Building a data-driven culture: the complete UK guide for 2025

Here's a statistic that should concern every UK business leader: 70% of UK businesses are unable to effectively leverage data for strategic decision-making. Despite the UK data analytics market reaching £4.67 billion in 2024, most organisations are failing to translate their data investments into actual business value.

The problem isn't technology. Most UK businesses now have access to powerful analytics tools. The problem is culture. As David Waller, partner at Oliver Wyman Labs, noted in Harvard Business Review: "The biggest obstacles to creating data-based businesses aren't technical; they're cultural."

This guide examines what it really takes to build a data-driven culture in a UK organisation. We'll look at the current state of data culture across UK businesses, the frameworks that work, the barriers you'll face, and practical steps to transform how your organisation thinks about and uses data.

The UK data culture crisis: the numbers don't lie

Before we discuss solutions, let's understand the scale of the challenge facing UK organisations. The gap between data collection and data-driven action is enormous.

That means more than three-quarters of UK businesses collect data, but only one in five do anything meaningful with it. The rest are essentially hoarding information that never influences a single decision.

The situation becomes even more concerning when we look at advanced data usage. According to the UK Business Data Survey 2024, only 2% of UK businesses use data for artificial intelligence or automated decision-making. Even among large businesses, this figure rises to just 12%.

The decision-making gap

The consequences of this data culture deficit show up in how organisations make decisions:

• 70% of UK businesses are unable to leverage data for strategic decision-making
• Only 22.4% of organisations rate themselves as "very good" at leveraging data insights
• 72% report that data management challenges make decision-making slow
• 53.6% say these challenges make decision-making less strategic

The trust problem

Perhaps most telling is that business leaders don't trust their own data:

• Only 47% of UK analytics and IT leaders are completely confident in their data accuracy
• Only 40% of UK business leaders completely trust their data
• 77% of organisations experience data quality issues
• 91% report these quality issues negatively impact performance

When leaders don't trust the data, they fall back on intuition. KPMG research found that 67% of CEOs prefer to make decisions based on their own intuition and experience rather than insights from data analytics. This creates a vicious cycle: data isn't trusted, so it isn't used, so investment in data quality isn't prioritised, so data remains untrustworthy.

What is a data-driven culture?

A data-driven culture is more than having access to dashboards or running occasional reports. It's an environment where data-informed decision-making is the norm at every level of the organisation, not just among analysts or executives.

As Deloitte defines it, data culture means establishing "a mindset where making data-driven decisions is the norm, empowering every member of an organisation." This isn't about turning everyone into a data scientist. It's about ensuring that when decisions need to be made, people naturally ask: "What does the data tell us?"

A genuine data-driven culture has several characteristics:

• Data accessibility: Relevant data is available to those who need it, when they need it, without requiring specialist skills to access
• Data literacy: Employees across all functions can interpret data, understand its limitations, and apply insights to their work
• Trust in data: Data quality is maintained, and people have confidence that the information they're using is accurate and current
• Leadership example: Senior leaders visibly use data in their decision-making and expect the same from their teams
• Continuous learning: The organisation treats data as a source of learning and improvement, not just reporting

Why technology alone isn't enough

Many organisations make the mistake of thinking that implementing a business intelligence tool will automatically create a data-driven culture. It won't. Tools like Microsoft Power BI are powerful enablers, but without the cultural foundation, they become expensive reporting systems that few people use.

Research consistently shows that technology adoption without cultural change fails to deliver value. Deloitte's 2025 survey found that most organisations achieve satisfactory return on investment from AI initiatives only within two to four years, far longer than the seven to 12 months typically expected for technology investments. The difference between organisations that realise value quickly and those that struggle? It's not the technology. It's how well they've prepared their people and processes.

The four pillars of data culture

Building a sustainable data-driven culture requires attention to four interconnected pillars: Learning, Governance, Technology, and Innovation. Each pillar supports the others, and weakness in any area undermines the whole structure.

1. Learning

Learning focuses on empowering everyone in the organisation with the knowledge and confidence to incorporate data into their daily work. This doesn't mean everyone needs the same skills. Different roles require different levels of data capability.

