By Jason Foster and Barry Green
At any stage of your data journey it is essential to take a step back and ensure you have a solid approach to data and AI in your organisation. Crafting your data strategy effectively is mission-critical. This article outlines six essential pillars that bring together all the components required for an effective data strategy.
As AI tools advance and data sources grow, so too does the importance of a strategic approach to maximising these assets. In fact, a recent Cynozure report found that when asked their top three priorities for 2025, 58% of respondents chose data and AI strategy.
A data strategy enables you to deliver business value through the application of data and analytics. Defining that strategy helps you articulate what you plan to achieve and how you will get there through data and AI applications – but where do you start?
The following six pillars bring together all the components required when writing your data strategy. Miss any one of these in your thinking, planning or execution and you will struggle to reach your full potential, limit opportunity and slow your progress.
Vision and value
Your vision should describe the important role data plays in achieving success for your financial organisation and the specific value you hope to achieve by making data-guided decisions. Successful data strategies are purposeful in their attempt to deliver business outcomes.
Aligning to your wider business strategy, this pillar is where you identify the key pain points and opportunities that exist in your institution and through which data can be applied to improve the decision making and create better solutions.
This is the place to assess the impact of AI on your organisation and the broader industry. Work out which business problems it is and isn’t suitable to support. Highlight the areas for R&D and the areas for fast execution.
For example, a retail bank might want to offer more personalised financial products for greater engagement and revenue. By creating processes that implement AI-driven data insights, the bank can envision the outcome and value that this shall bring.
You can align the rest of the data strategy pillars to this vision and use this pillar as a way to generate buy-in for your strategy.
People and culture
Making decisions guided by data is about your people and organisation’s culture. This pillar looks at the skills needed to implement your strategy successfully and how best to organise them. This needs to cover data, technical, AI, commercial, operational and management skills. It includes defining the roles and responsibilities of teams, the individuals within those teams and the training needed to improve their data fluency, knowledge and capability.
For example, a global investment firm could introduce a data literacy program that ensures all team members understand how to interpret and use data gathered effectively, with a support team on hand for any queries.
This needs to be backed up by a culture of blending intuition, experience and insights, so you should define what you can do to start changing the business culture. This starts with people – what they do, how they behave, their skills, the collaboration opportunities and sharing of knowledge and projects.
Operating model
The approach used for defining and managing with pace and agility can make or break your ability to deliver maximum returns. This pillar emphasises the importance of how your teams collaborate to build data and AI products and deliver business outcomes.
This pillar should articulate which method you will use to prioritise the business outcomes to invest energy into; how you allocate that work to the teams you have set up; and the approach to building your data products and services to test ideas and scale the successful ones through to a live environment. It looks at how you involve leadership members to keep them up-to-date with the data strategy status, as well as how to measure and monitor progress at a macro strategy level and a micro projects level.
It should consider how you govern AI at a corporate level to ensure you are building it responsibly, inside legal and compliance frameworks and your own AI usage policies. For example, a central data governance committee can be created that oversees AI applications, ensuring processes are aligned with regulations.
Technology and architecture
With a variety of innovative technologies available, your ability to embed and adapt your technology platform as part of your strategy is a huge differentiator and one that can get insight to the right people and help them create a stronger business.
For example, a bank could look to adopt a cloud-based data pool that enables real-time analytics for areas such as loan underwriting and risk assessments.
This pillar looks at what technology you need in order to obtain, manage and use data and AI effectively. It should define the technology strategy and how they all integrate with your systems.
Data management
The key outcome of data management is to have trusted, secure, quality and well-managed data that is ready to be harnessed by humans and AI to make decisions. Without this, organisations face the challenge of relying on low-quality, uncontrolled and untraceable data, which stifles progression and blocks the rest of your strategy.
This pillar identifies the work required to improve the level of trust in your data, starting with the definitions of your key metrics, how they are calculated and who owns them. It should look at what data you have, how it’s captured, how it moves through applications and who owns each data set. It should look at how you secure data physically and through strong access controls. Also, it looks after the quality of your data and in particular how master data sets (like customer, products, assets etc.) are created and managed.
You will also need to ensure that regulatory controls are met, documented and communicated. For example, an insurance company can install a decentralised database that securely stores customers’ data, with access given to separate teams, reducing errors and following data compliance.
Roadmap
Your strategy is not complete without a clear picture of the stages you will go through to deliver value and build the necessary capabilities. You need an adaptable plan that allows you to communicate the journey and improve at pace.
This pillar is where you unify the other five pillars into a roadmap for delivering your data products, and the order you should build out the capabilities required to help you meet your objectives. Importantly, it will outline the culture change initiatives required to create a data-guided and AI-positive culture.
As this is strategy and planning, there is no need to answer every single question or go into huge detail, but don’t be tempted to stay in strategy forever. This is a tool to make decisions on what you need to do and in what order, so make sure you are putting it into action and iterating along the way.
About the Authors
Jason Foster is one of the authors of Data Means Business. He is a best-in-class data strategist who has helped hundreds of organisations to activate their data and technology to drive change. He is host of the ‘Hub and Spoken’ podcast, Founder and CEO at Cynozure and has been recognised as one of the most influential people in data.
Barry Green is one of the authors of Data Means Business. He is a future thinking transformation leader who is passionate about using data, AI and technology to drive digital efficiencies in both our business and personal lives. He has undertaken a number of CDO roles with a strong belief that resilience, cognitive diversity and people are the keys to successful change and transformation.
