Fuelling intelligence: turning data into AI-driven value

by Susannah Matschke - Head of Data & AI- Growth, Sopra Steria Next UK
| minute read

In summary:

  • Strong data foundations like quality, accessibility, and readiness, are essential for unlocking meaningful ROI from AI investments.
  • AI success depends on understanding and managing data as a strategic asset, not just a byproduct of operations.
  • A structured data transformation journey helps organisations assess their current state, define goals, and build toward AI maturity.

Data was the big focus of the third webinar of our AI Maturity Series. Across every industry, leaders know data is a hugely valuable asset. But turning it into actionable AI-driven insight is where many get stuck.   

Data can be vast, scattered, of poor quality, or even completely unknown in some organisations - particularly those who’ve never taken the time to assess it. Without the right foundations, even the smartest AI solutions won’t deliver real value.   

The organisations pulling ahead are the ones investing in getting their data ready. 

During the webinar, Susannah Matschke - Head of AI Foundations, discussed how organisations can unlock the full value of their data by improving quality, accessibility, and readiness for AI. 

In this blog we have pulled together the highlights of what she covered in the session. To watch the full webinar, you can access the recording here

Fuelling Intelligence: Turning data into AI-driven value

Demystifying data and AI 

To set the scene and establish some shared understanding, Susannah started by clarifying some key definitions that are often thrown around in discussions about data and AI. She emphasised the importance of a shared understanding to foster inclusive conversations and keep barriers to entry low.  

What is data?  

Data is generated by human action, collected through things like user input, sensors, logging, and calculations. Data could be in the form of numbers, words or symbols but most importantly it is something that can be made useful through analysis, which is in turn interpreted to reveal patterns, trends, and relationships, ultimately leading to meaningful information or knowledge.  

What is information?  

Information is data that has been “made useful” or given a purpose, through analysis, transformation, collation or any other form of processing. Information is processed and organised, able to provide context and meaning, and enables decision making.  

What is AI?  

At its core, AI is a series of algorithms with roots in statistics and probability. AI algorithms analyse data to identify patterns and relationships, which can then be used to make predictions, classify new data, or generate new content.   

What is an algorithm?  

The definition of an algorithm is “a set of instructions to be followed in calculations or other operations.” This applies to both mathematics and computer science. So, at the essential level, an AI algorithm is the programming that tells the computer how to learn to operate on its own. AI algorithms often work by taking in training data that helps the algorithm to learn. How that data is acquired and is labelled marks the key difference between different types of AI algorithms. 

Driving value 

Susannah challenged the traditional view of data as a mere byproduct of business operations. Instead, she encouraged organisations to treat data as a valuable asset in its own right - something to be nurtured and managed with intention. 

Collecting huge quantities of data isn’t the goal, she pointed out. The aim should be to unlock value from that data. Organisations that do well with AI, according to Stanford’s AI Index, are those that treat their data with this mindset. 

Critically, Susannah warned against rushing into AI investments without addressing the state of your data. AI systems rely on high-quality data - bad data will only lead to worse outcomes when amplified by AI. If you don’t understand what data you have, what it’s telling you, and where the gaps are, and where your insights are, what are you hoping to get out of your AI implementation? AI isn’t a magic wand, it won’t fix your data, it will only amplify and increase your existing data issues (whilst costing you more money). 

If we invest in AI without first investing in our data, the ROI will simply never exist. 

This graphic presents a visual comparison between the scale of investment in AI and the actual returns being realised by organisations. The image features a balance scale, with investment figures on the left and ROI outcomes on the right.

On the left are the following investment figures:

  • 78% of companies have invested in AI usage
  • An estimated $33.9 billion was invested in Generative AI in 2024
  • Over $100 billion was invested in AI overall in 2024

On the right are the following ROI outcomes:

  • Only a fraction of organisations report productivity gains
  • Many report less than 10% cost savings or less than 5% revenue lift
  • BCG analysis found only 25% of companies are seeing measurable ROIs from AI investments

Data foundations for AI success   

Susannah acknowledged that despite massive investments in AI in recent years, most organisations struggle to turn data into measurable ROI. 

The problem, she said, isn’t the AI models themselves but the poor state of the underlying data and how it is managed, governed and understood. 

When data foundations are strong, organisations can: 

  • Better understand their business. 
  • Define strategic direction. 
  • Identify gaps. 
  • Modernise infrastructure. 
  • Build trust in their data. 
  • Shape a strong data culture. 
  • Secure long-term success. 

To deploy successful AI, organisations must redefine their data foundation as a strategic differentiator and a critical success factor. 