The largest group in most organisations are "data citizens" - people who use data through reports and dashboards served to them but don't need to build models or write code. They need training on what analysis is available, how to navigate analytics tools, and how to interpret the visualisations they see.

Data analysts and specialists need different support, including user groups that cross departmental boundaries and more detailed training on producing analysis for wider consumption. The key is matching learning pathways to actual job requirements rather than providing generic training that doesn't connect to daily work.

Deloitte recommends the 70:20:10 model for developing data literacy programmes:

• 70% of learning from on-the-job experience
• 20% from formal training programmes
• 10% from exposure to external events like conferences and hackathons

2. Governance

Governance is about building trust. It means enabling users to access the right data, not restricting access through arbitrary security protocols. But to do this effectively, you need to understand your data estate and, most importantly, who owns data within the organisation.

Clear data ownership ensures data is correct at its source rather than through complex workarounds at the platform level. When people know who to contact about data quality issues and those owners are accountable for maintaining accuracy, trust in data improves across the organisation.

Governance process design should provide guidelines that empower users to seek access to data. Too often, governance becomes a barrier that slows down analysis and frustrates users. The goal is to make it easy to do the right thing with data.

3. Technology

Technology refers to the tools that enable interaction with data, including its storage, processing, and presentation. The specific technology chosen matters less than how well it's rolled out, adopted, and supported within the organisation.

Users must be empowered and set up for success by understanding what tools should be used for, how to use them effectively, and how to request support. "Data clinics" - regular open sessions where people can get help with data questions - are useful for creating rapport between technical and business teams.

For most UK organisations, Microsoft Power BI has become the standard business intelligence platform. Its integration with Microsoft 365, self-service capabilities, and relatively accessible learning curve make it well-suited to democratising data access. But even the best tool requires ongoing support and training to deliver value.

The most effective approach is "governed self-service", where IT owns the data modelling (ensuring standards and security) whilst business users own report generation and analysis. Users build reports on an IT-created data model, with clear processes for requesting new data attributes. This balance prevents the chaos of ungoverned self-service whilst avoiding the bottleneck of IT-controlled everything.

4. Innovation

Innovation may not be achievable at the start of a data culture journey, but it becomes possible as the other three pillars mature. As people become more comfortable with data, they naturally start identifying new opportunities for data-driven improvement.

Organisations can foster innovation through dedicated research and development initiatives, hackathons where teams explore new ideas and technologies, and a culture that celebrates experimentation even when experiments don't succeed. The members of the organisation are the incubator for new ideas that will drive the organisation forward.

UK success stories: organisations getting it right

While the UK statistics paint a concerning picture overall, several leading organisations demonstrate what's possible when data culture transformation is done well. These case studies show quantified business outcomes that prove the investment is worthwhile.

Lloyds Banking Group: machine learning at scale

Lloyds Banking Group, the UK's largest digital bank, invested £3 billion in people, technology and data as part of its commitment to helping Britain prosper. The bank recognised that its on-premise machine learning platform limited the number of data scientists who could effectively experiment with advanced models, constraining innovation at a critical time.

Between May 2023 and April 2025, Lloyds migrated 15 modelling systems comprising hundreds of individual machine learning models to Google Cloud's Vertex AI platform. The transformation yielded dramatic results:

• User base expanded from a limited cohort to over 300 data scientists and AI developers working simultaneously
• 80 new machine learning experiments initiated within six months of deployment
• Mortgage income verification reduced from multiple days to seconds
• Zero unplanned platform downtime since migration
• 18+ generative AI systems deployed to production
• 27 tonnes of CO2 emissions reduction through cloud migration

Ranil Boteju, Group Chief Data and Analytics Officer at Lloyds Banking Group, captured the strategic vision: "Data will play a critical part. It's a driving force behind our ambition to help Britain prosper. That's why we will continue to invest in a data-driven culture - one that harnesses the power of data and analytics to create better products and experiences for our colleagues and customers."

Sainsbury's: democratising data for 12,000 colleagues

Sainsbury's, with 1,400 stores, £28 billion in annual sales, and 189,000 colleagues, undertook a comprehensive three-year analytics transformation that represents the most systematic approach to data democratisation documented in UK retail.

