Conflict, social unrest, inflation, climate change, and much more, are leading the world towards a trend of de-globalisation on a scale not many anticipated as recently as 10 or 20 years ago.

The effect of this is that power and data are becoming fragmented.

The result is that businesses may well be aware of data fabrics, data meshes and modern data stacks and may feel inclined to rush towards these technology solutions, but unless they address the cultural obstacles within their organisations and embrace a socio-technical approach, their investments and efforts are likely to be in vain.

How did we get here? Our new ever-fragmented reality leads to uncertainty in technology, which is naturally reacting to what’s happening in the world.

As a result of the uncertainty and rapid changes in technology – which includes artificial intelligence (AI) and the decentralised blockchain and web 3.0 – the global skills shortage bites harder than ever before.

In addition to this, regulations have become more complicated with even more red tape around rules about where data may and may not be stored in our fragmented world.

The impact here, and most organisations would attest to the fact that they are likely flooded by large and complex datasets from various sources.

They have difficulty integrating and managing data from different systems precisely because the data may be stored in different formats, structures, and locations.

In addition to this, locating data from different systems in a large organisation can be challenging, while rapidly changing compliance regulations make it difficult to comply.

The only way to succeed, in 2024 and beyond, is to have a clear and comprehensive data management strategy.

Data management refers to the process of collecting, storing, organising and maintaining data to support analysis and decision-making. Integrating a decentralised data world means there has to be interoperability between platforms and applications.

But what does this mean for organisations that need a clear and comprehensive data management strategy? It means they need a fabric or mesh to help them govern and control data.

The more decentralised and fragmented the world gets, the more technology is trying to weave it back together.

Data fabrics, data meshes exist to ease the challenges of managing data and to make sense of a multipolar and decentralised world.

Fabric and meshes are two different approaches with the intention to ease the challenges of data management in a multi polar decentralised world.

Modern data stacks are a collection of tools that enable organisations to collect, process, store and analyse data. These form part of a mesh or fabric data management strategy.

Data fabric is a tech-centric architecture for data management that unifies and integrates data across multiple systems.

Data fabric uses a variety of approaches to create a unified data management system that allows organisations to access, process, and share data more efficiently.

Data mesh, on the other hand, is a decentralised data architecture where data is treated as a product and managed by dedicated data product owners.

This approach transfers the responsibility from the central data team to the business units that create and consume data.

To improve the odds of successfully building an effective data management strategy, working with a trusted and experienced data partner to help shift the organisation’s data culture is a crucial – and often missing – step.

The Data and Analytics Leadership Annual Executive Survey 2023 found that cultural factors are the biggest obstacle to delivering value from data investments.

Data fabrics, meshes and modern data stacks will continue to consolidate an increasingly decentralised world by making the management of data easier. However, to ensure control over security and governance, and to extract value from data that is trustworthy requires a tactical shift to what we call a socio-technical approach. In other words, any strategy must be made up of an investment in people, process and technology to be successful.

This is because data management involves more than the technical aspects of data storage, processing and analysis.

It also includes the social aspects of data governance, change management, data quality management, user upskilling and collaboration between different teams. Organisations that know how to use technology the best will have an edge over their competitors.

Organisations would do well to engage with data partners who embrace a socio-technical approach to data management if they’d like to improve their odds of deriving value from data and extracting insights that can help them make better business decisions.

The urgency of implementing enterprise data management strategies comes from an acute awareness that businesses need to put their data to work effectively to influence business decisions and efficiency, and ultimately help steer the direction of the business.

C-suites understand that this urgency is not a fad, that an effective enterprise data strategy is a fundamental component of their digital transformation journeys. 

However, unless the data strategy is driven from the top of the organisation, there will – at some point – become a discrepancy between the business and data strategy. This disconnect can result in far more serious consequences than merely frustration – business outcomes as envisioned by the C-suites run the risk of becoming unattainable.

The data strategy must start at the executive level to ensure it aligns with the business outcomes. A data strategy being driven at lower levels in the organisation, separate from the C-suite which is designing the business strategy, is a recipe for an unsustainable disconnect. 

