologies like AI and ML. This insight is invaluable for anyone aiming to align their governance efforts with evolving business goals, ensuring your organisation remains adaptable and forward-thinking.

Stronger focus on business alignment and business agility

Organisations must focus their data governance efforts on the most value adding and highest priority areas. As business needs evolve with technological advances and shifting market trends, it is crucial to maintain a flexible and agile data governance strategy. This approach should align closely with overarching business goals, ensuring that governance efforts are not merely procedural.

Data governance has moved away from traditional models to a more dynamic and responsive framework. By aligning governance activities with business objectives, you can maintain agility as needs change. A proactive mindset is essential, with data governance activities triggered by significant organisational shifts, such as regulatory changes or market fluctuations.

Amplifi Voices Data Mesh Blog Image 1

It's also important to demonstrate how data governance contributes to your organisation’s performance. By initially aligning data governance with business objectives, its impact and value become clearer, providing transparency, and enhancing the effectiveness of your process.

Greater appreciation for norms and nuances

While general principles and best practices in data governance are foundational, the real skill lies in adeptly adapting these to your organisation's unique context. A one-size-fits-all approach simply doesn't work in data governance, as it often overlooks the distinct characteristics that set each company apart.

It's crucial to recognise the nuances within your industry or organisation that might define your competitive edge. Adapting data governance to align with your organisational culture and competitive strategy is essential. Sometimes, this involves rethinking terminology, like using 'data area' instead of 'data domain', or 'data ambassador' instead of 'data owner'. Such adjustments can significantly influence how data governance is perceived and embraced within your organisation.

Additionally, implementing a rigid framework across all areas can lead to pushback. It's vital to understand how different teams handle data and to customise your data governance strategy to fit these varied practices. This tailored approach not only eases implementation but also improves how quickly and easily data governance will be adopted across your organisation.

Stronger emphasis on culture – especially related to federated governance

It's becoming increasingly common for organisations to adopt a federated approach to data governance. This is where the responsibility for data is distributed across various business units rather than being centralised. This shift necessitates a strong data culture throughout the organisation, where everyone understands the importance of data, has a good level of data fluency, and knows their role in managing and using data.

As this data culture strengthens, it facilitates data democratisation—making data accessible across different levels of the organisation. This approach not only unlocks vast possibilities but also highlights the importance of equipping individuals with the necessary tools, training, and techniques to use data effectively.

Data governance initiatives are increasingly focused on enabling self-service analytics, enhancing data literacy, and promoting an environment conducive to data-driven decision-making. By embedding data deeply into daily business activities, organisations can maximise the benefits of federated governance and ensure its effective integration across the business.

Metadata management

Metadata management is becoming increasingly vital across all data disciplines, including governance, due to the vast amounts of data that organisations need to manage, understand, and derive value from. Effective metadata management practices are critical for comprehending what data exists, its location, its meaning, its usage, and how to extract value from it.

In the context of emerging concepts like data fabric and data mesh, metadata management is even more critical. Governing these frameworks requires access to accurate and accessible metadata to make sense of data and develop data products effectively. Additionally, the push towards data democratisation underscores the importance of metadata management. For people to confidently interact with data and make informed decisions, a robust level of metadata management is necessary.

Metadata management encompasses techniques such as data lineage, which tracks the journey of data from its source to consumption. This is vital for managing data quality and compliance and for monitoring issues related to data processes, all of which support a comprehensive data governance framework. Automating some traditional, manual data governance processes through metadata management tools can also enhance efficiency. These tools can automatically classify data, generate policies, and flag anomalies, streamlining governance operations and reducing resource demands.

Data quality

Data quality remains an essential aspect of data governance, though our understanding and methods for managing it are evolving. During a recent customer event, we explored new concepts like data observability and AI-driven data quality techniques. These discussions, led by Lee Morgan, Data Quality & Migration Practice Lead at Amplifi and myself, are shaping our approaches to integrating data quality within our governance frameworks. While the fundamental role of data quality has not changed, the tools and strategies employed are rapidly advancing.

We are moving towards a proactive approach to data quality, primarily through data observability, which alerts stakeholders to deviations in data patterns that could indicate errors or anomalies. This method, embedded within a data governance framework, allows for quicker and more effective responses.

The advent of AI-driven techniques such as inference engines, automated data classification, and interactive AI tools for report analysis is transforming traditional Data Quality processes. These innovations bring the risk of over-automation, where human oversight is still crucial. Effective governance is necessary to balance automation with human judgement, ensuring that enhancements in Data Quality do not lead to unethical or inaccurate outcomes.

Risk and trust

The integration of AI and machine learning into business processes highlights critical issues of risk, trust, and ethics in data governance. AI practices, especially those involving the use of large language models (LLMs), brings both challenges and opportunities, requiring a balance between innovative potential and stringent governance.This shift necessitates new roles to drive policies, govern ethical frameworks, mitigate AI bias, and audit algorithms, which are becoming essential parts of data governance.

It's also important to acknowledge that issues such as privacy and regulatory compliance are becoming increasingly complex, often developing more slowly than the proactive policies organisations can implement themselves. This highlights the importance of proactive data governance. Additionally, the advent of modern technologies like data fabric and AI-driven data cataloguing introduces new governance challenges. It is crucial to maintain high ethical standards and transparency in data usage, necessitating a consistent framework that is integrated into broader data governance strategies.

New technologies, new architectures

Emerging trends in data management are reshaping the need for robust data governance across various domains. One key trend is data mesh - highlighted by its four main pillars, one of which is federated governance. This approach necessitates a well-governed operating model that supports self-service data management while maintaining essential central capabilities.

Another is data fabric, which also requires stringent data governance, particularly in managing automated metadata management, or 'active metadata management'. This ensures effective oversight across different systems.

With the shift towards cloud-based data management, traditional data governance practices must adapt to new environments, ensuring they remain relevant and effective.

The integration of 'Data Ops' and 'Dev Ops' emphasises the need for collaboration and automation, necessitating the incorporation of data governance to ensure well-managed data pipelines and effective data lifecycle management during the deployment of new applications. This approach ensures governance continuity and compliance across data processes.

Closing thoughts

Practising good data governance allows you to stay ahead of potential risks, particularly when integrating advanced technologies like AI and machine learning. If you're seeking to enhance your organisation's approach to data governance, (or if you're interested in exploring our Data Governance Framework), please get in touch with our data governance experts. You can also read our Amplifi top tips on better Data Governance in our guide below.

Download Guide: 5 tips for better Data Governance

Amplifi Better Data Governance Guide Mockup 1
Amplifi Arun Blog Image

About the author

Arun Chandar is VP, Strategic Service Development.