To bring this to life, she referenced Maslow’s Hierarchy of Needs. AI is the top of the pyramid - equivalent to self-actualisation - but without a solid base, which is food and shelter in Maslow’s hierarchy, and clean data in our hierarchy, you’ll never reach the top. 

 

This graphic presents two pyramids side by side, each illustrating a structured hierarchy. The left pyramid represents Maslow’s Hierarchy of Needs, a psychological framework for human motivation. The right pyramid represents the Data and AI Hierarchy of Needs, a model outlining the foundational and advanced requirements for successful data and AI implementation.

In Maslow’s Hierarchy of Needs Pyrmaid, progressing from foundational to aspirational human needs states the following:

  1. Physiological Needs
    Breathing, food, water, shelter, sleep
  2. Safety and Security
    Health, employment, family, social stability
  3. Love and Belonging
    Friendship, relationships, connection
  4. Self-Esteem
    Confidence, achievement, respect, uniqueness
  5. Self-Actualisation
    Morality, creativity, purpose

For the Data and AI Hierarchy of Needs pyramid, progressing from foundational data practices to advanced AI capabilities states the following: 

  1. Collect
    Data generation, sensors, user-generated content, logging
  2. Move and Store
    Data storage, pipelines, lifecycle management
  3. Explore and Transform
    Cleaning, standards, quality, preparation
  4. Aggregate and Label
    Analytics, metrics, monitoring, sharing, visualising
  5. Learn and Optimise
    Artificial Intelligence (AI), prediction, deep learning

Data transformation journey 

To help firms get started, Susannah introduced a structured view of the data transformation journey – an end-to-end approach that encapsulates all the essentials for using data well within an organisation. This framework encourages organisations to identify where they are, where they want to go, and what they’re missing. 

By using this model, teams can take stock of their current capabilities and map out a clear path to AI-readiness, rather than jumping into solutions without foundational clarity. 

This graphic outlines a structured journey for organisations aiming to transform their data capabilities. It is divided into four progressive stages—Envision, Enable, Elevate, and Evolve—each contributing to a series of outcomes that support long-term data maturity and value creation.

The envision stage: 

  • Mission.
  • Vision.
  • Data strategy and objectives.
  • Data discovery and gap analysis.
  • Defined outcomes and benefits.
  • Impact assessment.
  • Transformation roadmap.
  • Outcome one: business case.

The enable stage starts with the data maturity assessment and then breaks off into two branches. The first branch consists of:

  • Data proofing.
  • Common standards and models.
  • Processes and procedures.
  • Data engineering.

The second branch consists of:

  • Change strategy and plan.
  • Organisational structure.
  • Team and skills development.
  • Data literacy programme

The branch joins back together for the following:

  • Data governance and policies.
  • Outcome two: target operating model achieved.

The elevate stage:

  • Data management, quality and cleansing.
  • Data visualisation and business intelligence.
  • Outcome three: data enrichment.

The evolve stage:

  • Artificial Intelligence and predictive analytics
  • Insight-driven decision making
  • Benefits and value measurement
  • Data maturity progression report / comparison
  • Ongoing data harnessing


Checklist for impactful AI   

Drawing on her experience with clients, Susannah shared a checklist of actions that can lead to more meaningful and impactful AI initiatives: 

  • Understand your why. 

  • Speak to users. 

  • Define the problem to be solved.  

  • Understand your data.  

  • Define data requirements. 

  • Build or refine your data foundation.  

  • Define your evaluation and success criteria.  

  • Explore solution options.  

  • Implementation. 

AI should solve real problems for real people. Data is generated by people (whether that’s users or customers) so it is a direct reflection of human behaviours and needs. Speaking to users will help organisations to better understand their data, and their data will help them to better understand their users – these two things have to coexist to be successful. 

Unlocking the real value of AI 

Susannah wrapped up the session by reinforcing a few key messages: 

  • Massive AI investments haven’t paid off for many companies because they’ve overlooked the importance of strong data foundations. 

  • To see ROI from AI, start with data - structure it, understand it, and treat it as a critical asset. 

  • Don’t aim for AI “self-actualisation” if you haven’t yet secured your organisation’s basic data needs. 

Data is not a byproduct; it’s a core strategic asset. And without a solid foundation, no amount of AI magic will deliver real impact. 

Know where you stand 

Understanding your organisation’s AI maturity is the first step to scaling it effectively and responsibly. Our AI Maturity self-assessment helps you do just that, giving you a clear view of your current strengths and areas to improve, and providing a tailored roadmap to guide your next steps. 

It takes less than 10 minutes, and the insight is immediate. 

Search