The ideal state is where the lower levels in the organisation implement under the direction of the C-suite as this allows for the right mix of levers to be pulled for an effective implementation. The executives, being the drivers, will enable budgets to access new technologies, and – crucially – initiate effective change management. This is the only way to ensure buy-in across the organisation. 

Effective change management also needs a top-down approach. It’s imperative for leaders to identify people in the organisation that have a natural knack for using data and then groom them to become champions for change. This approach, a business decision, encourages a culture of knowledge sharing and becomes exponential.

Obstacles to adoption of enterprise data strategies

Budgetary, infrastructure and skill constraints are global phenomena. Africa is no different, but in many cases some of these obstacles are more pronounced. 

On this continent, infrastructure and data literacy continue to be the biggest hurdles. However, in both these instances there are reasons to be optimistic. Starlink, for example, being available in Nigeria and Mozambique, will no doubt enable far more adoption among enterprises to leverage their data more effectively.

Data literacy is also improving exponentially. Today we get requests from countries north of South Africa’s border for assistance in regression analysis and correlation coefficients. This didn’t happen five years ago. 

Geography and industry also matters on the continent. For example, across regions and countries, the further one ventures from major centres, the less reliable – or even available – connectivity becomes, and this makes implementing a data strategy in an agricultural business, for example, more challenging. However, advances in technology are closing these gaps and driving increased adoption. 

Corporate South Africa, despite the country’s energy challenges, has been more fortunate in that infrastructure has been more advanced and stable for a longer period of time. This has accelerated the widespread acceptance by enterprises that they absolutely have to invest in their data strategies to remain competitive and make accurate decisions. 

Key components of an enterprise data strategy

Security, governance and enabling users are crucial components when implementing an enterprise data strategy. 

Security

Discussions about security are usually centred on criminality – such as breaches or data theft. Of course these are vital, and necessitate effective defense mechanisms and backup and recovery. However, security is multidimensional and data integrity is a crucial component. An organisation that has poor data quality is going to make incorrect assumptions, and this is a business and security risk.

In addition to making sure data is protected, backed up and quickly available and ensuring data integrity for effective decision-making, there also needs to be sufficient investment in the actual infrastructure to manage dynamic loads necessitated by peaks in user activity, for example. Failing to do this can result in data loss, which is a security risk.

People, or users, are another important cog in the multi-dimensional discussion about security. Ensure the right people have the appropriate level of access to the right data. Beyond this, user education is paramount.

Governance

Good data governance should enable users to leverage their data in the most efficient way possible, while still ensuring integrity, security and appropriate accessibility. There needs to be a healthy balance between access control and ease of access and use. Beyond this, the right skill sets need access to varying degrees of actionable data insights to enable reports that directly influence business decisions. 

User adoption 

The key is effectiveness and ease of use. An enterprise with thousands of users needs a platform that is as intuitive and uncomplicated as possible.

There is little use in one department pulling data from a spreadsheet, and another using some or other BI interface. The data champions identified in the change management have a far easier time helping their colleagues when using a unified, intuitive platform. 

Different levels of the organisation need different levels of insights. A platform like Qlik holds all data in its memory and enables this easy-to-use, single version of the truth across the organisation. 

An executive may want a bird’s eye view of what’s happening, while another may request the “why” to drive important decisions. An intuitive platform like Qlik enables another user – with access to the same unified tool – to dig deeper and then easily drag and drop charts to explain and narrate a story about the business. The same principle holds true when leveraging the predictive power of Qlik – extracting insights for the business must be intuitive and simple.

About the writer:

Lee Wearne is a Senior Business Intelligence Consultant at Insight Consulting. He has 12 years of business intelligence related experience across multiple industries including medical, agriculture, finance, manufacturing, HR and commercial sales. A key part of Lee’s experience has been within the agriculture and manufacturing industry across several African countries at an enterprise level, namely Mozambique, Zambia, Tanzania, Malawi, Eswatini and South Africa.