The retailer deliberately consolidated its analytics technology to just three platforms: Snowflake for data warehouse infrastructure, Amazon Web Services for cloud platform services, and MicroStrategy for analytics delivery. This architectural simplicity enabled rapid scaling of analytics access across the organisation.

The flagship store-level analytics application, deployed across all 1,400 stores, provides real-time visibility into sales, waste, shrink, and margin performance. Store managers access insights directly from handheld mobile devices on the shop floor, making data-informed decisions part of daily operations rather than a back-office activity.

During COVID-19, Sainsbury's analytics teams deployed new applications within just one week that enabled rapid reallocation of resources to address staffing and supply chain shortages. The infrastructure enabled geospatial analysis to minimise colleague travel, store hour optimisation, and priority-based allocation of essential supplies.

Tesco: AI-powered personalisation at massive scale

Tesco, with 23 million Clubcard loyalty programme members and 30 years of Clubcard history, deployed artificial intelligence-driven personalisation that demonstrates what's possible when data culture and technology combine effectively.

In May 2024, Tesco launched Clubcard Challenges, using EagleAI's machine learning to assign individually personalised purchasing challenges to members. The AI executes over 190 intelligent decisions per individual to assign each challenge, analysing purchase history, category engagement, spending patterns, and reward thresholds.

The results across four campaigns in 2024 were remarkable:

• 10 million customers targeted per campaign
• 289 million personalised vouchers delivered to 7.6 million customers
• 76% conversion rate from page visitors to active participants
• 98% open rate on app notifications
• Over half a billion extra Clubcard points collected by customers

CEO Ken Murphy noted: "Our customers continue to benefit from increasing personalisation and great value with Clubcard." The programme contributed to Tesco's "biggest ever Christmas performance" in 2024.

John Lewis Partnership: analytics for 74,000 employee-owners

John Lewis Partnership, the UK's largest employee-owned business with over 74,000 employee-owners across John Lewis department stores and Waitrose food shops, demonstrates how complex, multi-channel retail operations can successfully democratise data access.

The organisation centralised customer data through Salesforce implementation, then used Tableau to train 8,000 users in the first implementation phase. This enabled 10,000 Partners to independently explore real-time sales and inventory data without depending on specialist analytics teams.

NHS Wrightington, Wigan and Leigh Trust: data improving patient care

The healthcare sector shows how data culture directly impacts patient outcomes. Wrightington, Wigan and Leigh NHS Foundation Trust, serving 320,000 people in Greater Manchester with 4,800 professionals, migrated from specialist analytics to a self-service model.

The Trust's philosophy shift was fundamental: moving from 80% of business intelligence team time spent developing requested applications to 80% spent exploring data and identifying new insights. This reversal enabled data specialists to focus on value-added analysis rather than report generation.

The measurable outcomes directly affected patient care:

• 30-minute reduction in A&E waiting times
• Propelled the Trust into the top 10 nationally on access metrics
• Better workforce scheduling through patient volume prediction
• Optimised operating theatre utilisation

Across the NHS more broadly, documented efficiency savings demonstrate the sector-wide impact. The Humber and North Yorkshire health system documented £4.1 million in efficiency savings through shared care record implementation, while South Yorkshire's integrated care system projects cumulative efficiency savings of £22 million over five years.

Building data literacy across your organisation

Data literacy is the ability and mindset to read, work with, analyse, and use data ethically to solve challenges, drive innovation, and create value. It's not just a technical skill. It includes critical thinking about data, understanding its limitations, and knowing when and how to apply insights.

The UK data literacy crisis

The scale of the UK's data literacy challenge is stark. According to FutureDotNow research:

• 52% of working-age adults cannot complete all 20 basic digital tasks that industry and government agree are essential for work
• One in two young people cannot perform all essential digital tasks
• One in two academics holding degrees or PhDs lack complete digital skills
• One in three people earning over £75,000 annually lack complete digital skills
• One in five people working in the technology sector lack all essential digital skills

Within organisations, the picture is equally concerning. 58% of data leaders report that most of their employees are not data literate. And half of UK workers had not received any data skills training within the previous two years, despite both high employer demand for data skills and expressed employee interest in development.

