From ambition to action: How a data strategy powers AI in the public sector

Artificial intelligence is rapidly becoming central to the digital ambitions of not just tech giants and global corporations but also public sector organisations. For the UK’s government departments, AI is increasingly viewed as a tool to drive efficiency, improve services and make smarter decisions, with the AI Opportunities Action Plan and State of digital government report highlighting its potential.

However, amid the rush to explore AI, one fundamental question is often overlooked: do we have the data strategy needed to support our AI ambitions?

How a data strategy supports AI-enabled transformation

Currently, 43% of public sector organisations do not have a data strategy – a comprehensive and long-term plan setting out how an organisation will collect, manage and unlock value from data. A thorough and considered data strategy acts as a roadmap, connecting organisational goals with the approaches, actions and activities required to achieve them, as well as the underlying data capabilities and cultural factors needed for success.

A data strategy sets organisations on the path to developing robust data foundations and cultivating a strong leadership that understands, values and champions the use of data across an organisation, both vital for accelerating AI adoption, unlocking real value and transforming public services.

A data strategy can:

• Provide a roadmap for AI. With organisations increasingly tasked with integrating AI into their operations, a data strategy ensures these efforts are grounded in real departmental objectives. It enables teams to focus on activities that are purposeful, coordinated and value-driven.
• Enable aligned autonomy. A strong data strategy provides a framework that allows teams to experiment and innovate – within clearly defined guardrails. With aligned autonomy, teams can reduce duplication, avoid anarchy and ensure that any innovation is safe and strategic.
• Promote transparency. Within the public sector, trust is everything and programmes are more effective when operated with clarity and accountability. An accessible, well-communicated data strategy ensures every stakeholder understands the direction of travel and their role in it, creating a cohesive data service.

What makes a good data strategy?

Developing a useful data strategy means thinking holistically about an organisation and its data. It’s not just about tech – a good strategy will also look at structure, people, skills, and governance. Key elements include:

• Cloud adoption alignment: Cloud adoption in the public sector has accelerated through dedicated budgets, but it remains inconsistent, with 73% saying they store more than half of their data on-premises. A good data strategy will align with the organisation’s cloud strategy and the limitations of legacy systems to support scalability and flexibility.
• Interoperability structures: Public services don’t exist in isolation, relying on the fast, effective exchange of data between departments to function effectively for citizens. Data strategies across the public sector need to support this seamless data exchange.
• Governance models: Data privacy, security and compliance are vital in any organisation but especially within the public sector due to the vast amounts of data departments can collect and manage about their citizens. Clear governance surrounding data ownership, storage and access protocols is crucial to maintaining data integrity and security.

Developing an effective data strategy

So, how do you go about writing a data strategy? It starts by asking the right questions, involving the right people and thinking holistically about the organisation, capabilities and goals. Key components include:

1. Business objectives: The first step is to define what you want to achieve and what you need to support those objectives, for example, the aims of any AI work and the data that is required for AI initiatives to be successful.
2. Technical architecture: Define the systems and tools that support data collection, storage, and analysis. This includes those already in use and what may need to be adopted, modified or migrated away from to support future ambitions.
3. Data quality and availability: An organisation needs to set out what data is required, the sources of this data and quality standards that need to be met as part of a strategy. Making data available, reliable and trustworthy are some of the biggest challenges modern organisations face and they need to be addressed.
4. Organisational structure and governance: A good data strategy maps out clear governance processes and the roles and responsibilities across departments within data management. This helps avoid the trap of data becoming everyone’s job but no one’s responsibility.
5. Skills and capabilities: New ideas often require new skills. A data strategy will also outline what capabilities exist in the organisation as well as plans to build teams with the necessary expertise to deliver on ambitious plans.
6. Testing environments: Innovation takes time and resources, as well as the freedom to try something new. Build safe spaces within the data strategy so teams feel confident to explore AI use cases, refine data initiatives and learn through iteration.
7. Measurement and KPIs: Creating a data strategy is one step, but how does an organisation know if it has been successful? Setting out key metrics and KPIs helps drive behaviour and enables organisations to gain valuable insights into how well it is using data assets to drive value. It can also help identify areas for improvement to drive success in the future.

Delivering real value with a data strategy

At Equal Experts, we specialise in working alongside public and private sector organisations to build a clear and effective data strategy, helping them turn their big ambitions into practical action and ultimately deliver real results.

Working with HMRC, we helped to develop the Customer Insight Platform (CIP), transforming vast, unstructured data into a scalable solution. With almost endless use cases for the platform, we helped HMRC develop a strategic roadmap for data and consider how a modern data platform could enhance fraud prevention, provide customer insights, and support smarter decision-making across UK government services. Read the HMRC CIP case study.

Similarly, we’ve also helped transform data across the John Lewis Partnership through a holistic data strategy that included building aligned autonomy and paved roads within teams. By empowering teams to self-serve within data, we helped the retailer unblock data expertise bottlenecks and use data more effectively and strategically across the organisation. Read the John Lewis Partnership data pipelines case study.

These examples show how a well-crafted strategy can unlock real, lasting impact.

Build a more effective data strategy: Start with a data health check

Although many public sector departments face similar challenges, no two organisations are the same. Each has its own unique mission, history and opportunities to provide better services to citizens. The first step in building a data strategy that empowers an organisation to use data more effectively is understanding its current data capabilities and what it aims to achieve.

Our data health check is a crucial tool in this process, providing an independent and bespoke assessment on which to build a data strategy. Government organisations gain actionable insights into their current situation and can confidently plan impactful next steps on the journey to greater data maturity.

While AI demands strong foundations and clean, reliable, interoperable data, its use in the public sector isn’t just a technical issue. It’s also strategic. As public sector AI adoption grows, a lack of strategy will see organisations risking wasted time and resources on initiatives that fail to deliver value.  A well-defined data strategy can act as a guiding roadmap, aligning efforts and supporting real action.

If your organisation wants to build a strong data strategy and unlock the real value of AI adoption, we can help. Contact us to arrange a data health check and take the first step towards better data use across your organisation.