Building a data literacy programme

The Prosci ADKAR model provides a useful framework for thinking about what individuals need to change their behaviour. ADKAR identifies five elements that must be present for change to succeed: Awareness (understanding why change is needed), Desire (genuine motivation to participate), Knowledge (knowing how to change), Ability (practical skills to implement new behaviours), and Reinforcement (mechanisms to sustain change over time).

Many data literacy programmes fail because they focus exclusively on the Knowledge element through training, whilst neglecting the other four. Employees complete training but lack genuine desire to change their working practices, or they have the knowledge but lack the ability to apply it in their daily work, or there's no reinforcement mechanism to sustain new behaviours once initial enthusiasm fades.

Deloitte's experience across numerous client projects has identified seven key components for building an effective data literacy programme:

1. Find a sponsor: Programmes with strong executive sponsorship are able to catalyse new behaviours and amplify change. Find a passionate and influential sponsor to drive your data literacy initiatives from the front.
2. Define a clear vision: Understand what cultural behaviours you want to drive and the corresponding skills which underpin them. You can't create a data-literate organisation unless you understand what skills your various personas need.
3. Understand the gap: Assess how data-literate your workforce currently is. A basic data literacy evaluation will help you understand how much time and effort is required to move between proficiency levels.
4. Map curriculum to learner personas: Ensure you understand your personas, such as frontline staff versus back office versus executives, so you can provide content and training which is pertinent and delivered in a way that will have impact.
5. Focus first on foundational data skills: While terms like generative AI and neural networks sound exciting, they are at the far end of the spectrum. Most employees need to understand basic data analysis principles before tackling advanced topics.
6. Seek out opportunities to apply new skills: Research shows 80% of training content is forgotten within a month unless the person has an opportunity to apply their new skills. As a leader, make space for employees to safely test out their new capabilities.
7. Celebrate successes: While your organisation is on their journey to becoming data-literate, there will be moments where it gets difficult. One way to encourage people to stick with it is to provide them with a "finish line" that celebrates their achievements, such as graduation ceremonies.

UK government skills initiatives

UK organisations have access to several government-backed initiatives for developing data skills. The Get Tech Certified programme provides all civil and public servants with access to over 200 free learning and certification pathways, covering data, cloud technologies, artificial intelligence, and more. These range from introductory courses through to advanced qualifications, delivered by providers including AWS, Google Cloud, IBM, and Microsoft.

Skills Bootcamps offer 16-week intensive training programmes for those aged 19 and over, with guaranteed job interviews upon completion. These bootcamps focus on practical competencies including data engineering, software development, and digital marketing. For longer-term capability building, apprenticeships combine work-based learning with formal instruction, with recent statistics showing a 90.8% first-attempt pass rate across apprenticeship end-point assessments.

The data champions model

One effective approach is to identify data champions within each department. These are people who have a natural affinity for data and analytics and can help their colleagues troubleshoot issues, develop skills, and build enthusiasm for data-driven approaches. Critically, champions should be identified and invited based on genuine enthusiasm rather than assigned through top-down mechanisms. Programmes that appoint champions without genuine commitment often exist on paper only.

When identifying potential champions, look for three types of people: "go-to colleagues" whom others naturally approach with questions, "domain masters" who possess deep knowledge of why data matters to their specific business area, and "enthusiasts" who view data and new technologies as opportunities rather than threats.

Data champions serve as the bridge between central analytics teams and the wider business. They understand both the technical possibilities and the practical realities of their department's work. They can translate between data speak and business speak, making analytics relevant and actionable for their colleagues.

Supporting champions requires genuine investment. They need early access to products and training so they can become experts capable of helping colleagues. They benefit from formal communities where they can share challenges and solutions with other champions. And they need visible recognition from leadership and integration of their champion activities into formal performance reviews, treating championship as a genuine career pathway rather than an extracurricular activity.

Leadership's role in data culture transformation

Data culture transformation cannot succeed without genuine leadership commitment. When asked about obstacles to achieving better data literacy, data leaders most frequently cite lack of leadership buy-in (23%), ahead of lack of time and resources (22%), deep-rooted cultural resistance (19%), and poor access to appropriate training (16%).

The leadership trust gap

Perhaps the most significant barrier is that many senior leaders don't trust or feel comfortable with data-driven approaches:

• 67% of executives report feeling uncomfortable accessing or using data from advanced analytics systems
• Even within companies with strong data-driven cultures, 37% of respondents express discomfort with analytics
• 67% of CEOs indicate they prefer to make decisions based on intuition and experience over data analytics

This creates an obvious problem. If leaders visibly prefer gut instinct to data, their teams will follow suit. Why invest time in data analysis if the boss is going to make decisions based on intuition anyway?

What effective data leadership looks like

Leaders who successfully drive data culture transformation do several things consistently:

• Model data-first behaviour: They visibly reference data in meetings, ask for evidence to support proposals, and share their own decision-making process
• Allocate resources: They ensure time and budget for data literacy training, tool implementation, and data quality improvement
• Remove barriers: They actively work to eliminate obstacles to data access and use, rather than adding bureaucratic controls
• Celebrate data-driven wins: They publicly recognise when data-informed decisions lead to positive outcomes
• Accept uncertainty: They understand that data provides insight, not certainty, and create psychological safety for teams to make recommendations based on imperfect information

Overcoming common barriers to data culture

Building a data-driven culture means confronting several persistent barriers. Understanding these challenges helps you address them proactively rather than being surprised when they emerge.

Data quality and trust issues

Poor data quality is the most frequently cited barrier to effective data use. When asked about the primary impediment to better data management, 59.9% of UK respondents identified inadequate data quality. Beyond quality issues, 44.7% cite lack of trust in data integrity, and the same percentage report that analytics departments fail to keep pace with business initiatives.

Practical problems compound these issues. Having multiple copies of datasets makes it impossible to identify data ownership (43.7%), and administrative friction when locating original datasets (31.8%) slows down analysis and introduces errors.

The solution: Start with data governance. Establish clear ownership for key datasets, implement quality monitoring, and create processes for reporting and resolving data issues. Don't wait for perfect data before starting your data culture journey - use Power BI dashboards to make quality issues visible, which often accelerates their resolution.

Resistance to change

Risk-averse mindsets and resistance to change from both workers and managers rank among the most frequently cited barriers to technology adoption. But resistance is often misunderstood. What leaders interpret as stubborn unwillingness to change frequently reflects legitimate concerns: fear of the unknown, uncertainty about how changes will affect roles, doubt about the capacity to adapt.

Deeper psychological factors also drive resistance. McKinsey research found that only 28% of executives believe their organisations make generally good strategic decisions. The root cause? Managers struggling with cognitive biases including confirmation bias (seeking data that confirms pre-existing beliefs) and overconfidence. Better data alone won't solve this if decision-making processes remain vulnerable to these biases. The quality of the decision process matters six times more than the quality of analysis.

People also hold deeply ingrained mental models about data that operate at implicit levels: "Data analysis is for specialists, not relevant to my work," "My intuition is more reliable than data," "Data is used for surveillance and control." These assumptions often go unexamined because people don't recognise them as assumptions rather than facts.

The solution: Frame data as a tool that enhances human judgement rather than replacing it. Involve sceptical team members early in initiatives, asking for their input on what data would be useful. When data confirms their intuition, celebrate it. When it contradicts expectations, use it as a learning opportunity rather than a criticism. Most importantly, address the process, not just the tools. Create structured decision-making processes that require actively seeking disconfirming evidence and accepting findings that conflict with prior beliefs.

Skills and talent shortages

IT and data skills remain the hardest skills to find in the UK labour market, a position unchanged for the last five years. According to the 2025 Talent Shortage survey, 51% of IT firms plan to hire, but 75% struggle to find qualified candidates. Almost half of UK businesses (46%) reported struggling to recruit for roles requiring data skills over the previous two years.

The solution: Focus on developing existing employees rather than relying solely on external recruitment. The data champions model allows you to leverage people who already understand your business context. Partner with training providers for structured upskilling programmes, and consider apprenticeship routes for building long-term capability.

Competing priorities

Among Chief Data Officers surveyed, 47% reported that competing organisational priorities hinder their ability to realise the full value of data, while 48% cite budget and resource limitations as key challenges. Data transformation initiatives often lose out to more urgent operational demands.

The solution: Connect data initiatives to business outcomes that matter to leadership. Show how improved data capabilities will address their priorities, whether that's cost reduction, revenue growth, or risk management. Start with quick wins that demonstrate value, building momentum for larger investments.

Getting started: practical steps for UK organisations

Building a data-driven culture is a multi-year journey, not a project with a defined end date. Here's how to get started.

1. Assess your current state

Before you can improve, you need to understand where you are. Conduct a data maturity assessment covering:

• What data do you collect, and how is it stored and managed?
• Who owns key datasets, and how is quality maintained?
• What analytics tools are in use, and how widely adopted are they?
• What level of data literacy exists across different functions?
• How do leaders currently use data in decision-making?

2. Start with a business problem

Don't try to transform everything at once. Identify a specific business challenge where better data could make a measurable difference. Perhaps it's understanding why customers churn, optimising inventory levels, or improving sales forecasting. Focus your initial efforts on solving this problem well.

3. Invest in the right tools

For most UK organisations, Microsoft Power BI provides the right balance of capability and accessibility. Its self-service features allow business users to create their own reports without IT dependency, while its integration with Microsoft 365 means it works with tools people already use. Power BI Pro costs £10.80 per user per month, making it accessible for organisations of all sizes.

4. Build capabilities systematically

Develop a learning programme that addresses different needs across your organisation. Train data citizens on interpreting dashboards and reports. Develop data champions who can support their departments. Build specialist capabilities where you need them. Remember the 70:20:10 principle: most learning should happen through applying skills on the job.

5. Measure and communicate progress

Define how you'll measure the success of your data culture initiative. This might include:

• Data literacy assessment scores before and after training
• Adoption rates for analytics tools
• Number of data-informed decisions documented
• Time saved on reporting and analysis
• Business outcomes from data-driven initiatives

Communicate progress regularly to maintain momentum and demonstrate value to stakeholders.

The future of data culture: AI and beyond

The organisations building data-driven cultures today are preparing for a future where artificial intelligence amplifies human analytical capability. By 2026, 75% of customer-facing applications will embed analytics capabilities, and conversational AI will enable business users to ask questions in plain English and receive instant visualisations and recommendations.

Microsoft's Power BI Copilot already enables natural language queries, allowing users to ask "What were our top-selling products last quarter?" and receive AI-generated insights without writing a single line of code. This isn't science fiction. It's happening now, and the organisations with strong data cultures will be positioned to leverage these capabilities immediately.

The World Economic Forum's Future of Jobs Report 2025 projects 78 million net new jobs this decade, with the fastest-growing roles in data analytics, AI, and related fields. But beyond technical roles, employers increasingly value skills that complement AI: creative thinking, resilience, curiosity, and analytical mindset. Building a data-driven culture today prepares your workforce for this future.

The AI skills gap is widening. Current employee AI literacy sits at 28%, but organisations will need 60% literacy by 2026. Workers with AI skills already command a 56% wage premium. The organisations that invest in data literacy now will find the transition to AI-augmented analytics far smoother than those scrambling to catch up.

Building data culture is a journey, not a destination

The UK faces a genuine data culture crisis. Most businesses collect data but don't use it effectively. Most employees lack basic data literacy. Most leaders prefer intuition to evidence. The costs of this deficit are measured in billions of pounds of lost productivity and missed opportunity.

But the organisations that get this right see transformative results. Faster decisions. Better innovation. Stronger employee engagement. Real competitive advantage that's difficult for competitors to replicate because it's embedded in how the organisation thinks and operates.

The technology exists. Tools like Power BI make powerful analytics accessible to everyone. What's needed is the cultural change to make data a natural part of how your organisation works.

Ready to start building a data-driven culture in your organisation? Our Business Intelligence team at Red Eagle Tech helps UK organisations implement Power BI and develop the capabilities to use it effectively. Contact us today for a free consultation.

Frequently asked questions